ADVANCED TUNNEL FACE STABILITY

ANALYSES CONSIDERING FREE SPAN,

REINFORCEMENT AND DRAINAGE

CONGKE YI
Supervisors: Rafael Jiménez Rodríguez

Salvador Senent Domínguez

E.T.S.I Caminos, Canales y Puertos

Universidad Politécnica de Madrid

This dissertation is submitted for the degree of Doctor of Philosophy

Madrid, 2021
Título de la Tesis:

Advanced tunnel face stability analyses considering free span, reinforcement and

drainage

Autor: Congke Yi

Director: Rafael Jiménez Rodríguez; Salvador Senent Domínguez

Tribunal nombrado por el Mgfco. y Excmo. Sr. Rector de la Universidad Politécnica

de Madrid, el día ……. de …………………….... de …………..

TRIBUNAL CALIFICADOR

Presidente: ……………………………………………………………………

Vocal 1º: ……………………………………………………………………

Vocal 2º: ……………………………………………………………………

Vocal 3º: ……………………………………………………………………

Secretario: ……………………………………………………………………

Realizado el acto de defensa y lectura de la tesis el día ……. de ……………

de ………….. en Madrid, los miembros del Tribunal acuerdan otorgar la calificación

de: …………………………………………………………………………
 Abstract

ABSTRACT

This PhD thesis investigates several aspects that affect the stability of the tunnel face,

such as the free span and pore water pressures. It also analyzes stabilization measures

for these situations: a forepole umbrella is adopted as the representative reinforcement

measure for tunnels with a free span, whereas advance drainage boreholes are

considered for tunnels constructed under the water table.

The thesis analyzes the tunnel face stability employing different collapse mechanisms

developed in the framework of the Limit Analysis method. Numerical simulations (built

in the FLAC3D and OptumG2 codes) are used to support the Limit Analysis methodology

and to validate its main results: the failure geometry and the collapse pressure of the

tunnel face. Moreover, other analytical solutions proposed in the literature, both in the

frameworks of Limit Equilibrium and of Limit Analysis are used as benchmarks to

illustrate the capabilities of the proposed approach.

For tunnels excavated with a free span, an advanced Limit Analysis collapse

mechanism is proposed first; then, the influence of the unsupported length on the

stability of the tunnel face is analyzed, depending on the soil strength parameters. The

effects of a forepole umbrella are investigated by introducing its reinforcement

v
 Abstract

contribution into the free span mechanism with the assumption of two failure modes

for the forepole element: shear failure and bending moment failure.

For tunnels under the water table with advance drainage boreholes, the thesis firstly

studies the face stability under steady-state conditions, proposing a Limit Analysis

collapse mechanism that employs the pore water pressure distribution computed by

numerical modeling. The effects of the advance drainage boreholes (on the failure

geometry and on the collapse pressure) are studied considering the borehole

arrangement, water level height and tunnel overburden. Then, the stability of the tunnel

face under transient conditions is studied. Finally, the relationship between surface

settlements and support pressure applied on the tunnel face is used to study the effects

of advance drainage on excavation performance and stability.

vi
 Resumen

RESUMEN

Esta tesis doctoral investiga dos aspectos que afectan a la estabilidad del frente del túnel:

el vano libre (o longitud del túnel sin revestir) y las presiones de poro, estudiando

posibles medidas de estabilización antes estos factores. De manera concreta, se analiza

el efecto de los paraguas de micropilotes, al considerar el efecto del vano libre, y el

drenaje mediante perforaciones en el frente, en el caso de túneles ejecutados bajo el

nivel freático.

La tesis doctoral analiza la estabilidad del frente del túnel mediante diferentes

mecanismos de rotura desarrollados en el marco del Análisis Límite. Asimismo, se

emplean simulaciones numéricas (mediante los códigos FLAC3D y OptumG2) como

apoyo a las metodologías de Análisis Límite y para validar los resultados obtenidos por

los mecanismos de rotura: la geometría de fallo y la presión de colapso del frente del

túnel. Igualmente, se utilizan otras soluciones analíticas propuestas en la literatura,

tanto en el marco del Equilibrio Límite como del Análisis Límite, para mostrar las

posibilidades de los desarrollos propuestos y para comparar los resultados de los

diferentes planteamientos.

Para el estudio sobre el vano libre se propone, en primer lugar, un mecanismo de rotura

(como solución de contorno superior) formado por bloques traslacionales y rotacionales.

Mediante dicho mecanismo se estudia el efecto de la longitud sin sostenimiento en la

vii
 Resumen

estabilidad del frente en función de los parámetros resistentes del material donde se

excava el túnel. A continuación, se investiga el efecto estabilizador del paraguas de

micropilotes. Para ello, se introduce su contribución en el mecanismo de rotura

asumiendo dos modos de fallo del micropilote: fallo por cortante y fallo por flexión. Se

propone una metodología, desarrollada a partir de simulaciones numéricas, para la

estimación del espesor de la superficie de rotura en el mecanismo de fallo por flexión,

pudiendo, a partir de dicha estimación, obtener el valor de la presión de colapso del

frente de un túnel reforzado con un paraguas de micropilotes.

Para túneles ejecutados bajo el nivel freático, el estudio se centra, en primer lugar, en

la estabilidad del frente bajo la condición de flujo estacionario. Para ello se propone

una metodología a partir de un mecanismo de rotura en el marco del Análisis Límite y

en la interpolación de las presiones de poro calculadas mediante un modelo numérico.

Se estudia el efecto del drenaje sobre la geometría de la rotura y sobre la presión de

colapso en función de la disposición de los drenes, de la altura del nivel freático y del

recubrimiento sobre clave. Seguidamente, se estudia la estabilidad del frente del túnel

bajo la condición de flujo transitorio. Teniendo en cuenta la relación entre las

subsidencias producidas en superficie, debidas a la deformación de la excavación y al

drenaje, y la presión de sostenimiento aplicada, se estudia el efecto del drenaje en la

estabilidad del frente.

viii
 Acknowledgements

ACKNOWLEDGEMENTS

I would like to thank genuinely my supervisor Prof. Dr. Rafael Jimenez for his constant

guidance and unwavering support throughout this research. His foresight and resolute

decision contribute crucially to the delivery of the present work. Thanks to his

encouragement and endorsement, I have more confidence in making bald decisions in

the face of study and life challenges. Working with him is a highly rewarding

experience, even for a lifetime.

I am also sincerely grateful to Prof. Dr. Salvador Senent for his generosity and patience

to be my co-tutor. The completion of this research cannot be separated from his great

earnestness and devotion. I benefited a lot from his constructive advice and

collaborative engagement in the settlement of difficulties facing this work.

Financial supports from the China Scholarship Council (CSC) and the Spanish Ministry

of Economy, Industry and Competitiveness are greatly appreciated.

I enjoyed so much working in an inclusive atmosphere with colleagues and staff who

have different cultural backgrounds in the Geotechnics Laboratory. I appreciate all the

moments when we stay together to discuss academic issues or share daily delight. I was

impressed by their enthusiasm for work and passion for life.

ix
 Acknowledgements

I am very grateful to my friends in Spain, we always encourage each other to overcome

every difficulty, even in the toughest time facing the world-Coronavirus Pandemic.

Thanks also for everything that Spain brings to me, Spain has become my second home.

Last but not the least, I express my heartfelt gratitude to my family who shares

happiness and hardship with me, their supports are not absent at any time.

In memory of my grandpa, who has been warming my heart as I grew up.

x
 Contents

CONTENTS

Abstract .........................................................................................................................v

Resumen ..................................................................................................................... vii

Acknowledgements ..................................................................................................... ix

Contents ...................................................................................................................... xi

List of Symbols ...........................................................................................................xv

List of Tables ............................................................................................................ xix

List of Figures ........................................................................................................... xxi

1. Introduction ..........................................................................................................1

1.1 Problem statement ..........................................................................................1

1.1.1 Tunnel face stability considering the free span ..................................1

1.1.2 Tunnel face stability considering reinforcement measures ................2

1.1.3 Tunnel face stability considering advance drainage ..........................3

1.2 State of the art ................................................................................................4

1.2.1 On the relevance of free span .............................................................6

1.2.2 On the relevance of reinforcement measures .....................................8

1.2.3 On the relevance of drainage..............................................................9

1.3 Objectives .....................................................................................................15

1.4 Structure of the thesis ...................................................................................16

1.5 Research output ............................................................................................17

2. Methodology .......................................................................................................19

xi
 Contents

2.1 Introduction.................................................................................................. 19

2.2 Analytical methodologies ............................................................................ 20

2.2.1 Limit Analysis method .................................................................... 20

2.2.2 Limit Equilibrium method ............................................................... 22

2.3 Numerical methods ...................................................................................... 24

2.3.1 The Finite Element Method (OptumG2) ........................................... 24

2.3.2 The Finite Difference Method (FLAC3D) ........................................ 25

3. Tunnel face stability analysis considering the free span ................................ 29

3.1 Introduction.................................................................................................. 29

3.2 Tunnel face stability 2D mechanism considering the free span .................. 30

3.2.1 Model description ............................................................................ 30

3.2.2 Collapse mechanism ........................................................................ 31

3.2.3 Formulation of the analytical solution ............................................. 33

3.3 Numerical validation ................................................................................... 38

3.3.1 Test Cases ........................................................................................ 38

3.3.2 Numerical model ............................................................................. 39

3.3.3 Results ............................................................................................. 40

3.4 Results and discussions ............................................................................... 44

3.5 Conclusions ................................................................................................. 51

4. Effect of a forepole umbrella on the stability of the tunnel face ................... 53

4.1 Introduction.................................................................................................. 53

4.2 Incorporation of a forepole umbrella into the Limit Analysis tunnel face
stability mechanism ..................................................................................... 54

4.3 Effects of forepole umbrella on the stability of the tunnel face .................. 57

4.3.1 Collapse pressure ............................................................................. 60

4.3.2 Failure geometry .............................................................................. 65

xii
 Contents

4.4 Conclusions ..................................................................................................67

5. Tunnel face stability considering advance drainage under steady-state

conditions ............................................................................................................69

5.1 Introduction ..................................................................................................69

5.2 3D Limit Analysis mechanism considering pore water pressure and the
effect of advance drainage ...........................................................................71

5.2.1 Model description.............................................................................71

5.2.2 Limit Analysis mechanism ...............................................................72

5.3 Numerical computation of pore water pressure distribution under steady-

state conditions .............................................................................................76

5.4 Numerical validation ....................................................................................80

5.4.1 Test Cases.........................................................................................80

5.4.2 Results of Limit Analysis .................................................................82

5.4.3 Results of numerical analysis ...........................................................84

5.4.4 Comparison ......................................................................................87

5.5 Discussions ...................................................................................................88

5.5.1 Effect of advance drainage on the stability of the tunnel face .........88

5.5.2 Comparison with other solutions from the literature .......................90

5.6 Conclusions ..................................................................................................95

6. Tunnel face stability considering advance drainage under transient

conditions ............................................................................................................97

6.1 Introduction ..................................................................................................97

6.2 Problem setting.............................................................................................98

6.3 Computational model ...................................................................................99

6.4 Time-dependent behaviors under transient conditions ..............................102

6.4.1 Pore water pressure evolution ........................................................102

6.4.2 Plastic failure zone evolution .........................................................103

xiii
 Contents

6.5 Effect of advance drainage on the tunnel face stability ............................. 105

6.5.1 Pore water pressure ahead of the tunnel face................................. 105

6.5.2 Stand-up time of the tunnel face .................................................... 105

6.5.3 Settlement ...................................................................................... 108

6.6 Conclusions ............................................................................................... 109

7. Conclusions and outlook ................................................................................. 111

7.1 Conclusions ............................................................................................... 111

7.2 Outlook ...................................................................................................... 114

References ................................................................................................................. 117

Appendices ............................................................................................................... 131

Appendix A MATLAB-code Limit Analysis mechanism with free span

considering the effect of forepole umbrella ............................................... 135

Appendix B MATLAB-code Limit Analysis mechanism considering the advance

drainage...................................................................................................... 145

Appendix C FLAC3D-code Numerical Simulation considering the advance

drainage...................................................................................................... 185

xiv
 List of Symbols

LIST OF SYMBOLS

Latin symbols

𝐴 cross-sectional area of the structural element of the forepole

umbrella

𝐵 excavation width

𝑐 soil cohesion

𝑐 undrained strength of soil

𝐶 overburden

𝑑 diameter of the structural element of the forepole umbrella

𝑑 diameter of drainage borehole

𝑑 ,𝑑 exterior and interior diameters of the structural element of the

forepole umbrella

𝐷 diameter of the tunnel

𝐸 Young’s modulus

𝑓 yield strength of the structural element

ℎ hydraulic head

ℎ initial hydraulic head at the tunnel axis

𝐻 excavation height

𝐻 water elevation from the tunnel crown

xv
 List of Symbols

𝑘 permeability coefficient

𝐾 water modulus

𝑙 length of the interface between points 𝑖 and 𝑗 of the 2D mechanism

for tunnels with a free span

𝐿 unsupported length

L length of drainage borehole

𝐿 maximum length of free span without support pressure to keep stable

𝑀 ultimate bending moment of the structural element

𝑛 number of drainage boreholes

𝑁 ultimate axial force of the structural element

𝑂 center of rotation of the 3D mechanism in the vertical plane of

symmetry of the tunnel

𝑂 center of rotation of the 2D mechanism for tunnels with a free span

𝑝 pore water pressure in the tunnel face

𝑃 applied support pressure on the tunnel face of numerical modeling

𝑃( ... ) points in the interface BF of the proposed 2D mechanism for tunnels

with a free span

𝑃 ,𝑃 initial face pressure values of the bisection method

𝑞 water discharge via the tunnel face

𝑟 distance from tunnel center to the borehole

𝑅 radius of spiral line of the 2D mechanism proposed by Zhang et al.

(2018)

(𝑅 , 𝛽 ) polar coordinates of point or center 𝑖

𝑅 threshold value of Maximum unbalanced force ratio reaching the

stability

𝑅 maximum unbalanced mechanical-force ratio

𝑠 interval of adjacent points in the interface AF of the proposed 2D

mechanism for tunnels with a free span

𝑆 area of Block 𝑖 (𝑖 =1, 2, and 3)

xvi
 List of Symbols

𝑆, area of the discretized element on the tunnel face

𝑆, ,𝑆 , areas of facet 𝐹 , and 𝐹 ,

𝑡 time

𝑇( ... ) points in the interface AF the proposed 2D mechanism for tunnels

with a free span

𝑢 pore water pressure

𝑢⃗̇ unit vector of the interface AF in Point 𝑃

𝑢, pore water pressure of the barycenter of the discretized element 𝑆 ,

𝑢 , ,𝑢 , pore water pressures in the location of the barycenter of triangular

facets 𝐹 , and 𝐹 ,

𝑣⃗ assumed velocity vector in the kinematically admissible velocity

field

𝑉 ultimate shear force of the structural element

𝑉 ,𝑉 velocities of Block 1 and Block 3

𝑉 relative velocity between Block 1 and Block 3

𝑉,̇ velocity of Block 2 in Point 𝑃

𝑉 ,̇ relative velocity between Block 2 and Block 3 in Point 𝑃

𝑉⃗ velocity of the neutral axis of the structural element at the velocity

jump surface

𝑉, ,𝑉, element volumes corresponding to facet 𝐹 , and 𝐹 ,

𝑊 work rate of forces or internal energy dissipations

𝑊 work of forces or internal energy dissipations

𝑦 distance from the tunnel face

𝑦 ,𝑧 Cartesian coordinates of Point 𝑃

𝑌,𝑍 Cartesian coordinates of point or center 𝑖

xvii
 List of Symbols

Greek symbols

𝛼 angle of the cross-section of the structural element with the failure surface

𝛼 angle in Point A of the interface AF of the proposed 2D mechanism for

tunnels with a free span with the horizontal

𝛼 angle of the velocity of Block 3 with the horizontal

𝛽 angle of the interface AD of the proposed 2D mechanism for tunnels with

a free span with the horizontal

𝛾 unit weight

𝛾 submerged unit weight

𝛾 saturated unit weight

𝛿 thickness of the continuous deformation zone

𝛿 𝛿 suggested by De Buhan and Salençon (1993) (Equation (4.4))

𝛿 .( . ) 𝛿 estimated employing Equation (4.5)

𝛿 𝛿 approximated from the numerical model

𝜀 computation precision of support pressure in the bisection method

𝜀 volumetric strain

𝜂 angle of the forepole umbrella with the horizontal

𝜇 non-dimensional parameter

𝜎 collapse pressure

𝜎 support pressure

𝜐 Poisson’s ratio

𝜑 friction angle

𝜓 dilatancy angle

𝜔 initial slope (or critical angle) of the failure surface

𝜔 angular velocity around 𝑂 of the Block 2

xviii
 List of Tables

LIST OF TABLES

Table 3.1. Test cases for the numerical validation of the proposed Limit Analysis
mechanism.................................................................................................. 41

Table 4.1. Properties considered to study the effect of the forepole umbrella on the
stability of the tunnel face .......................................................................... 59

Table 5.1. Test Cases considered for numerical validation of the Limit Analysis
mechanism. (Diameter:D=10m; strength parameters: c=0kPa and φ=35°;
saturated unit weight: γsat=15.6kN/m3). ..................................................... 81

Table 5.2. Test Cases considered to compare the Limit Analysis mechanism with the
Limit Equilibrium methodology presented by Zingg (2016). (Diameter:
D=10m; strength parameters: c=0kPa and φ=35°; saturated unit weight:
γsat=20kN m3). ............................................................................................ 92

xix
 List of Figures

LIST OF FIGURES

Figure 2.1. Limit analysis failure mechanism for tunnels with a free span (after Zhang
et al., 2018)................................................................................................. 22

Figure 2.2. Limit Equilibrium failure mechanism for tunnels under the water table
(after Zingg, 2016) ..................................................................................... 23

Figure 2.3. Flow chart of the bisection method in FLAC3D to compute the critical
support pressure ......................................................................................... 26

Figure 2.4. Process for fluid-solid coupled analysis in FLAC3D to study the time-
dependent behaviors of tunnel face stability .............................................. 28

Figure 3.1. Four heading failure mechanisms with free span in NATM a) bench
failures; b) crown failures; c) full-face failures d) local face failures (HSE,
1996) .......................................................................................................... 30

Figure 3.2. Problem setup for studying the stability of a tunnel with free span .......... 31

Figure 3.3. Outline of the proposed Limit Analysis mechanism to compute tunnel face
stability considering the free span .............................................................. 32

Figure 3.4. Hodographs: a) interface between Blocks 1 and 3; b) in Point A between

Blocks 2 and 3; c) interface between Blocks 2 and 3 ................................ 35

Figure 3.5. Geometry of the numerical model built in OptumG2 ................................. 39

Figure 3.6. Comparison of collapse face pressures computed with the proposed
mechanism and with the numerical model ................................................. 40

xxi
 List of Figures

Figure 3.7. Comparison of failure geometries obtained with the proposed mechanism
and with the numerical model for different strength parameters (φ, c) and
lengths of the free span (L) ........................................................................ 43

Figure 3.8. Collapse pressure vs length of the free span for different strength
parameters .................................................................................................. 45

Figure 3.9. Collapse pressure vs free span length for different values of cohesion
considering a supported and an unsupported free span: a) φ=30°; b) φ=35°
................................................................................................................... 46

Figure 3.10. Maximum unsupported span vs cohesion ............................................... 48

Figure 3.11. Variation of the failure geometry depending on the free span (L=1,2,3):
a) φ=30°, c=5kPa; b) φ=30°, c=15kPa; c) φ=35°, c=5kPa; d) φ=35°,
c=15kPa ..................................................................................................... 49

Figure 3.12. Collapse pressures computed, as a function of the free span, with (i) the
proposed mechanism, (ii) the Limit Analysis methodology presented by
Zhang et al. (2018), and (iii) an Upper Bound Analysis in OptumG2.
(D=10m; c=5kPa; γ=18kN/m3): a) φ=20° (the pressure applied on the free
span is constant and equal to 50kPa); b) φ=40° (the pressures applied on
the free span and on the tunnel face are equal) .......................................... 50

Figure 4.1. Forepole umbrella layout in an ongoing excavation tunnel. a) longitudinal

profile; b) Section A:A’ (after Lunardi, 2008) .......................................... 54

Figure 4.2. Failure modes considered for the forepole umbrella (modified from De
Buhan and Salençon, 1993) ....................................................................... 56

Figure 4.3. Outline of the analyzed case to study the effect of the forepole umbrella on
the stability of the tunnel face .................................................................... 58

Figure 4.4. Collapse pressure vs ultimate bending moment of the forepole umbrella
for the three cases analyzed: a) Case 24 (φ=35°; c=5kPa; L=3m); b) Case
23 (φ=35°; c=5kPa; L=2m); c) Case 27 (φ=35°; c=15kPa; L=1m) ........... 62

Figure 4.5. Analysis of Equation (4.5): a) comparison between δ values estimated

from the numerical model (δNM) and those computed with Equation (4.5)
(δEq.(4.5)); b) comparison of collapse pressures computed employing δNM
and δEq.(4.5) for the three cases analyzed ..................................................... 64

xxii
 List of Figures

Figure 4.6. Comparison of failure geometries obtained with the proposed mechanism
and with the numerical model in Case 24 (φ=35°; c=5kPa; L=3m) for
different amounts of supports: a) without umbrella; b) light umbrella; c)
heavy umbrella ........................................................................................... 66

Figure 5.1. a) Water level distributions in different hydraulic conditions; b) a drainage

borehole (Lunardi, 2008) ........................................................................... 70

Figure 5.2. Outline of the analyzed problem: a) in the vertical plane of symmetry of
the tunnel; b) in the tunnel face cross-section. (Boreholes layout: (I)
“upper” drainage; (II) “lower” drainage.) .................................................. 72

Figure 5.3. Generation of the Limit Analysis mechanism: a) cross-section at the tunnel
face boundary; b) longitudinal section along the vertical plane of
symmetry of the tunnel (simplified from Mollon et al. (2011a), Fig.3) .... 73

Figure 5.4. Point-by-point generation of the external surface of the mechanism

(simplified from Mollon et al. (2011a), Fig.4) ........................................... 74

Figure 5.5. Numerical model with advance drainage boreholes employed to calculate
the pore water pressure distribution and the collapse pressure .................. 77

Figure 5.6. Contours of the pore water pressure distribution in the numerical model
for different drainage configurations: a) without drainage; b) “upper”
drainage boreholes; c) “lower” drainage boreholes. (D=10m; C=2D;
Hw=3D). (Values shown indicate the hydraulic head in kPa) .................... 78

Figure 5.7. Distribution of hydraulic head along the tunnel axis for several Test Cases
(see Table 5.1). (h: hydraulic head; h0: initial hydraulic head; y: distance
from the tunnel face) .................................................................................. 79

Figure 5.8. Examples of the failure geometries obtained with the Limit Analysis
mechanism: a) Test Case 2 (without drainage); b) Test Case 6 (“upper”
drainage); c) Test Case 10 (“lower” drainage). (D=10m; C=2D; Hw=3D) 83

Figure 5.9. Comparison of failure mechanisms computed with the Limit Analysis
solution and with the numerical simulation for different drainage
configurations: a) without drainage (Test Case 2); b) “upper” drainage
(Test Case 6); c) “lower” drainage (Test Case 10). (D=10m; C=2D;
Hw=3D)....................................................................................................... 85

xxiii
 List of Figures

Figure 5.10. Comparison of failure mechanisms computed with the Limit Analysis
solution and with the numerical simulation for different water levels: a)
Hw=2D (Test Case 9); b) Hw=3D (Test Case 10); c) Hw=4D (Test Case 11);
d) Hw=5D (Test Case 12). (D=10m; C=2D; “lower” drainage) ................ 86

Figure 5.11. Comparison of failure mechanisms computed with the Limit Analysis
solution and with the numerical simulation for different overburdens: a)
C=2D (Test Case 6); b) C=3D (Test Case 13). (D=10m; Hw=3D; “upper”
drainage) .................................................................................................... 86

Figure 5.12. Variation of the collapse geometry depending on the drainage

configuration for different water levels: a) Hw=3D; b) Hw=4D. (Cases 2
and 3 without drainage; Cases 6 and 7 “upper” drainage; Cases 10 and 11
“lower” drainage). (D=10m; C=2D) .......................................................... 90

Figure 5.13. Collapse pressures computed with the Limit Analysis mechanism, with
the Limit Equilibrium methodology presented by Zingg (2016), and with a
numerical simulation in FLAC3D, for different drainage configurations.
(See Table 5.2) ........................................................................................... 93

Figure 5.14. Comparison of the failure geometries obtained with the Limit Analysis
mechanism and with the Limit Equilibrium methodology presented by
Zingg (2016): a) “upper” drainage (Test Case 6 in Table 5.2); b) “lower”
drainage (Test Case 10 in Table 5.2). (D=10m; C=2D; Hw=3D) .............. 94

Figure 6.1. Schematic of the analyzed problem: a) in the vertical plane of symmetry of
the tunnel; b) in the tunnel face cross-section ........................................... 99

Figure 6.2. Numerical model with advance drainage boreholes in the analysis........ 101

Figure 6.3. Pore pressures ahead of the tunnel face along the tunnel axis for different
times without advance drainage............................................................... 103

Figure 6.4. Plastic failure zone evolution without drainage boreholes...................... 104

Figure 6.5. Pore pressures ahead of the tunnel face along the tunnel axis for different
times after introducing advance drainage ................................................ 105

Figure 6.6. Evolution of surface settlement at the monitoring point for different
support pressures: a) without advance drainage; b) with advance drainage
................................................................................................................. 106

xxiv
 List of Figures

Figure 6.7. Stand-up time vs support pressure. (Limit Analysis results have been
computed using the pore pressure distributions under steady-state
conditions) ................................................................................................ 107

Figure 6.8. Longitudinal surface trough for different support pressures and drainage
configurations ........................................................................................... 109

xxv
 1. Introduction

1. Introduction

1.1 Problem statement

The stability of the tunnel face is a key aspect of tunnel design as heavy casualties and

economic losses caused by face collapses during tunnel construction are not unusual.

However, this is a complex problem, and previous theoretical studies in the literature

mainly focus on simplified tunnel face stability analyses, so that the effects of some

advanced aspects, like the effect of the free span or the excavation under the water table,

are often neglected for engineering design and construction. Moreover, improvement

measures under such conditions (e.g., forepole umbrellas and advance drainage) are

typically not studied in detail, although they play a key role to reduce the tunnel face

instability and its deformation.

The present PhD thesis deals with the shortage of such studies, providing an advance

of the state-of-the-art for the following problems:

1.1.1 Tunnel face stability considering the free span

The unsupported length next to the tunnel face is defined as the free span (or

unsupported span). The free span typically occurs in projects constructed by the New

Austrian Tunneling Method (NATM), and it imposes a risk contributing to tunnel face

1
 1. Introduction

instabilities that can extend significantly beyond the tunnel face or even up to the

ground surface (Baudendistel, 1985). (Such instabilities can occur despite the self-

bearing contribution produced by the soil arching mechanism that develops above the

unlined portions of the excavation). Similarly, the free span can affect the failure

mechanism. For instance, according to Casarin and Mair (1981), who carried out tests

in cohesive materials, two different instability mechanisms can occur depending on the

ratio between free span length and tunnel diameter (L/D): for lower L/D ratios, the

failure mechanism mainly affects the ground ahead of the tunnel face; whereas, for

larger L/D ratios, the mechanism develops above the free span, producing an almost 2D

failure. Moreover, even for stable faces, the unsupported length influences the

deformability of the excavation (e.g., Nemati Hayati et al., 2013) and consequently, it

has an impact on nearby structures (e.g., van Tol, 2006).

This importance of the free span has therefore led us to study the stability of the tunnel

face considering the free span.

1.1.2 Tunnel face stability considering reinforcement measures

To improve the face stability in tunnels with a free span, reinforcement measures can

be implemented. For shallow tunnels with poor ground conditions, forepole umbrellas

are one of the most common reinforcement systems (Muraki, 1997). In tunnels with a

free span, the umbrella can support the weight of the material above the free span, hence

avoiding the development of a progressive failure that could reach the surface. However,

despite its advantages, there is not an established methodology to consider the effects

of such reinforcement support on the face stability analysis (Volkmann and Schubert,

2007). For this reason, this thesis investigates the effect of a forepole umbrella as

reinforcement to improve the stability of tunnel faces with a free span.

2
 1. Introduction

1.1.3 Tunnel face stability considering advance drainage

Tunneling under the water table imposes great risks due to the seepage forces acting on

the excavated tunnel face. Therefore, the safety of tunneling works under high pore

pressures can be compromised and conventional tunneling methods can be unsuitable

in these circumstances. Advance drainage is defined as drainage assisted by drainage

elements at the face (normally, boreholes, but also pilot tunnels or a twin tunnel), that

aim to reduce seepage forces ahead of the tunnel face before its excavation. More and

more tunneling projects are carried out under the water table, and advance drainage is

a very practical tool to improve the face stability of subaqueous tunnels when the

required support pressure is too high to be practically achievable (Pelizza and Peila,

1993).

Seepage flows within the ground can theoretically be divided into two types: (i) steady-

state flow, in which pore water pressures and flow are constant over time; and (ii)

transient flow, in which they change with time. The tunnel face stability under the water

table is often studied under steady-state conditions since pore water pressure

distributions can be easily computed and stresses in the ground do not change. But the

stability of the tunnel face under transient conditions is time-dependent, and a study of

how long the tunnel face can maintain the stability, considering the drainage and the

support pressure, is needed.

This thesis studies the stability of tunnel faces under the water table, considering the

influence of advance drainage under both steady-state flow (which may be expected in

a relatively high permeability material) and transient flow (which may be dominant in

low to medium permeability soils).

3
 1. Introduction

1.2 State of the art

Many studies on tunnel face stability have been published in recent decades, and some

of the more relevant ones are discussed herein. Concerning theoretical failure

mechanisms, the prism-wedge soil model proposed by Horn (1961) has been widely

employed to compute, in the framework of the Limit Equilibrium method, the critical

collapse pressure of the tunnel face (i.e., the minimum pressure that needs to be applied

on the tunnel face to avoid its instability) (e.g., Anagnostou and Kovari, 1994; Broere,

1998; Chen et al., 2015; Zhang et al., 2020a). Similarly, Liu et al. (2019) proposed an

improved model for face stability analysis with a dual-failure-mechanism, which is also

based on the Limit Equilibrium method.

In the framework of the Limit Analysis method, a two-block failure model was first

proposed by Leca and Dormieux (1990), and later improved by Mollon et al. (2009b),

who introduced a multiblock failure mode. Mollon et al. (2010) proposed a

methodology to generate the failure surface of a multiblock translational failure

mechanism, which is generated “point by point’’ from the entire (circular) tunnel face.

This methodology was later extended to a one-block rotational failure mechanism

(Mollon et al., 2011a), obtaining better estimations of the critical collapse pressure and

of the failure geometry. Other collapse mechanisms in the framework of Limit Analysis

enriched the tools for stability analyses of the tunnel face considering, for example,

non-linear failure criteria (Senent et al., 2013), layered soils and the possibility of partial

collapse (Senent and Jimenez, 2015; Chen et al., 2019), soil arching (Zou et al.,2019),

a longitudinally inclined face (Zhao et al., 2017) and non-homogeneous soils (Zou et

al., 2019).

4
 1. Introduction

Numerical simulations have also been widely employed to analyze tunnel face stability.

In one of the most well-known works, Vermeer et al. (2002) conducted 3D stability

analyses using the Finite Element Method (FEM) that identified several relevant factors

that affect the tunnel face stability. The FEM was also employed, for example, to

compute the face stability of tunnels in soils with linearly increasing shear strength with

depth (Ukritchon et al., 2017), or to determine the required support pressure of tunnel

faces in layered soils (Alagha and Chapman, 2019).

Similarly, the 3D Finite Difference Method (FDM) has been used by Chen et al. (2013)

to verify the accuracy of critical support pressures obtained experimentally. Similarly,

Senent et al. (2013) also used the 3D FDM to investigate the face stability of tunnels in

fractured rocks subject to the Hoek-Brown criterion; and Senent and Jimenez (2015)

studied the partial collapse of a tunnel face in layered soil. Based on FDM numerical

simulations, Zhang et al. (2015) proposed a Limit Analysis failure mechanism

composed of four truncated cones. Other approaches are less common: for example, the

Discrete Element Method (DEM) has been used by Chen et al. (2011) to study tunnel

face stability in dry sand, whereas the 3D Kinematical Element Method (KEM) was

used to investigate the tunnel face stability by Qarmout et al. (2019).

Experimental tests have been also designed to study tunnel face stability. For example,

Takano et al. (2006) used an X-ray tomography scanner to ‘visualize’ the shape of the

face failure mechanism at single gravity. The development of the failure mechanism,

as a function of the support pressure applied at the tunnel face, was investigated by

Kirsch (2010) using small-scale model tests, with his results showing that the

overburden has a negligible effect on the extent and evolution of the failure volume.

Ahmed and Iskander (2012) presented soil deformations associated with various face

5
 1. Introduction

support pressures, as obtained from transparent soil models of tunnel faces. Senent et

al. (2019) studied the relationship between the tunnel face support and surface

settlements, utilizing a small-scale model at single gravity and the low-cost Structure

from Motion photogrammetric technique.

In addition to the general review of tunnel face stability analyses discussed above, the

literature review of some advanced issues (i.e., free span, reinforcement, and drainage)

is extended below.

1.2.1 On the relevance of free span

Some experimental works have studied the failure mechanism of the tunnel face, and

its relationship with the free span (e.g., Kimura and Mair, 1981; Chambon and Corte,

1994; Oblozinsky and Kuwano, 2004; Lee and Schubert, 2008). Kimura and Mair (1981)

conducted centrifuge tests with L/D ratios from 0 to 2, and they observed a significant

influence of L/D on the tunnel stability and on the shape of the failure mechanism

(which evolved from 3D to 2D behavior). Chambon and Corte (1994), who used model

tests with L/D ratios between 0.1 and 0.4, observed the substantial influence of the

unsupported length on the failure geometry and on the support pressure, so that the

failure mechanism noticeably extends to the ground surface when L/D > 0.1, becoming

a 2D mechanism for higher ratios. Oblozinsky and Kuwano (2004) conducted

centrifuge tests with L/D ratios equal to 0.1, showing that such a short unsupported

length had no significant effect on the critical support pressure. Lee and Schubert (2008)

discussed the failure modes of a tunnel face with unsupported span through a set of

small-scale model tests; in particular, they defined five modes according to the collapse

behaviors of the unsupported span and of the tunnel face, and suggested pre-support

and face support measures for each mode in the excavation plan.

6
 1. Introduction

Numerical models have been employed to study the stability of tunnels with a free span.

Among them, three-dimensional models using the Finite Element Method (e.g.,

Kielbassa and Duddeck, 1991; Vermeer et al., 2002), the Finite Difference Method (e.g.,

Costa et al., 2007) or the Discrete Element Method (e.g., Lee et al., 2005; Lee and

Schubert, 2008) can be found in the literature. Kielbassa and Duddeck (1991) carried

out elastic three-dimensional finite element analyses, showing that the free span affects

greatly the stress release, and that the effective ground pressures determine the lining

design. Vermeer et al. (2002) studied the stability of tunnel faces in cohesionless

materials. Their results illustrate that the stability of the face is almost independent of

the free span for low L/D ratios (i.e., L/D<0.3); however, they indicated that results

depend on the geometry of the tunnel, so that the influence of the free span is higher for

smaller circular tunnels. Costa et al. (2007) recognized the same two mechanisms

associated with the free span reported by Casarin and Mair (1981), and they pointed out

that the transition between both mechanisms depends on drainage, with the effect of the

free span on the collapse pressure being more relevant under undrained conditions.

Based on these experimental observations and numerical simulations, some analytical

approaches have been established to study the influence of the free span. For example,

Tamez (1984) proposed a Limit Equilibrium failure mechanism, similar to the classical

mechanism proposed by Horn (1961) (which is based on the silo theory of Janssen,

1895) but with a prismatic block above the free span. Cornejo (1989) provided a

formulation to determine the allowable tunnel advance length and the required face

support pressure, considering two ground types: (i) an isotropic and homogeneous

ground; and (ii) a stratified ground with strength parameters varying with depth. He

also summarized earlier formulations proposed by previous researchers (e.g., Broms

7
 1. Introduction

and Bennermark, 1967; Tamez, 1984; Ellstein, 1986) suggesting that all of them can be

used to calculate the stability of the tunnel face, so that the final result can be determined

eventually by comparison. Employing a similar model in the framework of the Limit

Equilibrium with the method of slices, Anagnostou and Perazzelli (2013) studied the

effect of the free span and of the support pressure distribution, on the stability of the

tunnel face. Based also on the Limit Equilibrium method, Yu et al. (2020) proposed a

failure mechanism that introduces a rotational log-spiral slip surface into the lower

wedge, instead of the translational surface assumed by Anagnostou and Perazzelli

(2013), to better approximate the actual failure modes observed in the laboratory. Very

recently, Tian et al. (2020) proposed an approach to determine the influence of the free

span, considering the stability of both the arch crown and the tunnel face.

In the framework of Limit Analysis, solutions to the tunnel face stability considering

the free span are rare. Senent and Jimenez (2017) presented a two-dimensional failure

mechanism that combines rotational and translational non-deformable blocks. Similarly,

Zhang et al. (2018) proposed a two-dimensional Limit Analysis failure mechanism,

composed of a triangular block above the tunnel crown and of a spiral block ahead of

the tunnel face; their results show that the collapse pressures obtained by the Limit

Analysis mechanism are clearly lower than those obtained with the Limit Equilibrium

approach proposed by Anagnostou and Perazzelli (2013).

1.2.2 On the relevance of reinforcement measures

To improve the stability of the tunnel face, a forepole umbrella is considered, since it

is one of the most common reinforcement systems employed when a tunnel passes

through soils or weak rocks (e.g., Peila, 1994; Wang and Jia, 2008; Kitchah and

Benmebarek, 2016; Ocak and Selcuk, 2017). The umbrella, as a pre-support

8
 1. Introduction

reinforcement measure in the classical NATM, supports the weight of the material

above the free span, hence avoiding the development of a progressive failure that could

reach the surface. However, bolts are usually the support measure considered to model

the effect of reinforcement on the stability of the tunnel face, while typically neglecting

the influence of the free span (e.g., Ng and Lee, 2002; Kamata and Mashimo, 2003;

Yoo and Shim, 2003; Anagnostou and Serafeimidis, 2007; Perazzelli and Anagnostou,

2013; Anagnostou and Perazzelli, 2015; Paternesi et al., 2017; Pan and Dias, 2017).

Some authors have studied the performance of forepole umbrellas at the tunnel face;

e.g., using centrifuge tests (e.g., Calvello and Taylor, 1999; Date et al., 2009; Juneja et

al., 2010; Le and Taylor, 2017), or numerical simulations (e.g., Elyasi et al., 2016;

Klotoé and Bourgeois, 2019). From an analytical approach, Zhang et al., (2020b)

analyzed the stability of a tunnel face reinforced with bolts and with an umbrella arch

using the Limit Equilibrium method and a strength reduction technique. In the Limit

Analysis literature, there are few works that analyze forepole umbrellas, as previous

works have mainly considered a bolt reinforcement –a support that only works by axial

force– (e.g., Pan and Dias, 2017). Pinyol and Alonso (2011, 2013) studied the effect of

a forepole umbrella on the stability of the tunnel face in undrained soils, employing one

of the translational mechanisms proposed by Leca and Dormiuex (1990). Qian et al.

(2019) combined the strength reduction technique and an advance rotational

mechanism to compute the safety factors of non-circular tunnel faces reinforced with

umbrella pipes. However, these works did not consider the influence of the free span.

1.2.3 On the relevance of drainage

To reduce the instability of a tunnel face excavated under the water table, advance

drainage measures are used to relieve the pore water pressure. As a result, the seepage

9
 1. Introduction

decreases to a point that makes it possible that the face support pressure required for

stability can be achieved in conventional tunneling. As stated in Section 1.1.3, this

thesis studies tunnel face stability considering different seepage conditions (steady-

state and transient). The relevant literature is summarized below.

1.2.3.1 Seepage under steady-state conditions: tunnel face stability

The contribution of seepage on tunnel face stability has been incorporated into several

analytical solutions that do not initially consider seepage forces (e.g., Anagnostou and

Kovari, 1994; Leca and Dormieux, 1990; Mollon et al., 2011a). For instance, in the

framework of the Limit Equilibrium method, Anagnostou and Kovari (1996) introduced

seepage forces into the wedge-prism failure mechanism proposed by Anagnostou and

Kovari (1994); such seepage forces were computed by integration of the numerically

determined hydraulic heads. Perazzelli et al. (2014) improved the methodology of

Anagnostou and Kovari (1996) to investigate the tunnel face stability under seepage

conditions using the “method of slices”, and approximating the hydraulic head

distributions from the numerical simulations by fitting equations. The wedge-prism

model proposed by Anagnostou and Kovari (1996) was also modified by Lü et al.

(2017), who modeled the effect of seepage on the inclination angle of the wedge and

employed the hydraulic head fitting equations proposed by Perazzelli et al. (2014) to

match better with the numerical results.

In the framework of Limit Analysis, the two-block translational failure mechanism of

Leca and Dormieux (1990) has been extended to analyze the effect of seepage forces

on tunnel face stability (e.g., Lee and Nam, 2001, 2004; Lee et al., 2003, 2004; Tu et

al., 2012). Similarly, Pan and Dias (2016) extended the 3D rotational failure mechanism

proposed by Mollon et al. (2011a) to consider the seepage forces, and further analyzed

10
 1. Introduction

the influence of different variables that affect the stability of a tunnel face under the

water table, like the permeability anisotropy. The face stability of tunnels under the

water table in weak rock masses with the Hoek-Brown criterion has also been studied

by Pan and Dias (2018). More recently, Li et al. (2021) assessed the effect of seepage

on tunnel face stability, by introducing analytically computed hydraulic head

distributions into the 3D rotational failure mechanism proposed by Mollon et al.

(2011a).

In addition, several works have employed numerical methods to study the effect of

seepage forces on tunnel face stability under steady-state conditions (e.g., Anagnostou,

1995; De Buhan et al., 1999; Li et al., 2011; Lü et al., 2014; Weng et al., 2020).

Anagnostou (1995), by reformulating and solving the diffusion equation under steady-

state conditions, pointed out that hydraulic head gradients were underestimated when a

high advance rate is applied. De Buhan et al. (1999) analyzed the tunnel face stability

using seepage forces computed with a variational formulation. Li et al. (2011) provided

a simple but effective tool to compute the safety factor of a tunnel face, by adding the

seepage forces calculated with the FEM to the stress field. Lü et al. (2014) studied the

required support pressure with the FEM, showing that a large part of the applied support

pressure is used to equilibrate the seepage force and that the critical support pressure

has a linear relationship with the water table. Based on numerical simulations with the

FEM, Weng et al. (2020) indicated that soil permeability has a negligible influence on

the seepage force for a tunnel excavated under steady-state conditions.

Some case histories and field observations (e.g., Pellet et al., 1993; Barla, 2000) also

provided important information about the role played by seepage forces in relation to

tunnel stability. Additionally, laboratory tests have been conducted to analyze the

11
 1. Introduction

stability of the tunnel face under seepage conditions (e.g., Lee et al, 2003; Chen et al.,

2018; Lü et al., 2018; Weng et al., 2020). Lee et al. (2003) conducted small-scale model

tests to measure the seepage forces acting on the tunnel face, showing that the

relationship between seepage pressure and groundwater level to tunnel diameter ratio

(H/D) is almost linear. Chen et al. (2018) conducted centrifuge tests to study the face

stability of tunnels with steady-state seepage, finding that pore water pressure changes

only occur within a distance of less than 0.75D ahead of the tunnel face. Lü et al. (2018)

studied the influence of the cover to diameter ratio (C/D) and of the water level on the

collapse pressure and on the failure geometry. More recently, Weng et al. (2020) carried

out centrifuge tests, showing that the seepage force is independent of the longitudinal

slope of the tunnel, whereas the critical support pressure significantly increases with

the longitudinal slope angle.

1.2.3.2 Seepage under transient conditions: tunnel face stability and settlements

In some grounds with low to medium permeability, the transient flow can control tunnel

face stability, due to the occurrence of negative excess pore water pressures that

dissipate over time (Broere, 2002, 2003). For practical operations under transient

conditions, the delayed failure time (or stand-up time) of the tunnel face and the surface

settlements are the two key indicators of tunnel face stability. The tunnel face stand-up

time under transient flow has been studied by several authors (e.g., Schuerch and

Anagnostou, 2013a, 2013b, 2014, 2015; Schuerch et al., 2016; Callari et al., 2017), and

several publications have also reported and analyzed field observations of the time-

dependent settlements (e.g., Attewell, 1988; Shirlaw, 1994; O’Reilly et al., 1999; Chai

et al., 2004; Mair, 2008; Shen et al., 2014; Soga et al., 2017). Similarly, numerical

models have been employed to study surface settlements under transient conditions

12
 1. Introduction

(e.g., Shin et al., 2002; Greenwood, 2003; Zhang et al., 2004; Kasper and Meschke,

2006a; Höfle et al., 2008; Wang et al., 2012; Wongsaroj et al., 2007,2013; Callari, 2015;

Cattoni et al., 2016; Laver et al., 2017).

The strong influence of face support pressure on surface settlements has been discussed

in the literature (e.g., Shirlaw et al., 2002; Kasper and Meschke, 2006b), and Broere

(2015) also pointed out that a proper support pressure is very important for a safe and

controllable operation. Some real cases, like the First Bangkok Subway Line (Phienwej

et al., 2006) or the First Metro Line of Ho Chi Minh City (Hieu et al., 2020), have

illustrated the relationship between the maximum surface settlements and the shield

face pressure, so that larger settlements tend to occur when the shields were operated at

lower face pressures. Nagel et al. (2012) analyzed the time-dependent ground

deformations due to tunneling in soft and water-bearing soil, showing that an ‘optimal’

range of support pressure can be chosen so as to avoid large surface settlements (for

low face support values) or surface heaves (for large face support values).

1.2.3.3 Drainage measures

The required face support pressure in tunnels under the water table cannot usually be

provided by a mechanized excavation system, or by other conventional support

measures. Therefore, alternative treatments to alleviate the seepage force ahead of the

tunnel face, such as drainage measures, are required (Anagnostou, 2014). Different

drainage measures have been applied in practice, such as the drainage boreholes

employed in Lake Mead Intake No 3 Tunnel (Anagnostou and Zingg, 2013) or in

Shenzhen Metro (Tang et al., 2017); pilot tunnels (see e.g., Zingg et al., 2013; Jurík et

al., 2017) or drainage curtains (Zingg, 2016) have also been employed in other projects.

Such drainage measures introduce elements with high permeability, so that the water

13
 1. Introduction

ahead of the tunnel face can flow out through these channels easily (Li et al., 2012),

hence reducing the seepage force acting on the tunnel face. For example, Hong et al.

(2007) studied, by means of numerical simulations, the effects of horizontal pre-

drainage boreholes on the pore water pressures and on the flow rates that occurred in

subsea tunnels. They demonstrated the relevance of drainage, also giving suggestions

on the length of drainage boreholes required for different ground permeabilities.

The literature about analytical solutions to investigate the stability of a tunnel face with

drainage measures is very scarce. Only the classical Limit Equilibrium mechanism

formed by a lower wedge and an upper prism (Anagnostou and Kovári, 1994, 1996)

has been extended by Zingg (2016) to study the effect of the drainage on tunnel face

stability, in a thorough analysis of the effectiveness of drainage boreholes and of other

designs (pilot tunnels, twin tunnels, drainage curtains, etc.). The stability of the tunnel

face in homogenous and non-homogenous permeability fields was also studied, in the

framework of the Limit Equilibrium method by Zingg and Anagnostou (2016, 2018).

On the contrary, Limit Analysis has never been employed to study the stability of a

tunnel face considering the influence of drainage.

The effect of drainage measures on surface settlements has been seldom addressed.

Tang et al. (2017) studied the surface settlements induced by the excavation of a

shallow tunnel with drainage boreholes, using a case history and a numerical model.

They observed that the safety of the tunnel face is improved after introducing the

drainage boreholes, but the total settlement was found to be slightly larger than without

advance drainage. The relationship between settlements and support pressure,

considering also the drainage measures applied to improve the stability of the tunnel

face, has not yet been studied in the literature.

14
 1. Introduction

1.3 Objectives

This thesis aims to analyze the following advanced aspects of tunnel face stability: the

free span, reinforcement and drainage. It also considers the effect of drainage on the

tunnel face stand-up time and on the surface settlement.

The specific objectives of this thesis, in relation to each aspect, can be summarized as

follows:

 Tunnel face stability considering the free span. In the framework of the Limit

Analysis method, this thesis aims to propose a collapse mechanism that allows

one to study the effect of the free span on tunnel face stability.

 Tunnel face stability considering reinforcement measures. A forepole umbrella

is chosen to investigate the effect of reinforcement measures on the stability of

the tunnel face. This work aims to extend the free span mechanism, to be able

to analyze the effects of reinforcement on the critical collapse pressure and on

the failure geometry.

 Tunnel face stability considering drainage measures. Drainage boreholes are

chosen as the advance drainage measure studied in this work. Under steady-

state conditions, the aim is to analyze the effect of drainage on the stability of

the tunnel face, through a collapse mechanism in the framework of the Limit

Analysis method; under transient conditions, the goal is to study the effect of

drainage on the tunnel face stand-up time and on the surface settlements.

15
 1. Introduction

1.4 Structure of the thesis

This PhD thesis is structured into seven chapters, including this one.

In Chapter 2, the methodologies employed in this thesis are introduced, dividing them

into two categories: (i) analytical methods, including the Limit Analysis method and

the Limit Equilibrium method; and (ii) numerical methods, based on the Finite Element

Method (OptumG2) and on the Finite Difference Method (FLAC3D). Also, some key

procedures are described in this Chapter, to help readers better understand the research

tools employed in this thesis.

In Chapter 3, a 2D face failure mechanism for tunnels with a free span is presented. The

mechanism comprises three blocks: a rotational block ahead of the tunnel face, a

translational block above the free span, and a transitional block connecting the two

previous blocks. The 2D failure mechanism is an upper bound limit analysis solution

in the framework of Limit Analysis, and it can be employed to compute the collapse

pressure considering the length of the free span as a variable. A numerical validation is

carried out using a Finite Element Limit Analysis code (OptumG2), considering both the

geometry of the failure mechanism and the associated collapse pressure.

In Chapter 4, the influence of a forepole umbrella to reinforce a tunnel face with a free

span is studied. Two failure modes of the forepole umbrella are considered, and they

are employed to extend the 2D Limit Analysis mechanism for tunnels with a free span.

Using this solution, the collapse pressures and failure geometries of tunnel faces with

different forepole umbrella designs are compared.

In Chapter 5, the effect of drainage on the stability of the tunnel face under steady-state

conditions is studied. Pore water pressure distributions under steady-state conditions

16
 1. Introduction

for different configurations (e.g., water level, tunnel overburden, drainage boreholes

layout) are computed using the Finite Difference code (FLAC3D). Such pore water

pressures can then be extracted and interpolated to the positions required by a face

collapse Limit Analysis mechanism, hence allowing us to consider the influence of pore

water pressure on tunnel face stability. A numerical validation is conducted using

FLAC3D, considering both the failure geometry and the collapse pressure. Finally, the

results obtained with the proposed methodology are compared with the results of a

Limit Equilibrium solution.

In Chapter 6, the evolution of pore water pressures under transient conditions is

investigated, considering different hydraulic configurations (i.e., with or without

advance drainage). The stand-up time of the tunnel face, and the ground surface

settlements produced after tunnel excavation (as a function of the applied face support

pressures applied), are computed using FLAC3D. Finally, the effects of advance

drainage on tunnel face stability and on settlements are discussed.

In Chapter 7, a summary of the main contributions of this thesis, and some suggestions

for future work are presented.

1.5 Research output

The main results of this thesis have been published in related journals or presented in

international conferences.

Chapter 3 and 4 have been published in:

 Senent, S., Yi, C. and Jimenez, R., 2020. An upper bound solution for tunnel face

stability analysis considering the free span. Tunnelling and Underground Space

Technology, 103, 103515.

17
 1. Introduction

Chapter 5 has been published in:

 Yi, C., Senent, S. and Jimenez, R., 2019. Effect of advance drainage on tunnel face

stability using Limit Analysis and numerical simulations. Tunnelling and

Underground Space Technology, 93, 103105.

and presented in:

 Yi, C., Senent, S. and Jimenez, R., 2019. Effect of advance drainage on the stability

of tunnel faces under the water table. Congress on Numerical Methods in

Engineering, CMN 2019, Guimaraes, Portugal.

The main results of Chapter 6 have been accepted to be presented in:

 Yi, C., Senent, S. and Jimenez, R., 2020. Tunnel face stability considering drainage

and surface settlements. 10th International Symposium on Geotechnical Aspects of

Underground Construction in Soft Ground, Cambridge, UK.

18
 2. Methodology

2. Methodology

2.1 Introduction

This thesis employs several methodologies to address the problems discussed in the

previous chapter. To study stability problems in soil mechanics, three well-known

analytical techniques can be employed: the Slip-line method, the Limit Equilibrium

method, and the Limit Analysis method. This thesis investigates the applicability of

Limit Analysis to analyze tunnel face stability problems, and highlights its advantages

in comparison with other approaches. The Limit Equilibrium method is also used and

compared with the proposed Limit Analysis method.

With the arrival of high-performance computers, several codes have been produced to

facilitate numerical simulations. In this thesis, FLAC2D/FLAC3D (Fast Lagrangian

Analysis of Continua in 2 or 3 Dimension) and OptumG2/OptumG3 (Finite Element

Limit Analysis in 2 or 3 Dimension) are employed. Numerical methods are used to

validate the results of the Limit Analysis analytical solutions (i.e., to compare critical

pressures and failure geometries computed), as well as to estimate or compute other

aspects–such as the thickness of the continuous deformation zone when forepole

umbrellas are considered, or pore water pressure distributions–needed for the analyses.

19
 2. Methodology

These methodologies are further discussed below.

2.2 Analytical methodologies

Two analytical methods are employed in this thesis to analyze the stability of the tunnel

face: (i) the Limit Analysis method and (ii) the Limit Equilibrium method.

2.2.1 Limit Analysis method

The Limit Analysis method is a common tool to solve geotechnical problems, given its

capability to always obtain a realistic value of the collapse load, independently of the

geometry or the loading condition of a problem (Chen, 1975). As indicated in Chapter

1, the Limit Analysis method has been employed in this thesis to study advanced tunnel

face stability problems, taking the free span and seepage into account.

There are two main theorems to compute limit loads by the Limit Analysis method: (i)

the lower bound theorem and (ii) the upper bound theorem. The lower bound theorem

states that, in a hypothetical statically admissible stress field, the external load is not

greater than the actual limit load, as long as the system fulfills the equilibrium equations

and the boundary conditions, and the stresses lie inside or on the yield surface in stress

space. An infinite number of limit load values can be obtained by the lower bound

theorem, with the largest one being closest to the actual limit load. For the upper bound

theorem, a kinematically admissible velocity field (i.e., which satisfies compatibility

condition, the flow rule and the velocity boundary conditions) is assumed, so that the

power of its external forces is shown to be greater than (or equal to) the dissipated

power inside the system when the structure fails. (As the failure mechanism has to

satisfy the associated flow rule, the velocity vectors must form an angle φ with the

velocity discontinuity surfaces–i.e., it has to fulfill the normality condition–). Again,

20
 2. Methodology

the number of limit load solutions obtained by the upper bound theorem can also be

infinite, with the smallest one being closest to the real solution. Hence, the actual limit

load of a structure analyzed with both theorems ranges between the lower bound

solution and the upper bound solution so that a more accurate solution is obtained for

narrower ranges.

In this work, the Limit Analysis method solutions proposed for the tunnel face stability

problem are upper bound solutions. However, although the upper bound theorem is

used, the direction of the support pressure is opposite to the velocity of the collapse

mechanism so that the obtained critical support pressure is a lower bound of the actual

limit load. Therefore, a collapse mechanism that provides a higher value of the collapse

pressure, over previous solutions, will be an improvement.

In this thesis, a Limit Analysis failure mechanism that considers the free span in a 2D

tunnel face stability analysis, is proposed first; it is a simplified mechanism that

improves the previous solution by Senent and Jimenez (2017). The new mechanism is

formed by one rotational block ahead of the tunnel face and two translational blocks

(one above the free span and the other connecting the other two blocks). The details

about the failure geometry and formulation of this mechanism are presented in Section

3.2.

The second Limit Analysis mechanism employed introduces the reinforcement effect

of a forepole umbrella into the Limit Analysis framework. The work of the resistant

forces (i.e., axial force, shear force and bending moment) applied by the forepole

element is integrated into the formulation to study the effect of reinforcement on the

stability of the tunnel face. The new formulations are explained in detail in Section 4.2.

21
 2. Methodology

The third Limit Analysis mechanism is used to analyze the effect of drainage on tunnel

face stability. The contribution of drainage boreholes is incorporated into the rotational

failure mechanism proposed by Mollon et al. (2011a), integrating the influence of

seepage forces on the stability of the tunnel face according to Pan and Dias (2016). The

new formulations can be found in Section 5.2.

All of these analytical mechanisms are implemented in Matlab (The MathWorks, 2012)

(see codes in Appendices A and B). In addition, a previous 2D Limit Analysis

mechanism that considers the free span proposed by Zhang et al. (2018) is also used for

comparison with the proposed methodology. The mechanism by Zhang et al. (2018)

consists of a trapezoid (or triangular) shear sliding block above the tunnel crown, and

of a block sliding along a spiral shear line ahead of the tunnel face (see Figure 2.1).

Figure 2.1. Limit analysis failure mechanism for tunnels with a free span (after
Zhang et al., 2018)

2.2.2 Limit Equilibrium method

As indicated by Chen et al. (2015), the Limit Equilibrium method is often used in

tunneling due to its simplicity. The Limit Equilibrium method obtains an approximate

22
 2. Methodology

solution of the factor of safety, by assuming a failure mechanism and balancing the

driving and resistant forces. It often involves searching for the critical failure surface,

via an optimization process. However, although the Limit Equilibrium methodology

shares some aspects with the upper bound solutions of Limit Analysis, it does not meet

the precise requirements of the upper bound theorem, and therefore the obtained result

is not an actual bound (Chen, 1975).

In this thesis, the tunnel face limit equilibrium mechanism published by Zingg (2016)

for tunnels under the water table is adopted to compare its results with the results of the

Limit Analysis methodology considering drainage measures. This mechanism

comprises a wedge at the tunnel face, and an overlying prism reaching the ground

surface, as shown in Figure 2.2. The geometry of the mechanism depends on the angle

ω that defines the geometry of the lower wedge, so that the critical failure geometry is

determined by maximizing the collapse pressure with respect to ω.

Figure 2.2. Limit Equilibrium failure mechanism for tunnels under the water
table (after Zingg, 2016)

23
 2. Methodology

2.3 Numerical methods

In this thesis, two numerical approaches are employed: (i) the Finite Element Limit

Analysis Method as implemented in OptumG2; (ii) the Finite Difference Method as

implemented in FLAC3D.

2.3.1 The Finite Element Method (OptumG2)

OptumG2 is a geotechnical analysis software that applies the Finite Element Limit

Analysis Method (OptumCE, 2019). In this thesis, it is used for two main goals: (i)

validation of the new Limit Analysis failure mechanism that considers the free span

(see details in Section 3.3); and (ii) estimation of the thickness of the continuous

deformation zone developed when a forepole umbrella is employed (see details in

Section 4.3).

OptumG2 employs the variational principles of Limit Analysis to solve stability

problems, with the advantage with respect to other analytical Limit Analysis

approaches that stress fields and failure geometries are “optimized” due to the use of a

finite element formulation. This code implements the finite element formulation of the

limit theorems of classical plasticity, as described in Sloan (1988, 1989), Lyamin and

Sloan (2002a, 2002b). OptumG2 provides “narrow” upper bound and lower bound limit

analysis solutions, and it is used as a convenient tool to obtain the critical tunnel face

pressure. Furthermore, the failure geometry can be obtained from the shear dissipation

distribution and compared to the analytical Limit Analysis mechanism. (The shear

dissipation contour also allows one to measure the thickness of the continuous

deformation zone, which is required to compute the ultimate load of the forepole

umbrella.)

24
 2. Methodology

2.3.2 The Finite Difference Method (FLAC3D)

FLAC3D implements an explicit Lagrangian Finite Difference formulation to model the

behavior of geomaterials (Itasca Consulting Group, 2009), and it is extensively used in

geotechnical modeling (plastic collapse, flow, etc.), due to its availability of soil

constitutive models and its good modeling performance. In this work, FLAC3D is used:

(i) to obtain pore water pressure distributions under steady-state conditions (see Section

5.3); (ii) to compute tunnel face collapse pressures and failure geometries to validate

the Limit Analysis results in tunnels under the water table (see Section 5.4); (iii) to

compute pore water pressures under transient conditions, and to compute settlements

with a coupled analysis that considers the face pressure (see Sections 6.4 and 6.5).

Three important aspects of FLAC3D modeling as applied in this thesis, are discussed in

the following sections.

2.3.2.1 Computation of pore water pressures under steady-state conditions

The 3D Limit Analysis mechanism to study the influence of advance drainage in tunnels

under the water table requires pore water pressure distributions along the failure region.

FLAC3D is employed to calculate pore water pressures numerically; in particular,

FLAC3D provides the pore water pressure distributions associated with seepage flow

under steady-state conditions by means of a “flow-only” analysis, in which the pore

water pressures at the excavation face are fixed to be zero and a fast water recharge is

assumed as the seepage boundary conditions (see the codes in Appendix C-

Modelo_V01.dat). The obtained pore water pressure distribution is also used to

numerically compute the critical support pressure in a mechanical calculation, in which

pore pressures are assumed to remain constant.

25
 2. Methodology

2.3.2.2 Computation of critical support pressures and failure geometries

FLAC3D cannot directly provide the critical support pressure of the tunnel face. The

improved bisection method, proposed by Mollon et al. (2009a), has been adopted to

compute the critical pressure in the numerical model, as schematically shown in Figure

2.3 (see also the FLAC3D codes in Appendix C- Modelo_V02.dat and Intervalo.dat).

The computation starts with two reasonable initial face pressures Pinf and Psup,

corresponding respectively to pressures for which the tunnel face is unstable and stable.

Next, the stability of the face is computed assuming the mean value (P= (Pinf + Psup)/2).

If the tunnel face is unstable (i.e., the unbalanced mechanical-force ratio Runbal is bigger

than a prescribed reference value Rref), Pinf is replaced by such mean value; otherwise,

Psup is replaced. By repeating the computation until a specific accuracy ε is reached, the

critical support pressure can be obtained.

Figure 2.3. Flow chart of the bisection method in FLAC3D to compute the critical
support pressure

26
 2. Methodology

Similar to OptumG2, the failure mechanism ahead of the tunnel face is estimated from

the distribution of shear deformations of the FLAC3D numerical results, so it can be

used to validate the critical failure geometry obtained by the Limit Analysis mechanism.

2.3.2.3 Computation of critical support pressures and settlements under transient

conditions (Fluid-solid coupled analyses)

Due to the time-dependent characteristics of seepage under transient conditions, and to

the interactions between pore water pressures and soil behavior and deformation, an

“only flow” analysis under transient conditions is not acceptable. To analyze the tunnel

face stability in this situation, the FLAC3D numerical model must employ the “fluid-

solid coupled” mode, in which the flow is simulated in parallel to the mechanical

calculations. Such fluid-solid interaction implies that pore pressure changes cause

volumetric strains; in turn, pore pressures can be affected by deformations (strains).

Figure 2.4 illustrates the three stages of such analysis: stress initialization, tunnel

excavation and fluid-solid coupled analysis. Different from the simulation under

steady-state conditions, the seepage boundary condition at the tunnel face under

transient conditions is a mixed boundary condition. As shown in Figure 2.4 (c), it is a

“no-flow” boundary if it has negative pore pressures (pw<0, qw=0), and a “seepage face”

otherwise (pw=0, qw>0) (see also Anagnostou et al., 2016). With this mixed boundary

condition, the simulation can be carried out to analyze the behavior of the excavation,

registering the evolution of surface displacements (i.e., settlements) with time.

The critical support pressure under transient conditions cannot be obtained with the

bisection method as it was done under steady-state conditions, due to the high

computational cost of the required simulations. Under transient conditions, the stability

of a tunnel face is assessed according to the time (called the stand-up time) during which

27
 2. Methodology

the tunnel face is stable, before the acceleration of displacements associated with failure.

Consequently, the critical support pressure corresponds to the pressure for which the

stand-up time tends to infinity. Therefore, the FLAC3D model is run (under the fluid-

solid coupled mode and for a set of applied face support pressures), and the critical

pressure is estimated based on the relationship between the applied support pressure

and the stand-up time.

a)

b)

c)
Figure 2.4. Process for fluid-solid coupled analysis in FLAC3D to study the time-
dependent behaviors of tunnel face stability

28
 3. Tunnel face stability analysis considering the free span

3. Tunnel face stability analysis

considering the free span

3.1 Introduction

The excavated tunnel face together with the free span greatly impact the tunneling

construction safety. HSE (1996) summarized the heading failure mechanisms through

real-life experiences in London clay. Figure 3.1 presents four of them, for which the

tunnel cover is sufficient so that the failure does not reach the surface. This figure also

shows that the free span is an important factor that affects the heading failure

mechanism. Similarly, Standing and Burland (2006) indicated that the unsupported

length has a great influence on the volume losses and, accordingly, on the required

support pressure to be applied on the tunnel face to control its movements. Despite the

above, most research on the stability of the tunnel face considers a completed unlined

tunnel (i.e., an infinite free span; see e.g., Davis et al., 1980), or a completely lined

tunnel (i.e., with a null free span; see e.g., Leca and Dormiuex, 1990).

This chapter investigates the effects of the free span on the tunnel face stability

(assuming that the mechanism does not outcrop at the ground surface) in terms of (i)

the collapse pressure; and (ii) the failure geometry ahead of the tunnel face and over the

29
 3. Tunnel face stability analysis considering the free span

free span. From the description of the problem setup (Section 3.2.1), a new failure

mechanism for tunnel faces with a free span is proposed in the framework of the Limit

Analysis method (Section 3.2.2) and its formulation is described in detail (Section

3.2.3). To validate the new analytical mechanism, a numerical model is used to compare

the face collapse pressure and the failure geometry obtained from both approaches in a

series of test cases (Section 3.3).

a) b)

c) d)

Figure 3.1. Four heading failure mechanisms with free span in NATM a) bench
failures; b) crown failures; c) full-face failures d) local face failures (HSE, 1996)

3.2 Tunnel face stability 2D mechanism considering the free span

3.2.1 Model description

A tunnel excavation of height H under a cover depth C in a Mohr-Coulomb material,

defined by its cohesion (c), friction angle (φ) and unit weight (γ), is considered. Note,

therefore, that the analysis is two-dimensional, corresponding to a longitudinal section

of the excavation; but interesting outcomes can be obtained from this type of analyses

30
 3. Tunnel face stability analysis considering the free span

(e.g., Mollon et al. 2011b), and it could be particularly useful for other underground

works, such as long-wall mining (e.g., Yan et al., 2020). Figure 3.2 illustrates more

details of an analyzed tunnel: the length of unsupported excavation (or free span) is L

and a uniform support pressure σT is applied on both the tunnel face and on the free

span. Nonetheless, the case with an unsupported free span is also considered in the

discussion.

Figure 3.2. Problem setup for studying the stability of a tunnel with free span

3.2.2 Collapse mechanism

Figure 3.3 shows the proposed Limit Analysis mechanism to study the stability of the

tunnel face considering the free span. It is formed by three non-deformable blocks: (i)

a translational triangular block above the free span (Block 1); a rotational block ahead

of the tunnel face (Block 2); and a translational block above them that closes the

mechanism (Block 3).

31
 3. Tunnel face stability analysis considering the free span

Figure 3.3. Outline of the proposed Limit Analysis mechanism to compute tunnel
face stability considering the free span

The failure of the material above the free span is analogous to the failure mechanism in

the trapdoor problem (Terzaghi, 1936), so that Block 1 is similar to the mechanism

proposed by Evans (1984) for this experiment. In this thesis, because of the boundary

conditions –i.e., the tunnel face–, an asymmetrical triangular block (Block 1), with a

vertically descending velocity (V1), is proposed. Its external boundary DC forms an

angle φ with the vertical to fulfill the Limit Analysis hypothesis of associated flow.

Another parameter, β1, is introduced to define its internal boundary with Block 3. This

simplifies the mechanism originally proposed by Senent and Jimenez (2017), in which

a symmetrical triangular block was used, hence requiring one more block on its side.

The proposed asymmetrical block eliminates the need for such additional block, so that

the mechanism becomes simpler, and its results are more similar to numerical and

experimental results (see e.g., Costa et al., 2007; Lee and Schubert, 2008). Block 2

32
 3. Tunnel face stability analysis considering the free span

suffers a rotational movement, with angular velocity ω2, around a center O2 whose

position is defined by two variables (βE, RE). Its geometry is similar to the 2D rotational

mechanism proposed by Mollon et al. (2011b), which has been shown to outperform

other translational mechanisms, both in the computed collapse pressure and in the

failure geometry. In this case, the lower boundary of Block 2 (curve BF) is a logarithmic

spiral of parameter φ that starts in the lower corner of the excavation (Point B); its upper

boundary (curve AF) is generated by an iterative process, as explained later,

considering the velocity of Block 3. The position of this internal boundary between

Blocks 2 and 3 is determined by a new parameter (α1) that defines its inclination in

Point A. Finally, the translational Block 3 is characterized by a velocity (V3), towards

the tunnel face, that forms an angle 𝛼 with the horizontal. The angle in the upper corner

of Block 3 (point G) is 2φ to fulfill the associated flow requirement in the external

boundary of the mechanism.

Considering the previous explanation, the mechanism is defined by 5 parameters (β1,

βE, RE, α1, α2). Given that the proposed analytical solution is an upper bound Limit

Analysis solution, and that the face pressure acts against the movement of the

mechanism, the collapse pressure must be maximized in relation to these variables

(Chen, 1975).

3.2.3 Formulation of the analytical solution

3.2.3.1 Coordinate systems, velocity field and geometry construction

Two coordinate systems are employed to produce the mechanism: (i) a Cartesian

system (Y, Z) with origin in Point A; and (ii) a polar system (R, β) with origin in O2 –

i.e., in the center of rotation of Block 2–. Since the location of O2 is defined by

parameters βE and RE, its Cartesian coordinates are:

33
 3. Tunnel face stability analysis considering the free span

𝑌 = 𝑅 ∙ cos(𝛽 ) − (3.1)

𝑍 = −𝑅 · sin (𝛽 ) (3.2)

The proposed free span mechanism comprises three blocks. Hence, it is necessary to

define the relative velocities between blocks, in such a way that all terms in the energy

balance equations are expressed as a function of a unique velocity that could be later

removed from the equations. The relative velocity between Blocks 1 and 3 (V13) must

form an angle with their interface (line AD) equal to the friction angle. Since the

interface orientation is defined by the angle β1, velocities V3 and V13 can be calculated

with the hodograph (Figure 3.4(a)) employing the following equations:

( )
𝑉 = ( )
∙𝑉 (3.3)

𝑉 = ∙𝑉 (3.4)
( )

Similarly, the angular velocity of Block 2 (ω2) can be obtained by imposing an

associated flow rule in the initial point of the interface between Blocks 2 and 3 (Point

A). Since, at Point A, the interface forms an angle α1 with the horizontal, and the

velocity of Block 2 in Point A is perpendicular to 𝑂 𝐴, the angular velocity of Block 2

can be expressed as (Figure 3.4(b)):

( )
𝜔 = ( )
· (3.5)

where RA and βA are the polar coordinates of point A:

𝑅 = 𝑍 +𝑌 (3.6)

𝛽 = 𝑡𝑎𝑛 (− ) (3.7)

34
 3. Tunnel face stability analysis considering the free span

a)

b)

c)

Figure 3.4. Hodographs: a) interface between Blocks 1 and 3; b) in Point A

between Blocks 2 and 3; c) interface between Blocks 2 and 3

The boundaries of Block 2 are not straight lines, and they must be computed first to

obtain Point F (see Figure 3.3). The logarithmic spiral emerging from point B can be

expressed in polar coordinates as:

35
 3. Tunnel face stability analysis considering the free span

𝑅 = 𝑅 ∙ exp ((𝛽 − 𝛽 ) tan𝜑) (3.8)

The interface AF is generated point by point starting from Point A and, discretizing it

in segments of length s, following the process shown in Figure 3.4(c). For each point

Pi of the interface (yi, zi in Cartesian coordinates), its velocity corresponding to Block

2 is 𝑉 ,⃗̇ and forms an angle of 90° with 𝑂 𝑃 ̇ :

𝑉 ,⃗̇ = (𝑦 − 𝑌 , 𝑧 − 𝑍 ) ∙ 𝜔 (3.9)

Since the relative velocity between Blocks 2 and 3 in Point 𝑃 is 𝑉 ,⃗̇ = 𝑉 ,⃗̇ − 𝑉⃗, and

since this velocity must form in Point Pi an angle with the interface equal to φ, the unit

vector of the next segment (of length s) of the interface is obtained rotating an angle φ

the vector 𝑉 ,⃗̇ :

cos 𝜑 sin 𝜑 ,⃗̇

𝑢⃗̇ = ∙ (3.10)
−sin 𝜑 cos 𝜑 ,⃗̇

Hence, the next point of the interface (Pi+1) can be computed as:

𝑃 ̇ 𝑂 ⃗ = 𝑃 ̇ 𝑂 ⃗ − 𝑢⃗̇ ∙ 𝑠 (3.11)

Finally, Point F (βF, RF in polar coordinates) can be computed as the intersection of the

two previous curves (BF and AF).

The remaining boundaries of the mechanism are straight lines, and they are easy to

obtain. Starting from the coordinates of Point C (-L,0), Point D can be obtained as the

intersection of two straight lines, the first emerging from Point C, with an inclination

of (π/2-φ), and the second from Point A, with an inclination of β1. Then, the coordinates

of Point D are:

𝑌 = −𝑍 · tan(𝛽 ) (3.12)

𝑍 = ·
(3.13)

36
 3. Tunnel face stability analysis considering the free span

Similarly, the coordinates of Point G can be obtained as the intersection of two straight

lines emerging (i) from Point D, with an inclination of (α2-φ), and (ii) from Point F,

with an inclination of (π-φ-α2). Consequently, the coordinates of Point G are:

( )· ( )·
𝑌 = ( ) ( )
(3.14)

𝑍 = 𝑍 + cot(𝛼 − 𝜑) · (𝑌 − 𝑌 ) (3.15)

Once all the internal and external boundaries of the mechanism have been obtained, it

is possible to compute the area of each block and the length of each boundary, and to

formulate the Limit Analysis energy balance needed to compute the collapse pressure.

3.2.3.2 Work equation

The external forces applied on the failure mechanism comprise the soil weight, and the

support force applied on the tunnel face and on the free span. It is assumed that the

mechanism never outcrops at the surface, so that the ground surcharge is not considered

in the formulation.

The rate of work of the soil weight is:

𝑊 = ∭ 𝛾⃗ · 𝑣⃗ · 𝑑𝑉 = 𝛾 ∙ (𝑆 ∙ 𝑉 + 𝑆 ∙ sin(𝛽 ) ∙ 𝑅 ∙ 𝜔 + 𝑆 ∙ sin 𝛼 ∙ 𝑉 ) (3.16)

where Si (i=1,2,3) is the area of Block i and, βK and RK are the polar coordinates of the

barycenter of Block 2.

The rate of work of the support pressure is:

𝑊 = ∬ 𝜎 ⃗ · 𝑣⃗ · 𝑑𝑆 = −𝜎 · (𝐿 ∙ 𝑉 + 𝑅 ∙ 𝜔 ∙ 𝐻 ∙ cos𝛽 ) (3.17)

Finally, the rate of the internal energy dissipated in the proposed mechanism is due to

the plasticity work that occurs along the internal and external velocity discontinuities:

37
 3. Tunnel face stability analysis considering the free span

𝑊 = −𝑐 · cos𝜑 · (𝑙 ·𝑉 +𝑙 ·𝑉 +∑ 𝑇 ̇𝑇 ̇ ⃗ · 𝑅 · 𝜔 + ∑ 𝑃 ̇𝑃 ̇ ⃗ ·

𝑉 , +𝑙 ·𝑉 +𝑙 ·𝑉 ) (3.18)

where lij represents the length between the points i and j in the boundary of the

mechanism (see Figure 3.3), and Ti (i=1…N) and Pi (i=1…M) are the points in which curves

BF and AF are discretized.

By equating the work rate of the external forces to the rate of internal dissipated energy,

the collapse pressure can be computed from:

𝜎 = 𝛾 ∙ (𝑆 ∙ 𝑉 + 𝑆 ∙ sin(𝛽 ) ∙ 𝑅 ∙ 𝜔 + 𝑆 ∙ sin𝛼 ∙ 𝑉 ) − 𝑐 · cos𝜑 · (𝑙 ·𝑉 +

𝑙 ·𝑉 +∑ 𝑇 ̇𝑇 ̇ ⃗ · 𝑅 · 𝜔 + ∑ 𝑃 ̇𝑃 ̇ ⃗ · 𝑉 , +𝑙 ·𝑉 +𝑙 · 𝑉 ) /(𝐿 ∙

𝑉 + 𝑅 ∙ 𝜔 ∙ 𝐻 ∙ cos𝛽 ) (3.19)

As previously pointed out, this expression must be maximized in relation to the five

variables that define the mechanism (β1, βE, RE, α1, α2).

3.3 Numerical validation

In this section, the main results of the proposed mechanism –the value of the collapse

pressure and the failure geometry– are compared to the results of a numerical simulation

conducted with the two-dimensional Finite Element Limit Analysis code OptumG2

(OptumCE, 2019).

3.3.1 Test Cases

An excavation of 10m height with a free span L varying between 0 to 3m is employed

for the validation of the proposed mechanism. (L=0 corresponds to a completely lined

tunnel, and it is used as a control case). Table 3.1 lists the 36 test cases used in the

numerical validation. Two different friction angles (30° and 35°) and three different

cohesions (5, 15 and 25kPa) are used. All the cases have the same unit weight

38
 3. Tunnel face stability analysis considering the free span

(γ=20kN/m3) and the same overburden (C=5D), which has been chosen to avoid that

the failure mechanisms reach the surface.

3.3.2 Numerical model

Figure 3.5 illustrates the model used to validate the analytical solution numerically. The

excavation is 10m high and the overburden is 50m. The dimensions of the model are

70m×80m, and the length of the model ahead of the tunnel face is 50m. The boundary

conditions are given by fixed displacements at the boundaries of the model, i.e., at its

lateral perimeter and at its base, as shown in Figure 3.5. Similarly, the tunnel support

has not been included and displacements at the tunnel excavation boundary have been

fixed except in the free span. (“Limit Analysis” has been selected as Analysis Type in

OptumG2. Similarly, the Element Type has been configured as “Upper” and “Lower”;

since the face pressure acts against the movement of the mechanism, this makes

OptumG2 compute, respectively, lower and upper bounds of the collapse pressure).

Figure 3.5. Geometry of the numerical model built in OptumG2

39
 3. Tunnel face stability analysis considering the free span

3.3.3 Results

3.3.3.1 Collapse pressure

Table 3.1 and Figure 3.6 list the collapse pressures obtained with the proposed

mechanism and with the numerical model. As it can be observed, the results are very

similar, in all cases with differences lower than 7.4kPa. Note also that the differences

are higher for the lower friction angle (i.e., for φ=30°) and that they are almost

independent of the cohesion. For the higher friction angle (φ=35°), the differences are

lower than 3.5kPa comparing to the results of the lower bound analysis, and lower than

2.0kPa comparing to the results of the upper bound analysis.

Figure 3.6. Comparison of collapse face pressures computed with the proposed
mechanism and with the numerical model

40
 3. Tunnel face stability analysis considering the free span

Table 3.1. Test cases for the numerical validation of the proposed Limit Analysis mechanism

Parameters Collapse Pressure [kPa]

δ
Case 𝜑 𝑐 Proposed mechanism Numerical model (1)
Difference (1)
Free span 𝐿 [m]
[°] [kPa] [m] 𝜎 , 𝜎 , 𝜎 , −𝜎 ,

1 0 33.2 35.7 - 37.0 2.5 - 3.8 1.70

2 0.5 33.6 36.7 - 38.0 3.1 - 4.4 1.74
3 1 34.2 37.8 - 39.2 3.6 - 5.0 1.83
5
4 1.5 35.0 39.2 - 40.6 4.2 - 5.6 1.86
5 2 36.0 40.6 - 42.2 4.6 - 6.2 1.87
6 3 38.7 44.2 - 46.0 5.5 - 7.3 1.97
7 0 15.8 18.4 - 19.7 2.6 - 3.9 1.70
8 0.5 16.5 19.4 - 20.6 2.9 - 4.1 1.77
9 1 16.8 20.5 - 21.9 3.7 - 5.1 1.82
30 15
10 1.5 17.6 21.8 - 23.3 4.2 - 5.7 1.90
11 2 18.7 23.3 - 24.9 4.6 - 6.2 1.83
12 3 21.3 26.9 - 28.7 5.6 - 7.4 2.00
13 0 0.0 1.0 - 2.4 1.0 - 2.4 1.73
14 0.5 0.0 2.1 - 3.4 2.1 - 3.4 1.81
15 1 0.0 3.2 - 4.6 3.2 - 4.6 1.84
25
16 1.5 0.3 4.5 - 6.0 4.2 - 5.7 1.87
17 2 1.3 6.0 - 7.6 4.7 - 6.3 1.86
18 3 4.0 9.6 - 11.4 5.6 - 7.4 1.98
(continued)

41
 3. Tunnel face stability analysis considering the free span

Parameters Collapse Pressure [kPa]

δ
Case 𝜑 𝑐 Proposed mechanism Numerical model(1) Difference(1)
Free span 𝐿 [m]
[°] [kPa] [m] 𝜎 , 𝜎 , 𝜎 , −𝜎 ,

19 0 24.3 24.9 - 25.9 0.6 - 1.6 1.16

20 0.5 24.5 25.4 - 26.3 0.9 - 1.8 1.23
21 1 24.8 25.9 - 26.9 1.1 - 2.1 1.25
5
22 1.5 25.2 26.5 - 27.6 1.3 - 2.4 1.36
23 2 25.8 27.3 - 28.5 1.5 - 2.7 1.36
24 3 27.4 29.3 - 30.6 1.9 - 3.2 1.42
25 0 10.0 10.6 - 11.6 0.6 - 1.6 1.17
26 0.5 10.2 11.1 - 12.0 0.9 - 1.8 1.25
27 1 10.5 11.6 - 12.6 1.1 - 2.1 1.25
35 15
28 1.5 10.9 12.2 - 13.3 1.3 - 2.4 1.38
29 2 11.5 13.0 - 14.2 1.5 - 2.7 1.40
30 3 13.1 15.0 - 16.3 1.9 - 3.2 1.44
31 0 0.0 0.0 - 0.0 0.0 - 0.0 -
32 0.5 0.0 0.0 - 0.0 0.0 - 0.0 -
33 1 0.0 0.0 - 0.0 0.0 - 0.0 -
25
34 1.5 0.0 0.0 - 0.0 0.0 - 0.0 -
35 2 0.0 0.0 - 0.0 0.0 - 0.0 -
36 3 0.0 0.7 - 2.0 0.7 - 2.0 1.38
(1) The left value corresponds to an upper bound analysis (Element type=Upper in OptumG2) and the right value corresponds to a lower bound
analysis (Element type=Lower in OptumG2).

42
 3. Tunnel face stability analysis considering the free span

3.3.3.2 Failure geometry

Similarly, Figure 3.7 compares the collapse geometries obtained with the analytical

solution and with the numerical model. (Failure mechanisms in the OptumG2 model

have been estimated considering the distribution of shear dissipation.) As it can be seen,

the failure geometry obtained with the proposed mechanism captures the shape of the

failure given by the numerical model ahead of the tunnel face and above the free span.

Moreover, the failure surfaces in the numerical model define, approximately, the three

blocks that compose the analytical solution; thereby, the new mechanism is an

improvement over the four blocks mechanism proposed in Senent and Jimenez (2017).

Despite the above, the mechanism in the transition zone (Block 3) seems to be

excessively sharp compared with the results of the numerical simulation. In general,

predictions of the failure mechanism geometry improve for cases with higher soil

strength.

Figure 3.7. Comparison of failure geometries obtained with the proposed

mechanism and with the numerical model for different strength parameters (φ,
c) and lengths of the free span (L)
43
 3. Tunnel face stability analysis considering the free span

These results suggest that the proposed Limit Analysis mechanism can be used to

predict the collapse pressure and the failure geometry of tunnel faces considering their

free span, especially for tunnels in materials with higher friction angles (i.e., φ≥35°).

Therefore, such a mechanism is employed next to further explore the influence of the

free span on the stability of the face.

3.4 Results and discussions

As shown in Figure 3.8, the free span affects the collapse pressure of the tunnel face.

As expected, the collapse pressure increases with the free span. Moreover, a larger free

span can cause the instability of a tunnel face that would be otherwise stable for shorter

free spans; this is illustrated by the case with c=25kPa and φ=30°, in which the face is

not self-stable for L larger than about 1m. (When the collapse pressure is null, the face

is self-stable, and it is not necessary to support it). However, the effect of the free span

is less relevant than the effect of the soil strength: a reduction in the friction angle from

35° to 30° produces, for the case of c=5kPa, an increase of the collapse pressure of

around 40%; whereas its increase due to a larger free span, from 0 to 3m, is less than

13%. These results are coherent with Vermeer et al. (2002), who obtained that there is

almost no effect of the free span for frictional materials and L/D ratios lower than 0.3.

Conversely, Anagnostou and Perazzelli (2013) show a higher effect of the free span,

even for lower L/D ratios. One reason for such different results could be that, in this

work and in Vermeer et al. (2002) the support pressure is applied both on the face and

on the free span; whereas Anagnostou and Perazzelli (2013) apply support only on the

face.

44
 3. Tunnel face stability analysis considering the free span

Figure 3.8. Collapse pressure vs length of the free span for different strength
parameters

Therefore, the proposed mechanism has also been employed to compute the collapse

pressure without considering any support applied on the free span. (Note that the

formulation presented in Section 3.2.3 continues to be valid, after removing the first

term between parentheses in Equation (3.17)). Figure 3.9 shows the results of this

analysis for the same Test Cases used in Section 3.3.1 (see Table 3.1). As shown in

Figure 3.9, the increase of the collapse pressure with L is noticeably higher for

unsupported free spans, especially for cases with lower strength parameters or, more

precisely, with lower cohesions.

45
 3. Tunnel face stability analysis considering the free span

a)

b)

Figure 3.9. Collapse pressure vs free span length for different values of cohesion
considering a supported and an unsupported free span: a) φ=30°; b) φ=35°

Figure 3.9 shows some cases for which the collapse pressure cannot be calculated with

the proposed mechanism (i.e., with an unsupported span (and L>1m), for c=5kPa (φ=30°

46
 3. Tunnel face stability analysis considering the free span

and 35°)). The reason is that the free span, when no pressure is applied, is unstable, no

matter how much pressure is applied on the tunnel face. Consequently, a maximum

length of the free span can be defined (LMAX), above which the free span is unstable if

no pressure is applied on it. Anagnostou and Perazzelli (2013), based on the silo theory,

proposed Equation (3.20) to calculate LMAX depending on cohesion (c), soil weight (γ)

and excavation width (B); Equation (3.20) can be reduced to Equation (3.21) if an

infinite excavation width is assumed.

𝐿 = (3.20)

𝐿 = (3.21)

Figure 3.10 plots the LMAX results computed with Equation (3.21) for a soil weight of

γ=20kN/m3 and compares them with the LMAX obtained with the proposed Limit

Analysis mechanism. As can be seen, the new analytical solution divided by two the

cohesion needed to keep the free span stable. This result, as expressed in Equation

(3.22), can be also derived from the Limit Analysis upper bound solution proposed by

Evans (1984) for a trapdoor problem of width L (Equation (3.23)), if a null pressure is

supposed (σT=0). The difference between both methodologies (Equations (3.21) and

(3.22)) is probably due to the associated flow rule hypothesis assumed in Limit Analysis.

𝐿 = (3.22)

𝜎 = cot 𝜑 · ·𝐿·𝛾−𝑐 (3.23)

47
 3. Tunnel face stability analysis considering the free span

Figure 3.10. Maximum unsupported span vs cohesion

Figure 3.7 also shows the effect of the free span on the collapse geometry. (In addition,

it shows that lower friction angles typically tend to produce larger mechanisms, and

that there is an almost negligible effect of cohesion on the failure geometry.) For

different free span lengths, the main variation occurs in the zone above the tunnel, as

the geometry of the mechanism ahead of the tunnel face remains almost equal. To

illustrate the influence of the free span better, Figure 3.11 directly compares the failure

mechanisms obtained with the proposed mechanism for the Test Cases used in Section

3.3.1 (see Table 3.1). Obviously, the failure always covers the free span completely,

but two other effects can be appreciated: first, the failure mechanism extends further

upwards as the free span length increases (this can be expected, because Block 3 is

triangular with a constant upper angle (2φ) and, for larger free spans, its base is bigger);

and, second, the velocity vector of Block 3 becomes slightly more horizontal for larger

free spans –i.e., angle α2 of the translational velocity of Block 3 (V3) reduces–.

48
 3. Tunnel face stability analysis considering the free span

Figure 3.11. Variation of the failure geometry depending on the free span
(L=1,2,3): a) φ=30°, c=5kPa; b) φ=30°, c=15kPa; c) φ=35°, c=5kPa; d) φ=35°,
c=15kPa

As previously mentioned, Zhang et al. (2018) studied the effect of the free span on the

stability of the tunnel face using a Limit Analysis mechanism. Figure 3.12 compares

the collapse pressures computed (i) with the mechanism proposed by Zhang et al.

(2018), (ii) with our mechanism, and (iii) with a numerical simulation in OptumG2

(Upper Bound Analysis) as a function of the free span. Results are computed for two

values of the friction angle: φ=20° (Figure 3.12(a); see Fig.14 of Zhang et al., 2018);

and φ=40° (Figure 3.12(b)). (Zhang et al. (2018) considered different values of

pressures being applied on the tunnel face and on the free span: for the results in Figure

3.12(a), the pressure on the free span is constant and equal to 50kPa; whereas in Figure

3.12(b) the pressures on both the free span and on the tunnel face are equal). Results

49
 3. Tunnel face stability analysis considering the free span

show that trends computed with both models are similar (i.e., there is an increase of the

collapse pressure with the free span), but our mechanism provides higher critical

pressures than those computed with the mechanism proposed by Zhang et al. (2018).

Consequently, since both are upper bound limit analysis solutions, our mechanism

improves this previous mechanism that also considers the free span.

a)

b)

Figure 3.12. Collapse pressures computed, as a function of the free span, with (i)
the proposed mechanism, (ii) the Limit Analysis methodology presented by
Zhang et al. (2018), and (iii) an Upper Bound Analysis in OptumG2. (D=10m;
c=5kPa; γ=18kN/m3): a) φ=20° (the pressure applied on the free span is constant
and equal to 50kPa); b) φ=40° (the pressures applied on the free span and on the
tunnel face are equal)

50
 3. Tunnel face stability analysis considering the free span

Figure 3.12 also shows that, for a friction angle equal to 20° (i.e., significantly lower

than 35°), the results of our mechanism are noticeably lower than those computed with

the numerical model; in contrast, both approaches provide very similar collapse

pressures for the case with higher friction angle (40°), hence suggesting that the

proposed analytical solution can be used to accurately predict the collapse pressure of

tunnel faces with a free span in materials with higher friction angles (i.e., φ≥35°).

3.5 Conclusions

In this chapter, a simplified 2D Limit Analysis mechanism, based on the one employed

by Senent and Jimenez (2017), is proposed to study the face stability of tunnels

constructed in Mohr-Coulomb materials. The main contribution of the new mechanism

is that it considers the influence of the free span (or the length of the tunnel without

support); it also provides more similar results to the numerical simulations and to the

experimental tests.

The mechanism is formed by three non-deformable blocks (two translational and one

rotational) in such a way that the interface between one of the translational blocks and

the rotational block is obtained by an iterative process that assures that the associated

flow rule imposed by Limit Analysis is fulfilled.

Results of the proposed mechanism are compared with results of a numerical model

built with the Finite Element Limit Analysis code OptumG2, showing that the proposed

analytical solution can be employed to reasonably predict (i) the value of the collapse

pressure, and (ii) the collapse geometry, for tunnels in materials with higher friction

angles (i.e., φ≥35°).

51
 3. Tunnel face stability analysis considering the free span

Results confirm the expected effect of the free span on the collapse pressure, as larger

free spans entail higher collapse pressures. However, the influence of the free span is

also shown to significantly depend on the assumption about how the support pressure

is applied, as the behavior can change significantly when (a) the support pressure is

applied only on the tunnel face but not on the free span, as opposed to (b) when the

support pressure is applied both on the tunnel face and on the free span.

The tunnel face can be self-stable up to a maximum free span length LMAX, above which

failure cannot be avoided no matter how much support pressure is applied on the tunnel

face. LMAX mainly depends on the soil properties: cohesion and unit weight. Moreover,

for a given set of soil properties, the Limit Analysis method provides the maximum

unsupported length so that the tunnel face is stable.

The failure geometry varies with the length of the free span, so that the free span length

mainly affects the failure geometry above the tunnel crown, as higher and more

horizontal mechanisms are produced for larger free spans. However, the influence of

the free span on the geometry of the rotational block that develops ahead of the tunnel

face is almost negligible, and results have shown that changes of this block are mainly

due to the well-known variation of the shape of the rotational failure surfaces that occur

associated to changes of the ground friction angle (see e.g., Mollon et al. 2011a).

52
 4. Effect of a forepole umbrella on the stability of the tunnel face

4. Effect of a forepole umbrella on

the stability of the tunnel face

4.1 Introduction

The unsupported advance length (i.e., the free span) can vary from half a meter to

several meters depending on the ground (strength parameters, homogeneity, stress

state), the system of excavation, and the support measures employed to stabilize the

excavation (Lunardi, 2008). For tunneling in soft grounds, the forepole umbrella

(Figure 4.1) is an efficient soil reinforcement method materialized by steel pipes, or

sometimes by grout injected via pre-perforated steel pipes (Le et al., 2015). FHA (2009)

stated that the forepole umbrella is a common measure in NATM for cohesive grounds

but is considered to be infeasible for sandy grounds. Figure 4.1 illustrates an example

of the structural layout of a forepole umbrella in an ongoing tunnel excavation. Lunardi

(2008) proposed that the reinforcement effect of a forepole umbrella is to provide a

series of “beams” that rest on the rib steel sets and on the ground, that help preserve the

integrity of the advance core during tunneling construction. However, the size and

quantity of steel pipes chosen for a particular tunnel usually were based on empirical

guidelines, or the contribution of the forepole umbrella on the stability is simply not

53
 4. Effect of a forepole umbrella on the stability of the tunnel face

defined (Takeuchi et al. 1999). Consequently, more studies are needed to enable

engineers to design the reinforcement measure in a methodical way.

a) b)

Figure 4.1. Forepole umbrella layout in an ongoing excavation tunnel. a)

longitudinal profile; b) Section A:A’ (after Lunardi, 2008)

Chapter 4 aims to improve our understanding of the influence of the forepole umbrella

on tunnel face stability developing a methodology to incorporate it into the Limit

Analysis mechanism proposed in Chapter 3. Section 4.2 describes the extensions

required for the formulation proposed in Section 3.2, considering two failure modes of

the forepole umbrella. Then, in Section 4.3 the effect of the umbrella on tunnel face

stability is studied, considering both the collapse pressure and the failure geometry.

4.2 Incorporation of a forepole umbrella into the Limit Analysis

tunnel face stability mechanism

Introducing the effects of a forepole umbrella into a Limit Analysis upper bound

solution is not straightforward, since the failure mode of the umbrella must be defined

in advance, so that its contribution to the work equation can be computed. Unlike nail

reinforcements that work by axial force only, and that can fail by tension or by pull-out

54
 4. Effect of a forepole umbrella on the stability of the tunnel face

from the soil (Michalowski, 2005), forepole umbrellas typically resist forces

perpendicular to its axis, and they, therefore, work by shear and by bending –i.e., like a

beam–. A similar problem can be found in slope stability analyses considering anti-

slide piles (e.g., Rao et al., 2017); however, in that case, the vertical piles receive

horizontal forces that depend on displacements (e.g., Winkler model), whereas in the

tunnel face stability problem the support element is subhorizontal and receives the

weight of the ground over it. Pinyol and Alonso (2011), for undrained soils, stated that

the micropiles forming the umbrella are characterized by a limiting (yielding) bending

moment, and that the soil reacts against the micropile with a force per meter of

micropile given by 9cud (Broms, 1964), where cu is the undrained strength of the soil

and d the diameter of the micropile. In Pinyol and Alonso (2013) the micropile is

idealized as a beam subjected to an imposed displacement, considering that it fails when

its tensile strength is reached, and considering no contribution of the bending moment

to the stabilizing work.

In this work, the methodology proposed by De Buhan and Salençon (1993) for slopes

is used to study the effect of a forepole umbrella on the face stability of tunnels

excavated in Mohr-Coulomb materials. Two failures modes are considered, assuming

perfect bonding between the soil and the structural element:

(i) The structural element would fail at the velocity jump surface defined by the Limit

Analysis mechanism (Figure 4.2(a)), and the maximum resistance energy produced

by it can be expressed as (De Buhan and Salençon, 1993):

𝑊 = 𝑉⃗ · [𝑁 · sin (𝜑 − 𝛼) + 𝑉 · cos (𝜑 − 𝛼)] /

(4.1)

where 𝑉⃗ is the velocity of the neutral axis of the structural element at the velocity

jump surface (see Figure 4.2(a)); N0 and V0 are the ultimate axial and shear forces

55
 4. Effect of a forepole umbrella on the stability of the tunnel face

of the umbrella; and α is the angle between the failure surface and the cross-section

of the structural element.

(ii) The structural element would fail after being affected by a continuous deformation

zone with a certain thickness (δ). Two plastic hinges are produced in the

reinforcement (Figure 4.2(b)) and its maximum resistance energy can be expressed

as (De Buhan and Salençon, 1993):

𝑊 = 𝑉⃗ · (𝑁 · sin(𝜑 − 𝛼) + 2 · cos(𝜑 − 𝛼)) (4.2)

where M0 is the ultimate bending moment of the structural element.

a)

b)

Figure 4.2. Failure modes considered for the forepole umbrella (modified from
De Buhan and Salençon, 1993)

56
 4. Effect of a forepole umbrella on the stability of the tunnel face

The collapse pressure of the tunnel face considering the effects of the forepole umbrella

can be computed incorporating Equation (4.1) or (4.2) into the work equation (Equation

(3.19)). As explained later, the forepole umbrella modifies the mechanism in such a

way that the velocity of Block 3 becomes vertical (α2=90°). Therefore, the interface

between Blocks 1 and 3 disappears and the resistance energy due to the forepole

umbrella only affects interfaces CD and FG (see Figure 3.3). Moreover, φ-α=η (the

angle of the umbrella with the horizontal, see Figure 4.2) and, considering the small

inclination with which umbrellas are usually installed (typically between 0 and 15°),

the contribution of the axial resistance is neglected in Equations (4.1) and (4.2).

4.3 Effects of forepole umbrella on the stability of the tunnel face

Umbrellas vary greatly in stiffness, costs and installation times (Oke et al., 2014).

Typically, umbrellas are divided into (i) light umbrellas, which consist of steel bars of

relatively small diameter (25-50mm) and a length shorter than the tunnel height; and

(ii) heavy umbrellas, formed by micropiles –i.e., steel pipes (usually of diameter 70-

150mm and wall thickness 6–15mm) filled with grout and with a length of around 20-

25m–. In this analysis, different reinforcements are considered, from a light umbrella

composed of 25 mm diameter bars spaced 1 m, to a heavy one composed of micropiles

with 139.7 mm external diameter (de), 14.2mm thickness and spaced 0.3 m. Because of

the 2D consideration in the whole analysis, the forepole umbrella structural unit is

adopted. Table 4.1 summarizes the geometry, weight, and ultimate loads of the

umbrellas considered. (Diameters and thickness have been selected from a commercial

brochure, trying to cover different amounts of support. Additionally, ultimate (yielding)

loads (N0, M0, V0) have been computed according to the Spanish Recommendations for

Micropile Design (Ministerio de Fomento, 1990) disregarding any reduction

57
 4. Effect of a forepole umbrella on the stability of the tunnel face

coefficients or safety factors and considering a yield strength (fy) of the steel of 355MPa;

Equation (4.3)). In all cases, a 20m long umbrella is supposed. (Note that, although this

is an excessive length for light umbrellas, this allows us to compare results for

analogous situations, in which the reinforcement completely passes through any

possible failure block). The umbrella is assumed to have been installed in the previous

excavation cycle –i.e., the initial point of the umbrella is located at a distance of 2L

from the tunnel face– and with an inclination angle of 5° (see Figure 4.3).

·
𝑁 =𝐴 ·𝑓 · .
;𝑀 = · 𝑓 ;𝑉 = ·√
·𝑓 (4.3)

Figure 4.3. Outline of the analyzed case to study the effect of the forepole
umbrella on the stability of the tunnel face

To carry on this analysis, three specific cases from Chapter 3 with different free span

lengths are chosen, in particular, Case 24 (φ=35°; c=5kPa; L=3m), Case 23 (φ=35°;

c=5kPa; L=2m) and Case 27 (φ=35°; c=15kPa; L=1m); their support pressures without

reinforcement are equal to 27.4, 25.8 and 10.5kPa respectively.

58
 4. Effect of a forepole umbrella on the stability of the tunnel face

For each case, the umbrellas defined in Table 4.1 are introduced into the numerical

model described in Section 3.2, and the corresponding collapse pressures are computed.

(For simplicity, in this analysis only lower bound values of the collapse pressure –

corresponding to upper bound analyses– have been computed). To model the umbrella

in OptumG2, a “plate” structural element is used, which “in plane strain is equivalent to

a standard Euler-Bernoulli beam” (OptumCE, 2019).

Table 4.1. Properties considered to study the effect of the forepole umbrella on
the stability of the tunnel face

Umbrella properties Ultimate loads

Case Center Weight Axial Bending Shear

Diameter Thickness
separation force moment force
[mm] [mm] [kg/m]
[m] [kN/m] [kNm/m] [kN/m]
1 - - - - - - -
2 25 - 1.0 4.0 223 1 90
3 25 - 0.5 8.0 446 3 180
4 76.1 6.3 1.0 10.9 446 11 180
5 76.1 6.3 0.5 21.8 892 22 360
6 82.5 7.1 0.5 26.4 1086 28 438
7 88.9 8.0 0.5 31.8 1312 38 530
8 95.0 8.8 0.5 37.8 1538 46 622
9 101.6 9.5 0.5 43.2 1774 58 718
10 108.0 9.5 0.5 46.2 1898 66 768
11 114.3 10.0 0.5 51.4 2114 78 856
12 121.0 10.0 0.5 54.8 2250 88 910
13 127.0 11.0 0.5 63.2 2588 106 1046
14 133.0 12.5 0.5 74.8 3054 130 1234
15 133.0 12.5 0.4 93.5 3818 163 1543
16 133.0 12.5 0.3 124.7 5090 217 2057
17 139.7 14.2 0.3 146.7 6023 267 2437

59
 4. Effect of a forepole umbrella on the stability of the tunnel face

4.3.1 Collapse pressure

Figure 4.4 shows the collapse pressures obtained, as a function of the ultimate bending

moment of the umbrella. As observed, collapse pressures gradually reduce as the

umbrella becomes stronger, converging towards a minimum value that corresponds to

a failure mechanism that develops below the umbrella (see Figure 4.6(c)). For

comparison, Figure 4.4 also shows the results computed with the proposed Limit

Analysis mechanism, and incorporating the effect of the umbrella according to the two

failure modes described previously. For the first failure mode (Figure 4.2(a); i.e., a

‘sharp’ failure by axial and shear forces at the velocity jump surface; see Mechanism I

in Figure 4.4), a fast decrease of the collapse pressure towards such minimum value

occurs in all cases, even for the lightest umbrella.

For the second failure mode (Figure 4.2(b); i.e., failure by bending moment), results

depend on the thickness (δ) of the continuous deformation zone. Figure 4.4 plots the

results obtained for several hypotheses. The first one (Mechanism II - δBS in Figure 4.4)

comes from the following expression suggested by De Buhan and Salençon (1993):

𝜇= 𝑑/𝛿 (4.4)

where d is the micropile diameter and μ is a non-dimensional parameter. A typical value

for μ is 1/10 (De Buhan and Salençon, 1993), so, since the forepole diameters range

from 25 to 140mm, δ varies between 0.1 and 0.6 m. The second one (Mechanism II –

δNM in Figure 4.4) is obtained after approximating the thickness of the failure zone

ahead of the tunnel face in the numerical model (see Figure 4.5) (values of 1.42, 1.36

and 1.25 m are considered for Cases 24, 23 and 27, respectively; see Table 3.1). A wider

thickness δ produces a smaller rotation at the hinges that develop within the structural

60
 4. Effect of a forepole umbrella on the stability of the tunnel face

element (see Figure 4.2(b)), hence mobilizing less resistance energy, and, consequently,

requiring higher support pressures to maintain the stability of the tunnel face.

When a continuous deformation band is considered, the reduction is more gradual than

when a ‘sharp’ velocity jump surface is assumed, although the obtained results are still

lower than those computed with the numerical model. In both cases, the minimum

values, for heavier umbrellas, correspond to the minimum values obtained with the

numerical model.

Results in Figure 4.4 discussed so far suggest that the failure modes displayed in Figure

4.2 are not completely accurate and may predict unsafe results. Differences may be

caused by the behavior supposed in the analytical solution that, considering the results

of the numerical model, seems imprecise. In the numerical model, the umbrella works

as a fixed-ended beam, with a length larger than the width of the mechanism, and

developing plastic zones at the fixed ends and center of the structural element. But the

proposed analytical model does not exactly reproduce this behavior: the first failure

model (by shear) mobilizes too much energy, hence providing low estimations even for

the lightest umbrella; and the second failure mode (by bending) develops four plastic

hinges (two at each of the two external boundaries of the mechanism crossed by it). (As

previously explained, introducing the forepole umbrella modifies the mechanism in

such a way that the interface between Blocks 1 and 3 disappears, so that the structural

element only crosses two velocity jumps).

61
 4. Effect of a forepole umbrella on the stability of the tunnel face

Figure 4.4. Collapse pressure vs ultimate bending moment of the forepole

umbrella for the three cases analyzed: a) Case 24 (φ=35°; c=5kPa; L=3m); b)
Case 23 (φ=35°; c=5kPa; L=2m); c) Case 27 (φ=35°; c=15kPa; L=1m)

62
 4. Effect of a forepole umbrella on the stability of the tunnel face

Notwithstanding the above, the value of δ can be adjusted to reproduce the results of

the numerical simulation. Figure 4.4 shows the collapse pressures computed with the

proposed mechanism, considering values for δ equal to 3.2 times the thickness of the

failure zone observed in the numerical model (δ=3.2·(δNM)) (i.e., values of 4.54, 4.35

and 4.00m for Cases 24, 23 and 27 respectively; see Mechanism II - 3.2·(δNM) in Figure

4.4). It can be observed that the agreement with the numerical results is excellent.

Therefore, it seems possible to estimate the effect of the forepole umbrella employing

the second failure mode from De Buhan and Salençon (1993) (Figure 4.2(b); failure by

bending moment) and a thickness of the continuous deformation zone equal to 3.2·(δNM).

(Note that this methodology is not a rigorous upper bound solution, so higher values

than the actual critical pressure can be obtained).

But note that the methodology just described requires an estimation of δNM, which is a

result from the numerical model. And the methodology would be useless if it requires

conducting numerical analyses that provide the collapse pressure result that one is

aiming to obtain. Therefore, a simple method to estimate δNM easily is required to make

the methodology useful in practice. To that end, the δNM values from the 36 validation

Tests Cases of Section 3.3.1 (Table 3.1) were employed, but considering three different

tunnel excavation height values (8, 10 and 12 m), to fit an equation that estimates δNM

as a function of the soil strength parameters (φ and c) and of the geometry of the tunnel

(H and L). Based on our fitting trials, the following equation is proposed.

( [ ]) . ·( [ ]) .
𝛿 [m] = ( [ ]) . ·( ) .
− 1.92, 𝐿 ≥ 0.5 (4.5)

63
 4. Effect of a forepole umbrella on the stability of the tunnel face

Figure 4.5. Analysis of Equation (4.5): a) comparison between δ values estimated

from the numerical model (δNM) and those computed with Equation (4.5)
(δEq.(4.5)); b) comparison of collapse pressures computed employing δNM and
δEq.(4.5) for the three cases analyzed

64
 4. Effect of a forepole umbrella on the stability of the tunnel face

Figure 4.5(a) compares the values of δ extracted from the numerical model (δNM) and

the values estimated from Equation (4.5) (𝛿 .( . ) ). The maximum difference is 0.14m;

considering that the sensitivity of the collapse pressure to specific values of δ is

relatively minor, these predictions are useful. To support this statement, Figure 4.5(b)

plots the values of the collapse pressure computed for the three cases analyzed (i.e.,

cases 24, 23 and 27) with the methodology proposed employing the 𝛿 .( . ) values

estimated with Equation (4.5). As shown, the adjustment between the analytical and the

numerical results continues to be excellent. Figure 4.5(a) discriminates between

different diameters and friction angles. As it can be observed, the value of δ reduces

with D and increases with φ.

4.3.2 Failure geometry

Comparing the displacements of the numerical model and the failure geometries of the

proposed approach provides some interesting observations. As an example, Figure 4.6

shows the displacement fields in the numerical model for Case 24 for three situations:

(i) without reinforcement (Figure 4.6(a)); (ii) reinforced with a light umbrella (Figure

4.6(b)); and (iii) reinforced with a heavy umbrella (Figure 4.6(c)). (The results of the

proposed Limit Analysis mechanism, employing the failure mode by bending (Figure

4.2(b)) and with a width given by 3.2·(δNM), are superimposed for comparison.) In the

case without reinforcement, the three blocks described in Chapter 3 can be clearly

identified. When the ‘light’ umbrella is introduced (Case 2 in Table 4.1), displacements

above the tunnel become vertical and only one block can be distinguished in this zone.

With the ‘heavy’ umbrella (Case 17 in Table 4.1), the displacements above the

reinforcement are practically negligible, and all the movement is located ahead of the

tunnel face, developing a typical rotational failure.

65
 4. Effect of a forepole umbrella on the stability of the tunnel face

a)

b)

c)

Figure 4.6. Comparison of failure geometries obtained with the proposed

mechanism and with the numerical model in Case 24 (φ=35°; c=5kPa; L=3m) for
different amounts of supports: a) without umbrella; b) light umbrella; c) heavy
umbrella

Similar results are obtained with the proposed Limit Analysis mechanism. When the

‘light’ umbrella is introduced into the mechanism, the velocity of Block 3 becomes

vertical –i.e., α≈90°– so that the velocity of Block 3 is equal to the velocity of Block 1

66
 4. Effect of a forepole umbrella on the stability of the tunnel face

and the interface between them disappears. (Because of this, a simpler mechanism

depending on only three variables (βE, RE, α1) can be used, as employed to obtain the

results of Figures 4.4 and 4.5.) For the case with the ‘heavy’ umbrella, the velocity of

Block 2 is much larger than the velocities of Blocks 1 and 3. Consequently, the overall

energy balance is mainly between Block 2 and the support pressure, denoting that the

ground falls below the umbrella. (The minimum collapse pressures shown in Figure 4.4

could also be obtained with a mechanism only composed of one rotational block ahead

of the tunnel face).

4.4 Conclusions

The stability of the tunnel face with a free span is studied considering the reinforcement

provided by a forepole umbrella. The contribution of the umbrella is introduced into

the mechanism considering two failure modes proposed by De Buhan and Salençon

(1993).

The results show that a direct implementation of such two failure modes tends to

overpredict the effect of the umbrella on the support pressure. In addition, the failure

mode considering the bending moment, as compared to the one considering the shear

force, gives better results even though more efforts are needed to obtain the thickness

of the continuous deformation zone developed across the umbrella. Consequently, a

new methodology has been proposed to compute the value of the collapse pressure more

accurately, and without the need to conduct numerical analyses to estimate the

deformation zone thickness, through a fitting function and an additional multiplier

adjustment. Moreover, the mechanism is modified to reduce the number of parameters

to be optimized, after finding that the failure geometry changes depending on the

67
 4. Effect of a forepole umbrella on the stability of the tunnel face

amount of support (i.e., three blocks for cases without reinforcement, two blocks for

light forepole umbrellas and just one rotational block for heavy umbrellas).

Results confirm that the proposed mechanism can capture the stability improvement

(i.e., the reduction of the collapse pressure) produced by the umbrella, as well as the

variation of the failure mechanism. Therefore, the proposed limit analysis mechanism

can reproduce the critical failure geometry associated with different amounts of support;

in particular, it can predict a local failure below the support for heavy umbrellas.

Hence, the free span mechanism has been successfully extended to be able to assess the

stability of a tunnel face with an additional reinforcement provided by a forepole

umbrella.

68
 5. Tunnel face stability considering advance drainage under steady-state conditions

5. Tunnel face stability considering

advance drainage under steady-
state conditions

5.1 Introduction

For tunnels excavated under the water table, the stability of the tunnel face is reduced

by the destabilizing effect of the water, which depends on the hydrodynamic conditions

(i.e., steady-state and transient conditions). Consequently, assessing the effect of the

water flow is crucial, so that the advance measures (e.g., support, or drainage) can be

designed and implemented.

Two main design approaches can be employed for tunnels under the water table:

waterproofing reinforcement (e.g., grouting and freezing) and drainage (e.g., drainage

borehole, pilot tunnel, twin tunnel and drainage curtain). The best choice for a particular

tunnel depends on aspects such as the water table recharge flow rate, hydraulic gradient,

geomechanical characteristics, etc., (Lunardi, 2008). For example, an impermeable

design should probably be selected when environmental restrictions apply, even though

that implies that the tunnel is excavated under hydrostatic pressures (Line a in Figure

69
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.1(a)); or the hydraulic conditions can be modified as a result of construction (Line b),

or even in some cases they should be changed even further (e.g., with drainage

boreholes, see Line c) if the hydraulic head is too high.

a)

b)

Figure 5.1. a) Water level distributions in different hydraulic conditions; b) a

drainage borehole (Lunardi, 2008)

In this Chapter, the stability of the tunnel face under the water table is investigated,

considering steady-state conditions and under the assumption of a fast aquifer recharge.

(Next Chapter 6 analyzes the transient condition). Moreover, drainage boreholes (see

70
 5. Tunnel face stability considering advance drainage under steady-state conditions

Figure 5.1(b)) are considered to study the effect of advance drainage, as they are one of

the most typical measures to reduce the hydraulic head in tunnels under the water table.

Section 5.2 extends the Limit Analysis mechanism proposed by Mollon et al. (2011a)

to analyze the effect of water pressures and (as an original contribution of this thesis)

of advance drainage measures on the stability of the tunnel face. The pore water

pressure distributions correspond to steady-state conditions, and they are computed,

considering the presence of drainage boreholes placed at the tunnel face, by numerical

simulations in Section 5.3. The proposed mechanism is validated using a FLAC3D

numerical model (Section 5.4). Finally, in Section 5.5, the results of the proposed Limit

Analysis methodology are compared with the results obtained by a Limit Equilibrium

solution from the literature.

5.2 3D Limit Analysis mechanism considering pore water

pressure and the effect of advance drainage

5.2.1 Model description

Figure 5.2(a) shows a longitudinal section of the problem analyzed: a circular tunnel

excavated under the water table and with advance drainage boreholes. The tunnel has a

diameter of D=10m and an overburden, C, of between 20 and 50m. Hw represents the

water elevation from the tunnel crown (it also ranges between 20 and 50m) and h0 is

the initial hydraulic head at the tunnel axis. As shown in Figure 5.2(b), we consider two

horizontal drainage boreholes of diameter db=0.2m and length Lb=30m distributed

symmetrically with an angle of 45° from the vertical and with a distance to the center

of the tunnel face r=3.8m, in two different locations: (I) “upper” drainage; and (II)

“lower” drainage.

71
 5. Tunnel face stability considering advance drainage under steady-state conditions

a)

b)

Figure 5.2. Outline of the analyzed problem: a) in the vertical plane of symmetry
of the tunnel; b) in the tunnel face cross-section. (Boreholes layout: (I) “upper”
drainage; (II) “lower” drainage.)

5.2.2 Limit Analysis mechanism

The Limit Analysis mechanism proposed by Mollon et al. (2011a) is employed,

considering the effect of the pore pressure distribution as proposed by Pan and Dias

(2016). The collapse mechanism is formed by a unique non-deformable block that

rotates around a horizontal line perpendicular to the vertical plane of symmetry of the

tunnel (see Figure 5.3(b)). The mechanism is bounded, in this plane of symmetry, by

two log-spirals emerging from the crown and invert of the tunnel face that have the

same center (Point O in Figure 5.3). To generate the 3D failure surface that originates

from the tunnel face and fulfills the associated flow rule, as required by the Limit

Analysis theorems, a “point by point” spatial discretization technique is used, as

72
 5. Tunnel face stability considering advance drainage under steady-state conditions

explained in Mollon et al. (2011a). This technique first discretizes the tunnel perimeter

in a set of Points 𝐴 (see Figure 5.3(a)) and defines a set of radial planes 𝛱 ,

perpendicular to the vertical plane of symmetry, that pass through the center of rotation

of the mechanism (see Figure 5.3(b)). Then, if two points 𝑃 , and 𝑃 , in Plane 𝛱 are

known, a third point 𝑃 , in Plane 𝛱 can be obtained, so that the triangular facet

𝐹 , (defined by 𝑃 , , 𝑃 , and 𝑃 , ) fulfills the associated flow rule hypothesis of

Limit Analysis (see Figure 5.4). Similarly, 𝐹′ , can be obtained from points 𝑃 , ,

𝑃, and 𝑃 , ). When the surface of the collapse mechanism is completely defined

by the set of facets 𝐹 , and 𝐹′ , , an upper bound of the collapse pressure can be

obtained by equating (i) the total rate of work applied to the system by the external

forces and (ii) the rate of energy dissipated in the system due to shear forces developed

by Mohr-Coulomb soils.

Figure 5.3. Generation of the Limit Analysis mechanism: a) cross-section at the

tunnel face boundary; b) longitudinal section along the vertical plane of
symmetry of the tunnel (simplified from Mollon et al. (2011a), Fig.3)

73
 5. Tunnel face stability considering advance drainage under steady-state conditions

Figure 5.4. Point-by-point generation of the external surface of the mechanism

(simplified from Mollon et al. (2011a), Fig.4)

In the employed mechanism, the external forces comprise the soil weight, 𝑊 ; the

support force applied on the tunnel face, 𝑊 ; and, if the failure mechanism outcrops at

the surface, the ground surcharge, 𝑊 ; additionally, the resultant force due to the

seepage and buoyancy forces, 𝑊 , is considered. The rate of work of the soil weight

(𝑊 ), of the support force applied on the tunnel face (𝑊 ), and of the internal energy

dissipated along the failure surface (𝑊 ) are formulated concisely in Equations (5.1),

(5.2) and (5.3) respectively (more details can be found in Mollon et al., 2011a, and in

Pan and Dias, 2016).

𝑊 = ∭ 𝛾⃗ · 𝑣⃗ · 𝑑𝑉 = 𝜔 · 𝛾 ·∑ ∑ 𝑅 , · 𝑉 , · sin 𝛽 , + 𝑅 , ·𝑉 , · sin 𝛽 ,

(5.1)

𝑊 = ∬ 𝜎⃗ · 𝑣⃗ · 𝑑𝑆 = −𝜔 · 𝜎 · ∑ (𝑆 , · 𝑅 , · cos 𝛽 , ) (5.2)

𝑊 = ∬ 𝑐 · 𝑣 · cos 𝜑 · 𝑑𝑆 = 𝜔 · 𝑐 · cos 𝜑 · ∑ ∑ (𝑅 , · 𝑆 , + 𝑅′ , · 𝑆′ , ) (5.3)

74
 5. Tunnel face stability considering advance drainage under steady-state conditions

where 𝑅 , , 𝑉 , , 𝛽 , , 𝑆 , , 𝑅′ , , 𝑉′ , , 𝛽′ , , 𝑆′ , , 𝑅 , , 𝛽 , , 𝑆 , are geometrical terms

defined in Mollon et al. (2011a). In addition, note that the saturated unit weight γsat is

used in this thesis as the collapse mechanisms considered always develop under the

water table. Furthermore, since they never outcrop at the surface, the rate of work due

to the ground surcharge is always null.

According to Viratjandr and Michalowski (2006), the rate of work due to the seepage

and buoyancy forces (𝑊 ) can be computed using the pore water pressures; following

Pan and Dias (2016), we have:

𝑊 = − ∭ 𝑢 · 𝜀 · 𝑑𝑉 − ∬ 𝑢 · 𝑛 · 𝑣 · 𝑑𝑆 = 𝜔 · ∑ 𝑢 , · 𝑆 , · 𝑅 , · cos 𝛽 , +

𝜔 · sin 𝜑 · ∑ ∑ 𝑢, ·𝑅, ·𝑆, +𝑢 , ·𝑅 ,

·𝑆 , + 𝜔 · ∑ (𝑢 · 𝑆 · 𝑅 · sin 𝛽 )

(5.4)

where the volumetric strain 𝜀 is equal to zero because of the rigid block assumption.

Additionally, the pore pressure at the tunnel face (𝑢 , ) is assumed to be atmospheric,

so that the first term in the right side of Equation (5.4) is equal to zero; and 𝑢 ---which

represents the pore water pressure at outcropped elements--- is also zero because the

mechanisms considered herein never outcrop at the surface. Then, the rate of work due

to pore water pressures becomes (Pan and Dias, 2016):

𝑊 = − ∬ 𝑢 · 𝑛 · 𝑣 · 𝑑𝑆 = 𝜔 · sin 𝜑 · ∑ ∑ 𝑢 , · 𝑅 , · 𝑆 , + 𝑢′ , · 𝑅 ,
·𝑆 ,

(5.5)

where 𝑢 , and 𝑢′ , represent the pore pressures in the triangular facets that form the

external surface of the mechanism. In this work, they are obtained interpolating in the

pore water pressure distributions computed with the numerical model described in the

next section.

75
 5. Tunnel face stability considering advance drainage under steady-state conditions

After applying the work equation, the collapse pressure can be expressed as (Pan and

Dias, 2016):
·∑ ∑ , · , · , , · , · , · ·∑ ∑ , · , , · ,
𝜎 = − +
∑ ( , · , · , ) ∑ , · , · ,
·∑ ∑ , · , · , , · · ,
,
(5.6)
∑ , · , · ,

The optimal value of 𝜎 can be obtained by maximizing Equation (5.6) with respect to

the two coordinates that define the position of the center of rotation of the mechanism

in the vertical plane of symmetry (𝛽 and 𝑅 ; see Figure 5.3(b)). The obtained value of

𝜎 represents a lower bound of the support pressure 𝜎 required to stabilize the face,

considering the influence of water seepage.

5.3 Numerical computation of pore water pressure distribution

under steady-state conditions

To obtain the pore water pressure distribution required by the Limit Analysis solution,

a numerical model built in FLAC3D is employed. In the numerical model, shown in

Figure 5.5, the tunnel has a diameter of D=10m and the overburden is C=20m. Taking

advantage of the symmetry of the problem, and to decrease the computing times, only

half of the model is considered. The dimensions of the numerical model are

55m×70m×50m in x, y and z directions, respectively. The model has 157 500 zones and

the mesh ahead of the tunnel face is refined for a more precise interpolation of pore

water pressures in the vicinity of the collapse mechanism.

76
 5. Tunnel face stability considering advance drainage under steady-state conditions

Figure 5.5. Numerical model with advance drainage boreholes employed to

calculate the pore water pressure distribution and the collapse pressure

The water table is assumed to maintain a constant elevation (Hw in Figure 5.2(a)), whose

value depends on the case considered. The lower and lateral external boundaries of the

numerical model, including the longitudinal plane of symmetry of the tunnel, are

impermeable. The tunnel perimeter is also assumed to be impermeable, so that the

tunnel face becomes the only surface into which the groundwater can flow. When the

advance drainage boreholes are considered, the “null” model is applied to zones

representing the location of the boreholes, and an atmospheric pressure (equal to zero)

is imposed to them.

As an example of the results of the numerical model, Figure 5.6 shows the distributions

of pore water pressures in the vertical plane of symmetry of the tunnel, when the water

table is 3 diameters above the tunnel crown (Hw=3D), and for the cases without drainage

and with “upper” and “lower” drainage. As expected, Figure 5.6 shows (i) that advance

drainage reduces pore water pressures ahead of the tunnel face and (ii) that the

distribution of the reduced pressures varies with the location of the boreholes.

77
 5. Tunnel face stability considering advance drainage under steady-state conditions

a)

b)

c)

Figure 5.6. Contours of the pore water pressure distribution in the numerical
model for different drainage configurations: a) without drainage; b) “upper”
drainage boreholes; c) “lower” drainage boreholes. (D=10m; C=2D; Hw=3D).
(Values shown indicate the hydraulic head in kPa)

78
 5. Tunnel face stability considering advance drainage under steady-state conditions

To illustrate the effect of drainage better, Figure 5.7 presents the hydraulic heads along

the tunnel axis for several Test Cases listed in Table 5.1. (These Test Cases are later

employed for the numerical validation of the mechanism). As it can be observed,

advance drainage (Cases 5, 8, 9 and 12) significantly reduces the pore water pressures

ahead of the tunnel face, so that such reduction extends up to the total length of the

drainage boreholes (30m) and even beyond. In addition, it shows that the “lower”

drainage layout (Cases 9 and 12) is more effective to reduce the hydraulic head along

the tunnel axis than the “upper” layout (Cases 5 and 8). And, although it is less

appreciable, this difference of efficiency is reduced when the water level is raised (See

Cases 5 and 9 for Hw/D=2 and Cases 8 and 12 for Hw/D=5). (Similar results have been

obtained by Zingg (2016), although she considered higher Hw/D ratios and, as explained

later, the effectiveness of drainage to stabilize the tunnel face depends on the water

level).

Figure 5.7. Distribution of hydraulic head along the tunnel axis for several Test
Cases (see Table 5.1). (h: hydraulic head; h0: initial hydraulic head; y: distance
from the tunnel face)

79
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.4 Numerical validation

5.4.1 Test Cases

To validate the proposed methodology, 15 Test Cases are employed (see Table 5.1), to

compare the results of the Limit Analysis solution–in particular, the critical pressure

and the collapse geometry–with the results of the numerical model in FLAC3D.

A tunnel with a diameter of 10m in a purely frictional material (φ=30°) with a saturated

unit weight of 15.6kN/m3 is assumed, considering different elevations of the water table

(Hw/D=2 to 5) and different drainage configurations (without drainage, and with “upper”

and “lower” drainage). In general, an overburden of two diameters is considered,

although three additional cases with other overburdens are employed to analyze its

influence.

80
 5. Tunnel face stability considering advance drainage under steady-state conditions

Table 5.1. Test Cases considered for numerical validation of the Limit Analysis mechanism. (Diameter:D=10m; strength parameters:
c=0kPa and φ=35°; saturated unit weight: γsat=15.6kN/m3).

Reduction with
Collapse pressure respect to “without Difference LA vs NS
drainage” case
Position of
Case 𝐻 /𝐷 𝐶 ⁄𝐷
drainage ( ) ( )
𝜎 ( ) 𝜎 ( ) 𝑅( ) 𝑅( ) 𝜎 ( ) −𝜎 ( )
( )
[kPa] [kPa] [kPa] [kPa] [kPa]
[%]
1 2 2 Without 87.4 83.1 - - 4.3 5.1
2 3 2 Without 118.4 112.6 - - 5.8 5.2
3 4 2 Without 151.2 142.7 - - 8.5 5.8
4 5 2 Without 184.1 173.0 - - 11.1 6.4
5 2 2 Upper 64.2 61.3 23.2 21.8 2.9 4.8
6 3 2 Upper 83.6 80.1 34.8 32.5 3.5 4.3
7 4 2 Upper 103.2 99.0 48 43.7 4.2 4.3
8 5 2 Upper 123.1 119.5 61 53.5 3.6 3.0
9 2 2 Lower 59.7 57.5 27.7 25.6 2.2 3.8
10 3 2 Lower 80.8 77.6 37.6 35 3.2 4.0
11 4 2 Lower 102.6 98.4 48.6 44.3 4.2 4.2
12 5 2 Lower 124.6 119.4 59.5 53.6 5.2 4.4
13 3 3 Upper 82.2 78.6 - - 3.8 4.5
14 4 4 Upper 103.3 99.7 - - 3.6 3.6
15 5 5 Upper 122.8 120.6 - - 2.2 1.8

Note. LA: Limit Analysis mechanism; NS: Numerical simulation with FLAC3D.

81
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.4.2 Results of Limit Analysis

To validate the Limit Analysis mechanism that considers the influence of pore water

pressure, and in particular, to compare its results with the results of a numerical model,

it would be enough to use the same pore water pressure distribution–using e.g., the pore

water pressure coefficient ru defined by Bishop and Morgenstern (1960)–and to

compare the results of both approaches. However, to be able to study accurately the

effect of advance drainage on the tunnel face stability, the FLAC3D numerical model

described in Section 5.3 is used to enable us to work with more realistic pore pressure

distributions.

To that end, pore water pressures are extracted from the numerical model using a FISH

routine (the built-in programming language in FLAC3D), and they are then transferred

to the code of the Limit Analysis solution (implemented in MATLAB, The MathWorks,

2012). And, since the positions of nodes in the numerical model generally do not

coincide with the positions of the centers of the triangular facets that define the collapse

mechanism (i.e., where they are required to compute the support pressure in Equation

(5.6)), pore water pressures at the centers of facets are interpolated from the

distributions computed with FLAC3D. Then, the critical collapse pressure is computed

as explained in Section 5.2.2. Table 5.1 lists the critical pressures obtained with the

Limit Analysis mechanism, and Figure 5.8 shows three examples of the collapse

geometries obtained with the analytical solution for the different drainage

configurations (without drainage, and with “upper” and “lower” drainage).

82
 5. Tunnel face stability considering advance drainage under steady-state conditions

a)

b)

c)
Figure 5.8. Examples of the failure geometries obtained with the Limit Analysis
mechanism: a) Test Case 2 (without drainage); b) Test Case 6 (“upper”
drainage); c) Test Case 10 (“lower” drainage). (D=10m; C=2D; Hw=3D)

83
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.4.3 Results of numerical analysis

The numerical model to compute the collapse pressure of the tunnel face considering

the influence of water is identical to that presented in Section 5.3. The constitutive

model for the ground is elastic-perfectly plastic, using the Mohr-Coulomb failure

criterion as implemented in FLAC3D, and with an associated flow rule so that it fulfills

the requirements of Limit Analysis. The model is configured to “mechanical calculation”

only, so that pore pressures do not change during the simulation. (Note, however, that

they are employed to compute effective stresses and to detect failure; Itasca Consulting

Group, 2009). As before, the pore pressure distribution is imported from a (flow-only)

numerical simulation. (In this way, the distribution is ensured to be the same for both

calculations so that it does not affect the computed results).

The improved bisection method, proposed by Mollon et al. (2009a), has been adopted

to compute the collapse pressure in the numerical model as described in Section 2.3.2.

According to Mollon et al. (2009a), a value of Rref=1EXP-07 is employed as the

convergence criterion for the unbalanced mechanical-force ratio and the collapse

pressure is computed with a precision ε=0.1kPa. The obtained values are listed in Table

5.1. The collapse geometry in the numerical model can be visualized using the

distribution of shear strain rates. Figures 5.9, 5.10 and 5.11 show the failure

mechanisms–as observed within the vertical plane of symmetry of the tunnel–computed

with the numerical model for different drainage configurations (Figure 5.9), water

levels (Figure 5.10) and overburdens (Figure 5.11); collapse geometries from the Limit

Analysis analytical solution have been added in these figures for comparison. (The

obtained geometries and the influence of the different variables will be discussed in

Section 5.5.)

84
 5. Tunnel face stability considering advance drainage under steady-state conditions

a)

b)

c)

Figure 5.9. Comparison of failure mechanisms computed with the Limit Analysis
solution and with the numerical simulation for different drainage configurations:
a) without drainage (Test Case 2); b) “upper” drainage (Test Case 6); c) “lower”
drainage (Test Case 10). (D=10m; C=2D; Hw=3D)

85
 5. Tunnel face stability considering advance drainage under steady-state conditions

a) b)

c) d)

Figure 5.10. Comparison of failure mechanisms computed with the Limit

Analysis solution and with the numerical simulation for different water levels: a)
Hw=2D (Test Case 9); b) Hw=3D (Test Case 10); c) Hw=4D (Test Case 11); d)
Hw=5D (Test Case 12). (D=10m; C=2D; “lower” drainage)

a) b)

Figure 5.11. Comparison of failure mechanisms computed with the Limit

Analysis solution and with the numerical simulation for different overburdens:
a) C=2D (Test Case 6); b) C=3D (Test Case 13). (D=10m; Hw=3D; “upper”
drainage)

86
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.4.4 Comparison

Results in Table 5.1 show that collapse pressures computed with the analytical solution

and the numerical model are very similar, with absolute differences that typically are

in the range of 3-8kPa, or with relative differences that are typically less than 5%.

Consequently, the proposed Limit Analysis mechanism can be employed to compute

face collapse pressures considering the influence of advance drainage. Note also that

the Limit Analysis results are higher than those computed with the numerical model.

Since this is an upper bound solution in the framework of Limit Analysis, and given

that the face pressure acts against the movement of the mechanism, the obtained values

should be lower than the ones given by the numerical simulation. Several reasons can

justify this incoherence, including the mesh size in the numerical model (e.g., Mollon

et al., 2011b), the locations of the model boundaries, or the precision of the method

employed to calculate the collapse pressure. Table 5.1 also shows that the differences

between the numerical model and the analytical solution tend to increase as the water

table rises; on the other hand, Test Cases with drainage show lower differences than

those without drainage. That is, differences between the analytical solution and the

numerical model increase as the pore water pressures ahead of the tunnel face increase.

Therefore, caution should be exercised if the Limit Analysis mechanism is employed

to study the face stability in tunnels under higher water pressures (i.e., with Hw>5D, for

a D=10m tunnel).

Finally, the shape of the failure mechanisms obtained with the Limit Analysis approach

agrees satisfactorily with the numerical simulation results (Figures 5.9, 5.10 and 5.11).

The agreement is excellent in the lower part of the mechanism and approximated in the

upper part. (However, note that, as indicated by Mollon et al. (2011b), the lower part of

87
 5. Tunnel face stability considering advance drainage under steady-state conditions

the mechanism has the main contribution to the computed face collapse pressure for a

constant applied pressure.) In the upper part of the mechanism, the numerical model

shows more rounded failure geometries, whereas the geometry of the analytical solution

is sharper. This is more relevant in the case without drainage (Figure 5.9(a)), in contrast

with other cases for which the agreement is better. In conclusion, the analytical solution

gives a good approximation of the collapse geometry in all cases and it can be employed

to reasonably predict the shape of the failure surface.

5.5 Discussions

5.5.1 Effect of advance drainage on the stability of the tunnel face

Results in Table 5.1 show that raising the water table increases the collapse pressure in

all drainage configurations (Cases 1 to 4 without drainage; Cases 5 to 8 and Cases 9 to

12 with “upper” and “lower” drainage respectively). As noted by Zingg (2016), even a

very simple drainage with only two boreholes significantly reduces the critical pressure

(around 25-30% of its value) due to the reduction of the pore water pressure ahead of

the tunnel face. Results also show that both drainage configurations (“upper” and

“lower”) produce similar reductions of the collapse pressure. Finally, the comparison

of Test Cases 13 to 15 and Test Cases 6 to 8 suggests that the overburden does not

significantly influence the value of the collapse pressure.

Figures 5.8 and 5.9 illustrate that the shape of the critical failure mechanism is affected

by the drainage configuration employed: in the case without drainage (Figures 5.8(a)

and 5.9(a)), the mechanism does not extend upwards and it evolves towards the region

directly ahead of the tunnel face. With “upper” drainage (Figures 5.8(b) and 5.9(b)), the

mechanism develops vertically towards the region above the tunnel crown, hence

developing a similar shape to that obtained in studies considering dry ground (e.g.,

88
 5. Tunnel face stability considering advance drainage under steady-state conditions

Senent et al., 2013). Also note that, without drainage, the mechanism extends further

ahead of the tunnel face than when “upper” drainage is employed (approximately 6m

vs 4m in the Limit Analysis mechanism). This is probably because the mechanism tries

to mobilize as much energy (from the external loads) as possible, so that it tends to

develop towards the region with higher pore water pressures when there is no drainage;

with drainage, however, the effect of water is less important, and the mechanism tries

to optimize the relationship between its volume and its surface. Finally, with “lower”

drainage (Figures 5.8(c) and 5.9(c)), the reduction of pore pressures ahead of the upper

part of the face is less significant (see Figure 5.6(c)), so that the mechanism tries again

to develop towards the region with higher pore pressures.

Figure 5.10 suggests that the variation of the failure geometry discussed in the previous

paragraph is more relevant as the water level increases, so that when the pore pressures

ahead of the tunnel increase, the mechanism tends to become more horizontal, trying to

develop itself deeper into the region ahead of the tunnel face. To better show these

results, Figure 5.12 plots a comparison between the three drainage configurations

considered–without drainage, “upper” drainage and “lower” drainage–when Hw/D=3

(Figure 5.12(a)) and Hw/D=4 (Figure 5.12(b)), clearly illustrating that (i) the shape of

the mechanism changes depending on the drainage configuration, and (ii) the

“horizontalization effect” is more relevant for cases without drainage and with “lower”

drainage, whereas it is almost irrelevant when the drainage is in the upper part of the

tunnel face.

89
 5. Tunnel face stability considering advance drainage under steady-state conditions

a) b)

Figure 5.12. Variation of the collapse geometry depending on the drainage

configuration for different water levels: a) Hw=3D; b) Hw=4D. (Cases 2 and 3
without drainage; Cases 6 and 7 “upper” drainage; Cases 10 and 11 “lower”
drainage). (D=10m; C=2D)

Finally, Figure 5.11 shows two cases with the same water table and with “upper”

drainage, but with different overburdens. As it can be seen, the collapse geometry is

almost equal in both cases, hence suggesting that overburden does not affect much the

geometry of the failure surface (and neither the collapse pressure, as shown in Table

5.1).

5.5.2 Comparison with other solutions from the literature

In this section, the proposed Limit Analysis mechanism is compared with a Limit

Equilibrium solution from the literature (Anagnostou and Kovári, 1994, 1996), which

was previously employed to study the effect of drainage on tunnel face stability (Zingg,

2016). Although Zingg studies the effect of different drainage designs (such as drainage

boreholes in different numbers and layouts, or pilot tunnels), our comparison is limited

to the two drainage layouts analyzed in this thesis. Consequently, a 10m diameter tunnel

excavated in a purely frictional material (with φ=35°) and with different water table

elevations (Hw/D from 2 to 5) is considered, with two drainage boreholes with L=30m

90
 5. Tunnel face stability considering advance drainage under steady-state conditions

length located at the upper and lower part of the tunnel face (they correspond to the

“upper” and “lower” drainage configurations presented in Figure 5.2(b)). (A third case,

i.e., tunnel without drainage, is also considered for reference and comparison). In

addition, note that, due to the limitation of the Limit Equilibrium mechanism to

compute the collapse pressure when the submerged weight of the soil is lower than a

threshold–in the case without advance drainage, for 𝛾 ≤7.9kN/m3–a saturated weight

of 20kN/m3 is employed for this comparison. Note also that Zingg (2016) studies

subaqueous tunnels with Hw>5D, whereas this work analyzes tunnels with lower Hw.

Consequently, the Limit Equilibrium results presented herein are not directly taken

from Zingg (2016); they are computed using an own MATLAB implementation of the

formulation presented in that work.

Table 5.2 lists the Test Cases employed for comparison, and the collapse pressures

obtained for each case with both analytical solutions (Limit Analysis and Limit

Equilibrium); results computed with a FLAC3D numerical model are reported as well.

These values are plotted in Figure 5.13. As previously, collapse pressures computed

with the Limit Analysis mechanism are similar to results from the numerical model.

Results from the Limit Equilibrium mechanism are, in all cases, higher than those

computed with Limit Analysis and with FLAC3D (with relative differences generally of

around 15-29%, and reaching 44% in the case without drainage), and with absolute

differences that increase with Hw/D; this, however, suggests that the Limit Equilibrium

approach provides a safe estimation. The reason why the results of the Limit Analysis

mechanism are closer to the results of the numerical model could be due to the fact that

the same hypotheses (mainly, the associated flow rule) are considered in both

approaches.

91
 5. Tunnel face stability considering advance drainage under steady-state conditions

Table 5.2. Test Cases considered to compare the Limit Analysis mechanism with the Limit Equilibrium methodology presented by
Zingg (2016). (Diameter: D=10m; strength parameters: c=0kPa and φ=35°; saturated unit weight: γsat=20kN m3).
Numerical Simulation
Limit Analysis Limit Equilibrium
Position of with FLAC3D
Case 𝐻 /𝐷 𝐶 ⁄𝐷
drainage 𝜎( ) 𝜎( ) 𝜎( )
𝑤 [°] 𝑤 [°] 𝑤 [°]
[kPa] [kPa] [kPa]
1 2 2 Without 90.9 50 135.3 64 83.5 46

2 3 2 Without 121.3 51 200.0 68 112.8 46

3 4 2 Without 153.1 50 264.4 70 142.6 45

4 5 2 Without 185.7 50 330.3 72 173.1 47

5 2 2 Upper 67.3 51 80.4 42 61.6 47

6 3 2 Upper 87.3 49 111.4 44 80.2 48

7 4 2 Upper 103.3 51 142.9 44 99.7 52

8 5 2 Upper 126.7 51 174.5 44 119.5 52

9 2 2 Lower 63.8 44 75.6 38 57.8 43

10 3 2 Lower 84.6 44 109.3 40 78.1 40

11 4 2 Lower 106.2 43 143.6 40 98.8 41

12 5 2 Lower 127.9 44 178.1 42 119.4 42

92
 5. Tunnel face stability considering advance drainage under steady-state conditions

Figure 5.13. Collapse pressures computed with the Limit Analysis mechanism,
with the Limit Equilibrium methodology presented by Zingg (2016), and with a
numerical simulation in FLAC3D, for different drainage configurations. (See
Table 5.2)

In any case, all the methodologies demonstrate that: (i) collapse pressures increase

dramatically with the rise of the water level (i.e., with Hw/D); (ii) advance drainage

greatly reduces the collapse pressure, particularly for higher water tables; and (iii) the

location of the drainage boreholes does not affect the value of the collapse pressure

significantly, although it has some influence on the shape of the critical failure geometry.

In addition, results show that “lower” drainage is more effective when the water table

is low (between 3 and 4 diameters depending on the method employed, for the values

considered in this study); whereas, “upper” drainage is more effective when the water

table is higher than 4D. These results are coherent with previous research; e.g., Hong

et al. (2007) recommended “upper” drainage in the case of a tunnel with Hw/D≈13.8.

93
 5. Tunnel face stability considering advance drainage under steady-state conditions

Finally, the failure geometry of the Limit Equilibrium solution–formed by a wedge and

an upper prism–is compared to the rotational failure mechanism obtained with Limit

Analysis. Figure 5.14 shows two examples, with C/D=2 and Hw/D=3 and with “upper”

and “lower” drainage (Figures 5.14(a) and 5.14(b) respectively). Note that both

analytical solutions predict very similar initial slopes (or critical angle, ω) of the

mechanism in its lower part, and that differences in the upper part are substantial. Table

5.2 summarizes the critical angles computed with the Limit Equilibrium solution for all

cases considered, and the angles extracted from the Limit Analysis mechanism and

from the numerical model. All approaches produce similar values, except for situations

without drainage, in which Limit Analysis and the numerical model produce more

similar results. For the cases considered, the critical angle for cases with “upper”

drainage is higher than for “lower” drainage. This result disagrees with Zingg (2016),

who obtained that “lower” drainage produced higher angles; the discrepancy is

probably due to the different Hw/D ratios considered (Hw/D≤5 are employed herein,

whereas Zingg (2016) employed Hw/D=13).

a) b)

Figure 5.14. Comparison of the failure geometries obtained with the Limit
Analysis mechanism and with the Limit Equilibrium methodology presented by
Zingg (2016): a) “upper” drainage (Test Case 6 in Table 5.2); b) “lower”
drainage (Test Case 10 in Table 5.2). (D=10m; C=2D; Hw=3D)
94
 5. Tunnel face stability considering advance drainage under steady-state conditions

5.6 Conclusions

The effect of advance drainage boreholes is incorporated into the Limit Analysis

rotational collapse mechanism developed by Mollon et al. (2011a). The aim is to study

the face stability of a circular tunnel, with D=10m, constructed under the water table,

with particular attention on their failure geometries and on their collapse pressures, for

different drainage configurations (without drainage, and with drainage in the upper and

in the lower part of the tunnel face). The pore water pressure on the external surface of

the collapse mechanism is interpolated from pore water pressure distributions computed

with FLAC3D. Critical pressures and failure geometries calculated with the analytical

solution reasonably reproduce the results from the numerical simulations, especially

when the height of the water table above the tunnel crown is lower than 5 diameters.

Results show an expected increase of the critical pressure when the water level rises,

and highlight the importance of drainage measures on the stability of the face. The

location of the drainage boreholes is not so relevant, for the studied cases, although a

“lower” drainage is more effective when the water table is lower (Hw/D<3), whereas

an “upper” drainage is recommended when the water table is higher (Hw/D>4). Results

also show that the failure geometry is clearly influenced by the drainage configuration,

so that for cases without drainage and with “lower” drainage the mechanism develops

more horizontally into the ground ahead of the tunnel face, whereas with “upper”

drainage the mechanism develops more vertically, towards above the tunnel crown. The

shape of the critical collapse mechanism, particularly the tendency to develop more

horizontally (or not), is also heavily affected by the position of the water table.

Comparing the proposed methodology with the classical Limit Equilibrium mechanism

of Anagnostou and Kovári (1994, 1996) (Zingg, 2016), formed by a wedge and an upper

95
 5. Tunnel face stability considering advance drainage under steady-state conditions

prism, the Limit Analysis approach gives lower values of the collapse pressure (with

differences up to 30-45% mainly depending on the water level and on the drainage

configuration), that generally agree better with the numerical simulation results.

Furthermore, similar values of the initial slope of the mechanism in its lower part are

obtained with both methodologies.

96
 6. Tunnel face stability considering advance drainage under transient conditions

6. Tunnel face stability considering

advance drainage under transient
conditions

6.1 Introduction

The loss of hydraulic head after excavation does not take place immediately. The tunnel

face stability under steady-state conditions, as considered in Chapter 5, is usually an

idealized analysis for soils with relatively high permeability and tunnels with slow

excavation advance rate. As shown by Anagnostou (1995), the time required to reach

the steady-state condition in soils with lower permeability is higher and the time-

dependent seepage in this type of soils significantly modifies the face stability analyses.

Furthermore, time-dependent lining deformations and ground surface settlements, both

of which are in the interest of engineering design, can appear as a result of the changing

effective pressures (Soga et al., 2017). However, Broere and Van Tol. (2000) pointed

out that in most of the models proposed over the years to study tunnel face stability, the

effects of pore water pressures are rarely introduced; also, the development of the

excess pore water pressures, which can be of major importance during construction

97
 6. Tunnel face stability considering advance drainage under transient conditions

standstills, was not dealt with. Although pore water pressures are being incorporated

into tunnel face stability mechanisms, the literature about its effects is apparently scarce,

especially concerning time-dependent characteristics of pore water pressures. Chapter

6 extends the analysis of the tunnel face stability to consider transient conditions, in

which pore pressures change over time.

After outlining the problem (Section 6.2) and explaining the computational model

(Section 6.3), two important time-dependent features are addressed: (i) the pore water

pressure evolution (Section 6.4.1); and (ii) the plastic yielding zone evolution (Section

6.4.2). The effects of advance drainage on pore water pressure distributions (Section

6.5.1), as well as on the computed tunnel face stand-up time (Section 6.5.2), are studied

under transient conditions. Finally, Section 6.5.3 investigates the surface settlements

that occur in response to the tunnel face pressure and to the advance drainage.

6.2 Problem setting

A circular tunnel with diameter D=10m and overburden C=10m, excavated under the

water table, is considered. The water table is kept at the ground surface (i.e., a sufficient

recharge from the far-field is assumed) so that the initial hydraulic head at the tunnel

axis is Hw=10m (see Figure 6.1(a)). Two drainage boreholes are considered; they are

assumed to be symmetrical in relation to the vertical plane of symmetry of the tunnel,

being located in the upper part of the tunnel face with an angle of 45°measured from

the crown and with a distance to the center of the tunnel face of r=3.8m (See Figure

6.1(b)).

98
 6. Tunnel face stability considering advance drainage under transient conditions

a)

b)

Figure 6.1. Schematic of the analyzed problem: a) in the vertical plane of

symmetry of the tunnel; b) in the tunnel face cross-section

6.3 Computational model

Figure 6.2 shows the numerical model built in FLAC3D for these analyses. Due to the

symmetry of the problem, and to improve the computational efficiency, only half of the

model is considered, with dimensions 35m×55m×50m in the x, y and z directions,

respectively (136 850 elements). The mechanical boundary conditions are defined to

avoid displacements at the lateral and lower boundaries. Similarly, and for simplicity,

the tunnel perimeter is fixed to avoid its movements, hence representing a rigid lining.

99
 6. Tunnel face stability considering advance drainage under transient conditions

The fluid boundary condition at the tunnel face is defined using a mixed boundary

condition as described by Anagnostou et al. (2016), due to the negative pore water

pressures generated by the ‘‘instantaneous’’ excavation. Accordingly, the tunnel face

is set to be impermeable when the pore water pressure at the tunnel face is negative;

when the pore pressure is positive, an atmospheric pressure is imposed, therefore

allowing water to flow out from the ground (see Figure 2.4(c)). When advance drainage

is considered, the permeability of the elements forming the drain is increased by one

order of magnitude (i.e., 10 times higher than the permeability of the ground), without

removing them from the model to avoid any mechanical effect on the settlement

because of such removal. The soil is modeled as elastic-perfectly plastic, with the Mohr-

Coulomb (MC) failure criterion and a non-associated flow rule (with a dilatancy angle

ψ=0). The MC model is used for simplicity, because the interest is on the interaction

between face stability and drainage-induced settlements. (More advanced material

models, e.g., with strain-hardening or softening behavior, could be used for an

improved prediction of surface settlements). Also, as pointed out by Anagnostou et al.

(2016), the MC model is widely used in engineering practice; there is considerable

practical experience about its parameters, and it provides in general conservative

estimates of the stand-up time. The strength parameters are a cohesion c=20kPa, and a

friction angle φ=25° (these values are used so that the tunnel face is stable in the short

term but collapses in the long term.) The deformability of the ground is defined using

a Young’s modulus E=20MPa and a Poisson’s ratio ν=0.3. Finally, a permeability k=10-
7
m/s and a water modulus of Kf=108Pa are assigned. (The FLAC3D code of numerical

model is reproduced in Appendix C- Mixboundary.dat)

100
 6. Tunnel face stability considering advance drainage under transient conditions

Figure 6.2. Numerical model with advance drainage boreholes in the analysis

To simulate the support pressure, a uniform horizontal total stress is applied at the

tunnel face; although this corresponds to an excavated face supported with air pressure,

it is also used to simulate a face reinforced with, for example, bolts (Anagnostou et al.,

2016).

Three stages are implemented in the numerical modeling (see Figure 2.4): a) stress and

pore water pressure initialization; b) excavation and undrained mechanical equilibrium:

in this stage, fluid flow is not allowed so that mechanical equilibrium can be achieved

under the negative pore pressures generated ahead of the tunnel face (see explanations

below); c) fluid-solid coupled analysis: in this stage, fluid flow and mechanical

equations are solved with the mixed boundary condition explained above, so that the

evolution of pore water pressures and settlements can be monitored.

Two cases are computed and compared: (i) without advance drainage and (ii) with

advance drainage (modeling the drainage boreholes shown in Figure 6.1(b), and

101
 6. Tunnel face stability considering advance drainage under transient conditions

keeping all the other parameters constant). The most relevant results related to pore

water pressure, tunnel face stability, and settlement, are shown and discussed in the

following sections.

6.4 Time-dependent behaviors under transient conditions

6.4.1 Pore water pressure evolution

In the second modeling stage (excavation and undrained mechanical equilibrium), the

tunnel face extrudes due to the excavation. Consequently, negative excess pore

pressures develop, having a stabilizing effect on the tunnel face. They dissipate over

time due to seepage through the ground ahead of the tunnel face. Therefore, the

unsupported tunnel face is stable in the short term, but (depending on the ground

strength) it may fail in the long term due to the dissipation of such negative pore

pressures.

Figure 6.3 presents the evolution of pore pressures ahead of tunnel face (along the

tunnel axis) without advance drainage for different times. The pore pressure distribution

under steady-state conditions, calculated with a flow-only numerical calculation, is also

shown as reference. Figure 6.3 shows that, for t=0h (undrained mechanical equilibrium)

there is no seepage, producing a maximum negative pore pressure at the tunnel face

(y=0m). Far from the tunnel face, negative pore pressures reduce until they become

positive at around y=5m. The influence of the tunnel excavation on the pore water

distribution becomes negligible, with pore pressures being similar to the initial

hydraulic head, beyond approximately y=10m. This illustrates that the negative excess

pore pressures, and their stabilizing effect, develop in a limited region ahead of the

tunnel face and decrease with the distance to the tunnel face. (Note that the collapse

mechanism usually develops in this narrow region, see e.g., Pan et al., 2016, Lü et al.,

102
 6. Tunnel face stability considering advance drainage under transient conditions

2018). Over time, negative excess pore pressures dissipate due to seepage, and the

tunnel face may become unstable (see Section 6.5.2).

Figure 6.3. Pore pressures ahead of the tunnel face along the tunnel axis for
different times without advance drainage

6.4.2 Plastic failure zone evolution

Once the tunnel is excavated (without face support), the plastic zone starts to develop

from the bottom of the tunnel face due to the stress redistribution. Figure 6.4 shows the

plastic zone contours (in red color) at different times after excavation. The soil mass at

the lower part of the tunnel face is critical for the stability of the tunnel face after

excavation, because the highest shear stresses develop from there. While negative pore

water pressures ahead of the tunnel face dissipate with time, effective stresses within

the soil reduce; consequently, the plastic zone develops upwards (towards the ground

surface) and extends over a larger area. Two elastic regions within the failure

mechanism can be observed: an elastic zone close to the tunnel face occurs due to the

initial negative excess pore pressures, which gradually decrease as the seepage proceeds;

the other elastic zone is close to the ground surface, as it shrinks with the development

of the shear bands. Anagnostou et al. (2016) described similar observations and

103
 6. Tunnel face stability considering advance drainage under transient conditions

indicated that failure occurs when the plastic zone reaches the ground surface.

Therefore, the plastic failure zone evolution can be used to diagnose the stand-up time

of the tunnel face in numerical analyses.

t=0h t=2h

t=4h t=8h

t=10h t=15h

Figure 6.4. Plastic failure zone evolution without drainage boreholes

104
 6. Tunnel face stability considering advance drainage under transient conditions

6.5 Effect of advance drainage on the tunnel face stability

6.5.1 Pore water pressure ahead of the tunnel face

Figure 6.5 shows the evolution of pore pressures along the tunnel axis, for different

times, when advance drainage boreholes are considered. Dissipation of negative excess

pore pressures and drainage seepage proceed simultaneously, although the effect of

drainage at the tunnel axis (ahead of the face) is initially almost negligible. The later

effect of advance drainage on the pore pressure distribution can be observed along the

whole length of the drains (y≤15m) and even beyond (Zingg, 2016). Similarly, the

advance drainage boreholes do not accelerate the dissipation of pore pressures around

the tunnel face. This is because the drainage boreholes do not work when the soil around

them has negative pore pressures.

Figure 6.5. Pore pressures ahead of the tunnel face along the tunnel axis for
different times after introducing advance drainage

6.5.2 Stand-up time of the tunnel face

Since the negative pore pressures dissipate with time, collapse occurs after some time,

so the stand-up time can be defined as the time that the face is stable before its collapse.

Therefore, a support pressure needs to be applied on the tunnel face to prevent failure.

105
 6. Tunnel face stability considering advance drainage under transient conditions

Figures 6.6(a) and 6.6(b) show the evolution of settlements of a monitoring point at the

surface (see Figure 6.1(a)), for different support pressures and for the two cases

considered: without and with advance drainage. This figure shows that applying a

higher support pressure (i) delays the acceleration of settlements associated with failure

and (ii) can even avoid the collapse. Note that other criteria can be used to define the

failure of the tunnel face–such as the extension of the plastic zone or the rate of

volumetric strains (see Anagnostou et al., 2016)–can be adopted when the occurrence

of failure is ambiguous.

a)

b)

Figure 6.6. Evolution of surface settlement at the monitoring point for different
support pressures: a) without advance drainage; b) with advance drainage

106
 6. Tunnel face stability considering advance drainage under transient conditions

For the case without drainage considered herein, the face collapses in less than about 3

days for support pressures lower than 50kPa. For 50kPa, the rate of settlements within

the time interval considered is small and almost constant. Correspondingly, for the case

with advance drainage, the face is stable during the first 3 days even for a support

pressure of 37.5kPa. Consequently, drainage is clearly helping to stabilize the face, as

less support pressure is required. These results are coherent with previous works (e.g.,

Zingg, 2016) and with the results shown in Chapter 5.

Figure 6.7 shows the variation of the stand-up time as a function of the support pressure,

for the two drainage cases considered. As previously explained, increasing the support

pressure increases the stand-up time, especially for higher support pressures. Moreover,

for tunnel faces with advance drainage, the stand-up time is longer than without

drainage; and differences in the stand-up time enlarge for higher support pressures.

Figure 6.7. Stand-up time vs support pressure. (Limit Analysis results have been
computed using the pore pressure distributions under steady-state conditions)

107
 6. Tunnel face stability considering advance drainage under transient conditions

The reader can note that the stand-up timelines tend towards an asymptotic support

pressure, for which the stand-up time is infinite (i.e., the tunnel face is completely

stable). Critical pressures computed with the methodology in Chapter 5 (i.e., employing

an upper bound solution in the framework of Limit Analysis, and with the steady-state

pore pressure distribution), are incorporated into Figure 6.7 as two dotted vertical lines.

Although it seems that the analytical results correspond to the asymptotic values, more

research (and longer computations) is needed to analyze this fact. (Note that, as an

illustration, the time required to calculate the stand-up time in the case of advance

drainage and σT =25kPa on an Inter Code CPU i7-5930k 3.50-GHz PC is larger than 4

days). One possible explanation is that the Limit Analysis mechanism assumes an

associated flow rule, whereas a null dilatancy angle is employed in the numerical model,

as suggested by Anagnostou et al. (2016). Other works from the literature, with dry soil,

show a limited influence of dilatancy on computed values of critical pressure (see

Senent et al., 2013).

6.5.3 Settlement

As shown in Figure 6.6, settlements at the monitoring point develop faster for the

smaller support pressures, and they do not seem to be heavily influenced by drainage.

To analyze this problem, Figure 6.8 reproduces the surface settlement troughs (along

the tunnel axis) for the minimum support pressures required to avoid failure (50kPa and

37.5kPa, respectively, for the case without and with drainage respectively; see Figure

6.6). (These values are approximate, since a detailed analysis is not carried out

following, for example, the bisection method proposed by Mollon et al., 2009a). As it

can be observed, settlements for the case with advance drainage and applying the

minimum support required for stability are slightly larger than without drainage (with

108
 6. Tunnel face stability considering advance drainage under transient conditions

maximum values of 16.1mm and 12.3 mm respectively). Similarly, settlements for the

same support pressure (σT=50kPa) are also larger for the case with drainage.

Consequently, it is clear that drainage increases surface settlements and that the larger

settlements associated to advance drainage develop due to the combined effect of both

(i) the higher seepage flow and (ii) the lower required support pressure.

Figure 6.8. Longitudinal surface trough for different support pressures and
drainage configurations

6.6 Conclusions

In this chapter, the face stability of tunnels under the water table was studied,

considering transient conditions and the presence of advance drainage; in addition, the

resulting surface settlements were also analyzed. Employing a numerical model of a

shallow tunnel, the evolution of negative excess pore pressures ahead of the tunnel face

is analyzed, and their effect on its stability is also assessed. Results show that the

negative excess pore pressures at the tunnel face dissipate gradually, due to seepage,

with the plastic zone developing gradually, so that the tunnel face may become unstable

after some time, if the plastic zones reach the ground surface. Increasing the support

pressure applied on the tunnel face delays such failure and can even completely avoid

109
 6. Tunnel face stability considering advance drainage under transient conditions

it. Similarly, drainage increases the stability of the face, as less support pressure is

required.

The Limit Analysis failure mechanism proposed in Chapter 5 is used to compute the

critical pressure in both drainage situations, employing the steady-state pore water

pressure distribution. Although more research is needed, results show that the analytical

solution can serve to preliminarily estimate the support pressures required for long-term

face stability analyses, avoiding an extremely time-consuming computation under

transient conditions.

Finally, advance drainage leads to larger surface settlements due to the combined effect

of (i) the higher seepage flow and (ii) the reduced support pressure required for stability.

Therefore, tunnels with advance drainage require lower support pressures to maintain

their face stability, but lead to larger surface settlements that may introduce additional

deformation-related risks that should be considered.

110
 7. Conclusions and outlook

7. Conclusions and outlook

7.1 Conclusions

In this thesis, three advanced face stability problems–i.e., considering the free span,

reinforcement with a forepole umbrella, and drainage of pore water pressures in tunnels

under the water table–have been investigated.

In the first part of this thesis, a failure mechanism considering the free span, in the

framework of the upper bound theorem of Limit Analysis, is provided (Chapter 3); then,

it is extended to analyze the reinforcement effect of a forepole umbrella (Chapter 4).

The main findings can be summarized as follows:

1. The proposed 2D analytical Limit Analysis failure mechanism provides an

efficient tool to obtain the face failure geometry, and the face collapse pressure,

of tunnels with a free span. As expected, the increase of the free span reduces

the stability of the tunnel face, and the mechanism is shown to be more accurate

for soils with higher frictional angles (i.e., φ≥35°). However, the effect of the

free span on the collapse pressure is less relevant than the effect of the soil

strength (for typical values considered in tunneling). It is also shown that the

length of the maximum self-stable free span is proportional to the soil cohesion.

111
 7. Conclusions and outlook

2. The length of the free span barely affects the rotational failure geometry ahead

of the tunnel face, but the failure mechanism above the free span extends further

upwards as the free span length increases; similarly, the velocity vector of the

transition block becomes slightly more horizontal for larger free spans. The

cohesion of the soil mass has an almost negligible effect on the failure geometry,

but (as expected) lower friction angles cause larger failure mechanisms.

3. The failure modes of the forepole element (shear and bending moment) can be

integrated into the Limit Analysis mechanism with a free span, to study the

reinforcement effect of the forepole umbrella to improve the tunnel face stability.

The collapse pressure of the tunnel face is reduced as the forepole umbrella

becomes heavier, although a minimum support pressure corresponding to a

failure mechanism developing below the umbrella–may still be required for

stability, independently of the umbrella design. An accurate prediction of the

collapse pressure can be computed when a bending moment failure, with a

thickness of the continuous deformation zone, is assumed. The thickness of the

continuous deformation zone can be estimated using an equation presented in

the thesis, that was obtained through the fitting of numerical simulation results.

In the second part of this thesis, the 3D Limit Analysis failure mechanism of Mollon et

al. (2011a) is extended to consider pore water pressures at the tunnel face, and such

extended mechanism is employed to study the effect of advance drainage boreholes

under steady-state conditions (Chapter 5); then, the stand-up time and the surface

settlements that occur in response to the face support pressure are computed,

considering transient flow conditions and (if applicable) the influence of advance

112
 7. Conclusions and outlook

drainage boreholes installed at the tunnel face (Chapter 6). The principal conclusions

of this part can be summarized as follows:

1. The proposed methodology is based on a 3D Limit Analysis failure mechanism,

and it employs as input the results of a numerical simulation (under steady-state

conditions) of the pore water pressure distribution ahead of the tunnel face; then,

it allows one to obtain, easily and accurately, the collapse pressure and the

failure geometry of the tunnel face considering advance drainage boreholes.

2. The layout of the advance drainage boreholes affects the distribution of pore

water pressures ahead of the tunnel face, and hence the face support pressure; it

also affects the failure geometry, since the collapse mechanism at the tunnel

face tends to develop towards the region with higher pore water pressures.

Similarly, and according to the collapse pressure obtained after considering the

advance drainage, a “lower” drainage design is more effective when the water

table is lower (Hw/D<3), whereas an “upper” drainage design is recommended

when the water table is higher (Hw/D>4).

3. Under transient conditions, the advance drainage can increase the stand-up time

of the tunnel face, particularly when higher support pressures are applied on the

tunnel face. However, larger surface settlements occur when drainage boreholes

are introduced to reduce the pore water pressures ahead of the tunnel face.

Therefore, engineers must balance the different available measures to control

the stability of the tunnel face (e.g., drainage and support pressure) while still

preventing excessive settlements from developing.

4. The Limit Analysis methodology assuming steady-state conditions can be

employed to preliminarily predict the required support pressure in the long term.

113
 7. Conclusions and outlook

This may provide an adequate tool for preliminary analyses, especially since the

computations of the face stand-up time under transient conditions are very time-

consuming.

7.2 Outlook

Future works could be conducted in the following areas.

1. The failure geometries obtained by the analytical solutions for both analyses

(i.e., free span and drainage) are sharper in the upper part than those computed

by the numerical model. Moreover, the failure geometry obtained by the

proposed free span mechanism is not completely consistent with the numerical

results in the region of the transitional block that forms above the tunnel crown.

Consequently, a deeper study is recommended to improve predictions of failure

geometry for these mechanisms.

2. The proposed Limit Analysis failure mechanisms, employed to study the

stability of the tunnel face considering the free span and drainage, are adequate

to reproduce the behavior of shallow tunnels, but they are not applicable to deep

tunnels. This is because the soil arching effect that develops in deep tunnels is

not yet considered into these Limit Analysis mechanisms (see e.g., Zou et al.,

2019; Li et al., 2020).

3. A more advanced soil constitutive model could be used to analyze the coupled

(fluid-solid) tunnel face stability behavior that develops under transient

conditions, due to the limited applicability of the Mohr-Coulomb model for

predictions of long-term surface settlements.

114
 7. Conclusions and outlook

4. The influence of several properties that may affect the behavior of the tunnel

face under transient conditions–such as the permeability coefficient, soil

strength parameters, tunnel advance rate, etc.–were not investigated in this

thesis, although interesting practical results may be obtained from such analyses.

115
 References

References

Ahmed, M., & Iskander, M., 2012. Evaluation of tunnel face stability by transparent
soil models. Tunnelling and Underground Space Technology, 27(1), 101-110.

Alagha, A. S., & Chapman, D. N., 2019. Numerical modelling of tunnel face stability
in homogeneous and layered soft ground. Tunnelling and Underground Space
Technology, 94, 103096.

Anagnostou, G., 1995. The influence of tunnel excavation on the hydraulic head.
International Journal for numerical and analytical methods in geomechanics, 19(10),
725-746.

Anagnostou, G., 2014. Some critical aspects of subaqueous tunneling, ITA-AITES

Wolrd Tunnel Congress (WTC2014). Muir Wood Lecture, Iguassu, Brazil, 1-20.

Anagnostou, G. & Kovári, K., 1994. The face stability of slurry-shield-driven tunnels.
Tunnelling and underground space technology, 9(2), 165-174.

Anagnostou, G. & Kovári, K., 1996. Face stability conditions with earth-pressure-
balanced shields. Tunneling and underground space technology, 11 (2), 165-173.

Anagnostou, G. & Perazzelli, P., 2013. The stability of a tunnel face with a free span
and a non‐uniform support. Geotechnik, 36(1), 40-50.

Anagnostou, G. & Perazzelli, P., 2015. Analysis method and design charts for bolt
reinforcement of the tunnel face in cohesive-frictional soils. Tunnelling and
Underground Space Technology, 47, 162-181.

Anagnostou, G., Schuerch, R., Perazzelli, P., Vrakas, A., Maspoli, P. & Poggiati, R.,
2016. Tunnel face stability and tunnelling induced settlements under transient
conditions. Eidgenössisches Departement für UVEK, Bundesamt für Strassen, Zurich.
(Technical report)

Anagnostou, G. & Serafeimidis, K., 2007. The dimensioning of tunnel face

reinforcement. ITA-AITES World Tunnel Congress (WTC2007), Prague, 291-296.

117
 References

Anagnostou, G. & Zingg, S., 2013. On the stabilizing effect of advance drainage in
tunnelling/Über die statische Auswirkung der vorauseilenden Gebirgsdrainagen im
Tunnelbau. Geomechanics and Tunnelling, 6(4), 338-354.

Attewell, P.B., 1988. An overview of site investigation and long-term tunnelling-

induced settlement in soil. Geological Society, London, Engineering Geology Special
Publications, 5(1), 55-61.

Barla, G., 2000. Lessons learnt from the excavation of a large diameter TBM tunnel in
complex hydrogeological conditions, GeoEng2000. Technomic Publishing Company,
Melbourne, Australia, 938-996.

Baudendistel, M., 1985. Significance of the unsupported span in tunnelling. Tunneling

85, 103-108.

Bishop, A.W. & Morgenstern, N.R., 1960. Stability coefficients for earth slopes.
Géotechnique, 10(4), 129-150.

Broere, W., 1998. Face stability calculation for a slurry shield in heterogeneous soft
soils. Tunnels and metropolises, 23, 215-218.

Broere, W., 2002. Influence of excess pore pressures on the stability of the tunnel face.
International Symposium of Geotechnical Aspects of Underground Construction in
Soft Ground, Toulouse, France, 179–184.

Broere, W., 2003. Influence of excess pore pressures on the stability of the tunnel face.
(Re) Claiming the Underground Space, Amsterdam, 759-765.

Broere, W., 2015. On the face support of microtunnelling TBMs. Tunnelling and
Underground Space Technology, 46, 12-17.

Broere, W. & Van Tol, A.F., 2000. Influence of infiltration and groundwater flow on
tunnel face stability. International Symposium of Geotechnical aspects of underground
construction in soft ground, 1, 339-344.

Broms, B.B., 1964. Lateral resistance of piles in cohesive soils. Journal of the Soil
Mechanics and Foundation Engineering, 90(2), 27–64.

Broms, B.B. & Bennermark, H., 1967. Stability of clay at vertical openings. Journal of
the Soil Mechanics and Foundation Engineering, 193(SM1):71–94.

Callari, C., 2015. Numerical assessment of tunnel face stability below the water table.
Proceedings of the 14th International Conference of International Association for
Computer Methods and Recent Advances in Geomechanics, Kyoto, 2007-2010.

Callari, C., Alsahly, A., & Meschke, G., 2017. Assessment of stand-up time and
advancement rate effects for tunnel faces below the water table. Proceedings of the 4th
International Conference on Computational Methods in Tunneling and Subsurface
Engineering (EURO: TUN 2017). University Innsbruck, Austria.

118
 References

Calvello, M. & Taylor, R.N., 1999. Centrifuge modelling of a spile-reinforced tunnel

heading. International Symposium of Geotechnical Aspects of Underground
Construction in Soft Ground, Tokyo, 345-350.

Casarin, C. & Mair, R. J., 1981. The assessment of tunnel stability in clay by model
tests. Soft Ground Tunneling, A. A. Balkema, Rotterdam, 33-44.

Cattoni, E., Miriano, C., Boco, L., & Tamagnini, C., 2016. Time-dependent ground
movements induced by shield tunneling in soft clay: a parametric study. Acta
Geotechnica, 11(6), 1385-1399.

Chai, J.C., Shen, S.L., Zhu, H.H., & Zhang, X.L., 2004. Land subsidence due to
groundwater drawdown in Shanghai. Géotechnique, 54(2), 143-147.

Chambon, P. & Corte, J.F., 1994. Shallow tunnels in cohesionless soil: stability of
tunnel face. Journal of Geotechnical Engineering, 120(7), 1148-1165.

Chen, G. H., Zou, J. F., & Qian, Z. H., 2019. An improved collapse analysis mechanism
for the face stability of shield tunnel in layered soils. Geomechanics and Engineering,
17(1), 97-107.

Chen, R. P., Li, J., Kong, L. G., & Tang, L. J., 2013. Experimental study on face
instability of shield tunnel in sand. Tunnelling and Underground Space Technology, 33,
12-21.

Chen, R. P., Tang, L. J., Ling, D. S., & Chen, Y. M., 2011. Face stability analysis of
shallow shield tunnels in dry sandy ground using the discrete element method.
Computers and Geotechnics, 38(2), 187-195.

Chen, R. P., Tang, L. J., Yin, X. S., Chen, Y. M., & Bian, X. C., 2015. An improved
3D wedge-prism model for the face stability analysis of the shield tunnel in
cohesionless soils. Acta Geotechnica, 10(5), 683-692.

Chen, R., Yin, X., Tang, L., & Chen, Y., 2018. Centrifugal model tests on face failure
of earth pressure balance shield induced by steady state seepage in saturated sandy silt
ground. Tunnelling and Underground Space Technology, 81, 315-325.

Chen, W.F., 1975. Limit Analysis and Soil Plasticity. Elsevier, Amsterdam.

Cornejo, L., 1989. Instability at the face: its repercussions for tunnelling technology.
Tunnels & tunnelling, 21(4), 69-74.

Costa, A., Cardoso, A.S., & Sousa, J.A., 2007. The influence of unlined excavation
length in face stability. In Applications of Computational Mechanics in Geotechnical
Engineering V, CRC Press, 228-235.

Date, K., Mair, R.J., & Soga, K., 2008. Reinforcing effects of forepoling and facebolts
in tunnelling. International Symposium of Geotechnical Aspects of Underground
Construction in Soft Ground, Shanghai, 635-641.

119
 References

Davis, E.H., Gunn, M.J., Mair, R.J., & Seneviratine, H.N., 1980. The stability of
shallow tunnels and underground openings in cohesive material. Géotechnique, 30(4),
397-416.

De Buhan, P., Cuvillier, A., Dormieux, L., & Maghous, S., 1999. Face stability of
shallow circular tunnels driven under the water table: a numerical analysis.
International Journal for Numerical and Analytical Methods in Geomechanics, 23(1),
79-95.

De Buhan, P. & Salencon, J., 1993. A comprehensive stability analysis of soil nailed
structures. European Journal of Mechanics A/Solids, 12: 325–345.

Ellstein, R. A., 1986. Heading failure of lined tunnels in soft soils. Tunnels & tunnelling,
18(6), 51-54.

Elyasi, A., Javadi, M., Moradi, T., Moharrami, J., Parnian, S. & Amrac, M., 2016.
Numerical modeling of an umbrella arch as a pre-support system in difficult geological
conditions: a case study. Bulletin of engineering geology and the environment, 75(1),
211-221.

Evans, C.H., 1984. An examination of arching in granular soils. Doctoral dissertation,

Massachusetts Institute of Technology.

Federal Highway Administration (FHA), 2009. Technical Manual for Design and
Construction of Road Tunnel–Civil Elements. Department of Transportation, U.S.

Greenwood, J.D., 2003. Three-dimensional analysis of surface settlement in soft ground

tunneling. Doctoral dissertation, Massachusetts Institute of Technology.

Health and Safety Executive (HSE), 1996. Safety of new Austrian tunnelling method
(NATM) tunnels. HSE Books, London.

Hieu, N.T., Giao, P.H. & Phien-wej, N., 2020. Tunneling induced ground settlements
in the first metro line of Ho Chi Minh City, Vietnam. Geotechnics for Sustainable
Infrastructure Development, Hanoi, 297-304.

Höfle, R., Fillibeck, J., & Vogt, N., 2008. Time dependent deformations during
tunnelling and stability of tunnel faces in fine-grained soils under groundwater. Acta
Geotechnica, 3(4), 309-316.

Hong, E.S., Shin, H.S., Kim, H.M., Park, C., & Park, E.S., 2007. Numerical study on
horizontal pre-drainage system using horizontal directional drilling in subsea tunnelling.
Chinese Journal of Rock Mechanics and Engineering, 26(S2), 3697-3703.

Horn, M., 1961. Alagutak homlokbiztositására ható vizszintes földnyomásvizsgálat

néhány eredménye. Az országos mélyépitóipari konferencia elóadásai, Közlekedési
Dokumentációs Vállalat, Budapest. (In Hungarian)
.
Itasca Consulting Group, 2009. FLAC3D manual, Version 4.0, Minneapolis.

120
 References

Janssen, H.A., 1895. Versuche fiber Getreidedruck in SilozeUen. Zeitschrift des

Vereins deutscher Ingenieure, Band XXXIX, No. 35, 1045-1049.

Juneja, A., Hegde, A., Lee, F.H., & Yeo, C.H., 2010. Centrifuge modelling of tunnel
face reinforcement using forepoling. Tunnelling and Underground Space Technology,
25(4), 377-381.

Jurík, I., Grega, L., Valko, J., & Janega, P., 2017. New drainage tunnel of the tunnel
Višňové–design and excavation. Conference Series: Materials Science and Engineering,
IOP Publishing, Vol.236:012086.

Kamata, H. & Mashimo, H., 2003. Centrifuge model test of tunnel face reinforcement
by bolting. Tunnelling and Underground Space Technology, 18(2-3), 205-212.

Kasper, T. & Meschke, G., 2006a. A numerical study of the effect of soil and grout
material properties and cover depth in shield tunnelling. Computers and Geotechnics,
33(4-5), 234-247.

Kasper, T. & Meschke, G., 2006b. On the influence of face pressure, grouting pressure
and TBM design in soft ground tunnelling. Tunnelling and Underground Space
Technology, 21(2), 160-171.

Kielbassa, S., & Duddeck, H., 1991. Stress-strain fields at the tunnelling face–Three-
dimensional analysis for two-dimensional technical approach. Rock mechanics and
rock engineering, 24(3), 115-132.

Kimura, T. & Mair, R.J., 1981. Centrifugal testing of model tunnels in soft clay. 10th
International Conference of Soil Mechanics and Foundation Engineering, Stockholm,
319–322.

Kirsch, A., 2010. Experimental investigation of the face stability of shallow tunnels in
sand. Acta Geotechnica, 5(1), 43-62.

Kitchah, F. & Benmebarek, S., 2016. Finite difference analysis of an advance core pre-
reinforcement system for Toulon's south tube. Journal of Rock Mechanics and
Geotechnical Engineering, 8(5), 703-713.

Klotoé, C.H. & Bourgeois, E., 2019. Three dimensional finite element analysis of the
influence of the umbrella arch on the settlements induced by shallow tunneling.
Computers and Geotechnics, 110, 114-121.

Laver, R.G., Li, Z., & Soga, K., 2017. Method to evaluate the long-term surface
movements by tunneling in London clay. Journal of Geotechnical and
Geoenvironmental Engineering, 143(3), 06016023.

Le, B.T. & Taylor, R.N., 2017. The reinforcing effects of Forepoling Umbrella System
in soft soil tunnelling. Proceedings of the 19th International Conference on Soil
Mechanics and Geotechnical Engineering, Seoul, 1709-1712.

121
 References

Leca, E. & Dormieux, L., 1990. Upper and lower bound solutions for the face stability
of shallow circular tunnels in frictional material. Géotechnique, 40(4), 581-606.

Lee, I.M., Lee, J.S., & Nam, S.W., 2004. Effect of seepage force on tunnel face stability
reinforced with multi-step pipe grouting. Tunnelling and underground space technology,
19(6), 551-565.

Lee, I.M. & Nam, S.W., 2001. The study of seepage forces acting on the tunnel lining
and tunnel face in shallow tunnels. Tunnelling and Underground Space Technology,
16(1), 31-40.

Lee, I.M. & Nam, S.W., 2004. Effect of tunnel advance rate on seepage forces acting
on the underwater tunnel face. Tunnelling and Underground Space Technology, 19(3),
273-281.

Lee, I.M., Nam, S.W., & Ahn, J.H., 2003. Effect of seepage forces on tunnel face
stability. Canadian Geotechnical Journal, 40 (2), 342-350.

Lee, Y.Z. & Schubert, W., 2008. Determination of the round length for tunnel
excavation in weak rock. Tunnelling and underground space technology, 23(3), 221-
231.

Lee, Y.Z., Schubert, W., & Chang-Yong, K., 2005. The influence of the round length
on the stability of tunnel face and unsupported span. ITA-AITES World Tunnel
Congress (WTC2005), Istanbul, 211-216.

Li, C., Miao, L., & Lv, W., 2011. Numerical analysis of face stability during shield-
driven tunneling under groundwater table. Proceedings of the Geo-Frontiers 2011:
Conference: Advances in Geotechnical Engineering, Dallas, 3796-3806.

Li, P., Zou, H., Wang, F., & Xiong, H., 2020. An analytical mechanism of limit support
pressure on cutting face for deep tunnels in the sand. Computers and Geotechnics, 119,
103372.

Li, W., Zhang, C., Tan, Z., & Ma, M., 2021. Effect of the seepage flow on the face
stability of a shield tunnel. Tunnelling and Underground Space Technology, 112,
103900.

Li, Z.K., Huang, Y.R., Wang, J.Y., & Zi, Q., 2012. 3-D FEM seepage simulation of
channel with drain pipes. Applied Mechanics and Materials. Vol.137, 198-204.

Liu, X. Y., Wang, F. M., Fang, H. Y., & Yuan, D. J., 2019. Dual-failure-mechanism
model for face stability analysis of shield tunneling in sands. Tunnelling and
Underground Space Technology, 85, 196-208.

Lunardi, P., 2008. Design and construction of tunnels: Analysis of Controlled

Deformations in Rock and Soils (ADECO-RS). Springer, Berlin.

122
 References

Lü, X., Wang, H., & Huang, M., 2014. Upper bound solution for the face stability of
shield tunnel below the water table. Mathematical Problems in Engineering 2014,
2014:1-11.

Lü, X., Zhou, Y., Huang, M., & Li, F., 2017. Computation of the minimum limit support
pressure for the shield tunnel face stability under seepage condition. International
Journal of Civil Engineering, 15(6), 849-863.

Lü, X., Zhou, Y., Huang, M., & Zeng, S., 2018. Experimental study of the face stability
of shield tunnel in sands under seepage condition. Tunnelling and Underground Space
Technology, 74, 195-205.

Lyamin, A.V. & Sloan, S.W., 2002a. Lower bound limit analysis using non-linear
programming. International Journal for Numerical Methods in Engineering, 55(5), 573-
611.

Lyamin, A.V. & Sloan, S.W., 2002b. Upper bound limit analysis using linear finite
elements and non-linear programming. International Journal for Numerical and
Analytical Methods in Geomechanics, 26(2), 181-216.

Mair, R.J., 2008. Tunnelling and geotechnics: new horizons. Géotechnique, 58(9),.695-
736.

Michalowski, R.L., 2005. Limit analysis in geotechnical engineering. State of the art
report on selected topics by TC-34: Prediction and simulation methods in geomechanics,
Technical Committee 34 of ISSMGE.

Ministerio de Formento, 1990. Guía para el proyecto y la ejecución de micropilotes en

obras de carretera. Centro de Publicaciones, Secretaría General Técnica, Ministerio de
Fomento, Madrid. (In Spanish)

Mollon, G., Dias, D., & Soubra, A. H., 2009a. Probabilistic analysis of circular tunnels
in homogeneous soil using response surface methodology. Journal of Geotechnical and
Geoenvironmental Engineering, 135(9), 1314-1325.

Mollon, G., Dias, D., & Soubra, A. H., 2009b. Probabilistic analysis and design of
circular tunnels against face stability. International Journal of Geomechanics, 9(6), 237-
249.

Mollon, G., Dias, D., & Soubra, A. H., 2010. Face stability analysis of circular tunnels
driven by a pressurized shield. Journal of Geotechnical and Geoenvironmental
Engineering, 136(1), 215-229.

Mollon, G., Dias, D., & Soubra, A.H., 2011a. Rotational failure mechanisms for the
face stability analysis of tunnels driven by a pressurized shield. International Journal
for Numerical and Analytical Methods in Geomechanics, 35(12), 1363-1388.

Mollon, G., Phoon, K.K., Dias, D., & Soubra, A.H., 2011b. Validation of a new 2D
failure mechanism for the stability analysis of a pressurized tunnel face in a spatially
varying sand. Journal of Engineering Mechanics, 137, 8-21.

123
 References

Muraki, Y., 1997. The umbrella method in tunnelling. Doctoral dissertation,

Massachusetts Institute of Technology.

Nagel, F., Stascheit, J., & Meschke, G., 2012. Numerical simulation of interactions
between the shield-supported tunnel construction process and the response of soft
water-saturated soils. International Journal of Geomechanics, 12(6), 689-696.

Nemati Hayati, A., Ahmadi, M.M., Hajjar, M., & Kashighandi, A., 2013. Unsupported
advance length in tunnels constructed using New Austrian Tunnelling Method and
ground surface settlement. International Journal for Numerical and Analytical Methods
in Geomechanics, 37(14), 2170-2185.

Ng, C.W.W. & Lee, G.T., 2002. A three-dimensional parametric study of the use of soil
nails for stabilising tunnel faces. Computers and Geotechnics, 29(8), 673-697.

Oblozinsky, P. & Kuwano, J., 2004. Centrifuge experiments on stability of tunnel face.
Slovak Journal of Civil Engineering, 3, 23-29.

Ocak, I. & Selcuk, E., 2017. Comparison of NATM and umbrella arch method in terms
of cost, completion time, and deformation. Arabian Journal of Geosciences, 10(7):177.

Oke, J., Vlachopoulos, N., & Marinos, V., 2014. Umbrella arch nomenclature and
selection methodology for temporary support systems for the design and construction
of tunnels. Geotechnical and geological engineering, 32(1), 97-130.

Optum Computational Engineering (OptumCE), 2019. OptumG2 manual, Version 2019,

Copenhagen.

O'Reilly, M.P., Mair, R.J., & Alderman, G.H., 1991. Long-term settlements over
tunnels: an eleven-year study at Grimsby. Proceedings of Conference Tunnelling,
Institution of Mining and Metallurgy, London, 55–64.

Pan, Q. & Dias, D., 2016. The effect of pore water pressure on tunnel face stability.
International Journal for Numerical and Analytical Methods in Geomechanics, 40(15),
2123-2136.

Pan, Q. & Dias, D., 2017. Safety factor assessment of a tunnel face reinforced by
horizontal dowels. Engineering Structures, 142, 56-66.

Pan, Q. & Dias, D., 2018. Three dimensional face stability of a tunnel in weak rock
masses subjected to seepage forces. Tunnelling and Underground Space Technology,
71, 555-566.

Paternesi, A., Schweiger, H. F., & Scarpelli, G., 2017. Numerical analyses of stability
and deformation behavior of reinforced and unreinforced tunnel faces. Computers and
Geotechnics, 88, 256-266.

Peila, D., 1994. A theoretical study of reinforcement influence on the stability of a

tunnel face. Geotechnical & Geological Engineering, 12(3), 145-168.

124
 References

Pelizza, S. & Peila, D., 1993. Soil and rock reinforcements in tunnelling. Tunnelling
and Underground Space Technology, 8(3), 357-372.

Pellet, F., Descoeudres, F., & Egger, P., 1993. The effect of water seepage forces on
the face stability of an experimental microtunnel. Canadian Geotechnical Journal, 30(2),
363-369.

Perazzelli, P., Leone, T. & Anagnostou, G., 2014. Tunnel face stability under seepage
flow conditions. Tunnelling and Underground Space Technology, 43, 459-469.

Phienwej, N., Sirivachiraporn, A., Timpong, S., Tavaranum, S., & Suwansawat, S.,
2006. Characteristics of ground movements from shield tunnelling of the first Bangkok
subway line. Proceedings of the International Symposium on Underground Excavation
and Tunnelling, Bangkok, Thailand, 319-330.

Pinyol, N.M. & Alonso, E.E., 2011. Design of micropiles for tunnel face reinforcement:
undrained upper bound solution. Journal of geotechnical and geoenvironmental
engineering, 138(1), 89-99.

Pinyol, N.M. & Alonso, E. E., 2013. An undrained upper bound solution for the face
stability of tunnels reinforced by micropiles. Proceedings of the 18th International
Conference on Soil Mechanics and Geotechnical Engineering, Paris, 3395-3398.

Qarmout, M., König, D., Gussmann, P., Thewes, M., & Schanz, T., 2019. Tunnel face
stability analysis using Kinematical Element Method. Tunnelling and Underground
Space Technology, 85, 354-367.

Qian, Z., Zou, J., Pan, Q., & Dias, D., 2019. Safety factor calculations of a tunnel face
reinforced with umbrella pipes: A comparison analysis. Engineering Structures, 199,
109639.

Rao, P., Zhao, L., Chen, Q., & Li, L., 2017. Limit analysis approach for accessing
stability of three-dimensional (3-D) slopes reinforced with piles. Marine Georesources
& Geotechnology, 35(7), 978-985.

Schuerch, R. & Anagnostou, G., 2013a. Analysis of the stand-up time of the tunnel face.
ITA-AITES World Tunnel Congress (WTC2013), Geneva, Switzerland, 709-714.

Schuerch, R. & Anagnostou, G., 2013b. The influence of the shear strength of the
ground on the stand-up time of the tunnel face. Proceedings of International Symposium
on Tunnelling and Underground Space Construction for Sustainable Development (TU-
Seoul 2013), Seoul, Korea, 297-300.

Schuerch, R. & Anagnostou, G., 2014. Failure propagation and mesh-dependency in

coupled hydraulic-mechanical transient problems. Computer Methods and Recent
Advances in Geomechanics, 329-334.

125
 References

Schuerch, R. & Anagnostou, G., 2015. Delayed failure identification by coupled

hydraulic-mechanical numerical analyses. Proceedings of the 14th International
Conference of International Association for Computer Methods and Recent Advances
in Geomechanics, Kyoto, 1061-1066.

Schuerch, R., Poggiati, R. & Anagnostou, G., 2016. Design diagrams for estimating
tunnel face stand-up time in water-bearing ground. ITA-AITES World Tunnel Congress
(WTC2016), San Francisco.

Senent, S., García-Luna, R., Sánchez-Lázaro, M., & Jimenez, R., 2019. Tunnel face
stability laboratory tests in sand considering surface settlements. Proceedings of the
XVII ECSMGE-2019, Reykjavik.

Senent, S. & Jimenez, R., 2015. A tunnel face failure mechanism for layered ground,
considering the possibility of partial collapse. Tunnelling and Underground Space
Technology, 47, 182-192.

Senent, S. & Jimenez, R., 2017. Study of the tunnel face stability considering the free
span by means of numerical models and analytical solutions. Proceedings of the IV
International Conference on Computational Methods in Tunneling and Subsurface
Engineering (EURO: TUN 2017), Innsbruck University, Austria.

Senent, S., Mollon, G., & Jimenez, R., 2013. Tunnel face stability in heavily fractured
rock masses that follow the Hoek–Brown failure criterion. International Journal of
Rock Mechanics and Mining Sciences, 60, 440-451.

Shen, S.L., Wu, H.N., Cui, Y.J., & Yin, Z.Y., 2014. Long-term settlement behaviour of
metro tunnels in the soft deposits of Shanghai. Tunnelling and Underground Space
Technology, 40, 309-323.

Shin, J.H., Addenbrooke, T.I., & Potts, D.M., 2002. A numerical study of the effect of
groundwater movement on long-term tunnel behaviour. Géotechnique, 52(6), 391-403.

Shirlaw, J.N., Busbridge, J.R., & Yi, X., 1994. Consolidation settlements over tunnels:
A Review. Canadian Tunnelling, 253-265.

Shirlaw, J.N., Ong, J.C.W., Osborne, N.H., & Tan, C.G., 2002. The relationship
between face pressure and immediate settlement due to tunnelling for the North East
Line, Singapore. International Symposium of Geotechnical Aspects of Underground
Construction in Soft Ground, Toulouse, 306–311.

Sloan, S.W., 1988. Lower bound limit analysis using finite elements and linear
programming. International Journal for Numerical and Analytical Methods in
Geomechanics, 12(1), 61-77.

Sloan, S.W., 1989. Upper bound limit analysis using finite elements and linear
programming. International Journal for Numerical and Analytical Methods in
Geomechanics, 13(3), 263-282.

126
 References

Soga, K., Laver, R.G., & Li, Z., 2017. Long-term tunnel behaviour and ground
movements after tunnelling in clayey soils. Underground Space, 2(3), 149-167.

Standing, J. R., & Burland, J. B., 2006. Unexpected tunnelling volume losses in the
Westminster area, London. Géotechnique, 56(1), 11-26.

Takano, D., Otani, J., Nagatani, H., & Mukunoki, T., 2006. Application of x-ray CT on
boundary value problems in geotechnical engineering: research on tunnel face failure.
Proceedings of Geocongress, Proceedings of ASCE, Atlanta.

Takeuchi, H., Mitarashi, Y., Suzuki, M., Matsushige, M., & Nakagawa, K., 1999. A
Proposal of Design Criteria on Long Fore Piling, Proceedings of the Japan-Korea Joint
Symposium on Rock Engineering, Fukuoka, 199-206.

Tamez, E., 1984. Estabilidad de Túneles Excavados en Suelos, Trabajo de ingreso a la

Academia Mexicana de Ingeniería, México. (In Spanish)

Tang, X. W., Gan, P. L., Liu, W., & Zhao, Y., 2017. Surface settlements induced by
tunneling in permeable strata: a case history of Shenzhen Metro. Journal of Zhejiang
University-Science A, 18(10), 757-775.

Terzaghi, K., 1936. Stress distribution in dry and in saturated sand above a yielding
trap-door. 1st International Conference on Soil Mechanics and Foundation Engineering,
Cambridge, Massachusetts, 307-311.

The MathWorks, Inc, 2017. MATLAB (MATrix LABoratory), Version R2017b,

Massachusetts.

Tian, X., Song, Z., Zhou, G., & Zhang, X., 2020. A Theoretical Calculation Method of
the Unsupported Span for the Shallow Tunnel in the Soft Stratum. Advances in Civil
Engineering, 2020(8), 1–11.

Tu, X.M., Yang, Y.Y., & Wang, G.H., 2012. Face stability analysis of rectangular
tunnels driven by a pressurized shield. Advanced Materials Research, Trans Tech
Publications Ltd, 461-465.

Ukritchon, B., Yingchaloenkitkhajorn, K., & Keawsawasvong, S., 2017. Three-

dimensional undrained tunnel face stability in clay with a linearly increasing shear
strength with depth. Computers and Geotechnics, 88, 146-151.

Van Tol, A.F., 2006. The effects of tunnelling on existing structures. Geotechnical
Aspect of Underground Construction in Soft Ground; Bakker et al. eds. Taylor and
Francis. London. pp. 31-41.

Vermeer, P.A., Ruse, N., & Marcher, T., 2002. Tunnel heading stability in drained
ground. Felsbau, 20(6), 8-18.

Viratjandr, C. & Michalowski, RL, 2006. Limit analysis of submerged slopes subjected
to water drawdown. Canadian Geotechnical Journal, 43 (8), 802-814.

127
 References

Volkmann, G. M. & Schubert, W. 2007. Geotechnical model for pipe roof supports in
tunneling. ITA-AITES World Tunnel Congress (WTC2007), Prague, 755-760.

Wang, H. & Jia, J., 2008. Analytical method for mechanical behaviors of pipe roof
reinforcement. International Conference on Information Management, Innovation
Management and Industrial Engineering, Institute of Electrical and Electronics
Engineers (IEEE), 352-357.

Wang, Z., Wong, R.C., Li, S., & Qiao, L., 2012. Finite element analysis of long-term
surface settlement above a shallow tunnel in soft ground. Tunnelling and underground
space technology, 30, 85-92.

Weng, X., Sun, Y., Yan, B., Niu, H., Lin, R., & Zhou, S., 2020. Centrifuge testing and
numerical modeling of tunnel face stability considering longitudinal slope angle and
steady state seepage in soft clay. Tunnelling and Underground Space Technology, 101,
p.103406.

Wongsaroj, J., Soga, K., & Mair, R.J., 2007. Modelling of long-term ground response
to tunnelling under St James's Park, London. Géotechnique, 57(1), 75-90.

Wongsaroj, J., Soga, K., & Mair, R.J., 2013. Tunnelling-induced consolidation
settlements in London Clay. Géotechnique, 63(13), 1103-1115.

Yan, S., Wang, R., Bai, J., Wu, W., & Elmo, D., 2020. An alternative approach to
determine cycle length of roadway excavation in coal mines. Journal of Ambient
Intelligence and Humanized Computing, 11(2), 553-560.

Yoo, C. & Shin, H.K., 2003. Deformation behaviour of tunnel face reinforced with
longitudinal pipes—laboratory and numerical investigation. Tunnelling and
Underground Space Technology, 18(4), 303-319.

Yu, L., Zhang, D., Fang, Q., Cao, L., Zhang, Y., & Xu, T., 2020. Face stability of
shallow tunnelling in sandy soil considering unsupported length. Tunnelling and
Underground Space Technology, 102, p.103445.

Zhang, C., Han, K., & Zhang, D., 2015. Face stability analysis of shallow circular
tunnels in cohesive–frictional soils. Tunnelling and Underground Space Technology,
50, 345-357.

Zhang, D.M., Huang, H.W., & Hicher, P.Y., 2004. Numerical prediction of long-term
settlements over tunnels in clay. Tunnelling and Underground Space Technology, 19(4
5), 379-386.

Zhang, X., Wang, M., Wang, Z., Li, J., Tong, J., & Liu, D., 2020a. A limit equilibrium
model for the reinforced face stability analysis of a shallow tunnel in cohesive-frictional
soils. Tunnelling and Underground Space Technology, 105, 103562.

Zhang, X., Wang, M., Wang, Z., Li, J., Zhao, S., Tong, J., & Liu, D., 2020b. Stability
analysis model for a tunnel face reinforced with bolts and an umbrella arch in cohesive-
frictional soils. Computers and Geotechnics, 124, p.103635.

128
 References

Zhang, Z., Shi, X., Wang, B., & Li, H., 2018. Stability of NATM tunnel faces in soft
surrounding rocks. Computers and Geotechnics, 96, 90-102.

Zhao, L., Li, D., Li, L., Yang, F., Cheng, X., & Luo, W., 2017. Three-dimensional
stability analysis of a longitudinally inclined shallow tunnel face. computers and
Geotechnics, 87, 32-48.

Zingg, S. 2016. Static effects and aspects of feasibility and design of drainages in
tunnelling. Doctoral dissertation, ETH Zurich.

Zingg, S. & Anagnostou, G., 2016. An investigation into efficient drainage layouts for
the stabilization of tunnel faces in homogeneous ground. Tunnelling and Underground
Space Technology, 58, 49-73.

Zingg, S. & Anagnostou, G., 2018. Tunnel face stability and the effectiveness of
advance drainage measures in water-bearing ground of non-uniform permeability. Rock
Mechanics and Rock Engineering, 51(1), 187-202.

Zingg, S., Bronzetti, D., & Anagnostou, G., 2013. Face stability improvement by
advance drainage via pilot tunnel. ITA-AITES World Tunnel Congress (WTC2013),
Geneva, Switzerland, 701-708.

Zou, J., Chen, G., & Qian, Z., 2019. Tunnel face stability in cohesion-frictional soils
considering the soil arching effect by improved failure models. Computers and
Geotechnics, 106, 1-17.

Zou, J. F., Qian, Z. H., Xiang, X. H., & Chen, G.H., 2019. Face stability of a tunnel
excavated in saturated nonhomogeneous soils. Tunnelling and Underground Space
Technology, 83, 1-17.

129
 Appendices

Appendices

Appendix A. MATLAB-code Limit Analysis mechanism with the free span and

forepole umbrella

Appendix B. MATLAB-code Limit Analysis mechanism considering the advance

drainage

Appendix C. FLAC3D-code Numerical Simulation considering the advance drainage

131
 Appendices

List of MATLAB archives

Appendix A
Airetri.m To calculate the area of a triangle
Input.m To input parameters and to show the results
Sigma_C.m To calculate the collapse pressure considering the free span

Appendix B
Airetri.m To calculate the area of the Facets 𝐹 , and 𝐹′ ,

ComputeMC.m To compute each term of the Limit Analysis work energy

equation
Createfigures.m To create mechanism figures in different perspectives
Generate.m To generate the failure mechanism through the ''point by point''
technique
SigmaC.m To import the water pressure distribution data through the
interpolation function and to show the results
SigMC.m To calculate the collapse pressure with the optimized geometrical
parameters
Trace.m To prepare the data to plot the mechanism figures
ValueMC.m To calculate variables of the mechanism independent to the
''point to point'' technique

List of FLAC3D archives

Appendix C
Modelo_V01.dat To compute the pore water pressure distribution under
steady-state conditions
Modelo_V02.dat To compute the collapse pressure by the bisection method
Intervalo.dat Initial face support pressure for the bisection method
Mixboundary.dat Mixed boundary condition for the tunnel face under
transient conditions

133
 Appendices

Appendix A MATLAB-code Limit Analysis mechanism with free

span considering the effect of forepole umbrella

Airetri.m
function [sec]=Airetri(a,b,c)
aa=max([a, b, c]);
cc=min([a, b, c]);
if aa~=b & cc~=b
bb=b;
elseif aa ~= a & cc~=a
bb=a;
else
bb=c;
end
if b + c ~= a
sec = 0.25 * ((aa + bb + cc) * (-aa + bb + cc) * (aa - bb + cc) * (aa + bb - cc)) ^ (1 / 2);
else
sec = 0;
end

135
 Appendices

Input.m
global D L Phi Coh Gam N M Sep
global Bh1 BhE RaE Al1 Al2
global curvaB_xy curvad3_xy Dy Dz Fy Fz Gy Gz Omg
global Ita No Mo Vo Delta Umb
D = 10; % Tunnel diameter [m].
L = 3; % Free span [m].
Phi = 35; % Friction angle [°]
Coh = 5; % Cohesion [kPa].
Gam = 20; % Unit weight [kN/m3].
N = 100; % Number of points in logarithmic spirals [-]
Sep = 0.02; % Discretization length in interface between Blocks s and 3 [m]
% Umbrella parameters
Ita = 5; % Angle of the umbrella with the horizontal [°]
No = 1538; % Axial force [kN/m]
Mo = 46; % Bending moment [kNm/m]
Vo = 622; % Shear force [kN/m]
Delta = 0.11; % Thickness of the continuous deformation zone [m]
Umb = 0; % 1-with umbrella; 0-without umbrella
% Variables
BhE = 20; % Coordinate of the center of rotation of Block 2 (BethaE) [°]
RaE = 10; % Coordinate of the center of rotation of Block 2 (RE) [m]
Al1 = 5; % Geometrical variable (Alpha1) [°]
Bh1 = 30; % Geometrical variable (Betha1) [°]
Al2 = 70; % Geometrical variable (Alpha2) [°]
if Umb == 1
Bh1 = 90 - Phi;
Al2 = 90;
end
Phi = Phi/180*pi;
Ita = Ita/180*pi;
Bh1 = Bh1/180*pi;
BhE = BhE/180*pi;
Al1 = Al1/180*pi;
136
 Appendices

Al2 = Al2/180*pi;
options = optimset('TolX',0.01,'Tolfun',0.01);
if Umb == 1
Params_Geom = [BhE,RaE,Al1];
[Params_Geom,sig,exitflag,output] = fminsearch(@Sigma_C,Params_Geom,options);
else
Params_Geom = [Bh1,BhE,RaE,Al1,Al2];
[Params_Geom,sig,exitflag,output] = fminsearch(@Sigma_C,Params_Geom,options);
end
Sigma = -Sigma_C(Params_Geom)

137
 Appendices

Sigma_C.m
function [sig] = Sigma_C(Parametros)
global D L Phi Coh Gam N M Sep
global Bh1 BhE RaE Al1 Al2
global curvaB_xy curvad3_xy Dy Dz Fy Fz Gy Gz Omg
global Ita No Mo Vo Delta Umb
if Umb == 1
BhE = Parametros(1);
RaE = Parametros(2);
Al1 = Parametros(3);
if RaE <= 0 || BhE <= 0 || BhE >= acos(D/2/RaE)
sig = 10000000;
return
end
if Al1 <= 0 || Al1 >= pi/2
sig = 20000000;
return
end
else
Bh1 = Parametros(1);
BhE = Parametros(2);
RaE = Parametros(3);
Al1 = Parametros(4);
Al2 = Parametros(5);
if Bh1 >= pi/2-Phi || Bh1 <= 0 || RaE <= 0 || BhE <= 0 || BhE >= acos(D/2/RaE)
sig = 10000000;
return
end
if Al1 <= 0 || Al2 <= 0 || Al1 >= pi/2 || Al2 >= pi/2
sig = 20000000;
return
end
end
Z_0 = -RaE*sin(BhE);
138
 Appendices

Y_0 = RaE*cos(BhE)-D/2;
RaA = (Z_0^2+Y_0^2)^0.5;
RaB = (Z_0^2+(Y_0+D)^2)^0.5;
BhA = atan(-Z_0/Y_0);
BhB = atan(-Z_0/(Y_0+D));
V1 = 1;
V3 = 1;
if Umb == 0
V3 = sin(pi/2-Bh1-Phi)/sin(Al2+Bh1+Phi)*V1;
V13 = cos(Al2)/sin(pi/2-Bh1-Phi)*V3;
End

% Computation of Point F
P_0 = [Z_0 Y_0];
P_i = [0 0];
r_P = norm(P_0);
r_M = RaB;
Omg = V3/r_P*sin(Al1+Al2-Phi)/sin(Al1-Phi+BhA);
curvad3_xy = zeros(3,2);
curvad3_xy(1:2,1) = P_i;
M = 1;
while r_M > r_P
r_P = P_i - P_0;
v_2 = [r_P(2) -r_P(1)]*Omg;
v_3 = [-cos(Al2) -sin(Al2)]*V3;
v_23 = v_2 - v_3;
u = [cos(Phi) sin(Phi) ; -sin(Phi) cos(Phi)]*transpose(v_23)/norm(v_23);
u = transpose(u);
P_i = P_i - u*Sep;
curvad3_xy(3,M) = norm(v_23);
r_P = norm(P_i-P_0);
b_P = atan((P_i(1)-Z_0)/(-P_i(2)+Y_0));
r_M = RaB*exp((BhB-b_P)*tan(Phi));
if b_P <= 0
139
 Appendices

sig = 31000000;
return
end
if u(1) >= 0
sig = 32000000;
return
end
M = M+1;
curvad3_xy(1:2,M) = P_i;
end
if b_P>Al2-5/180*pi
sig = 40000000;
return
end
er_F = 1;
Sep_i = 0;
Sep_s = Sep;
while er_F > 0.0001
Sep_c = (Sep_i+Sep_s)/2;
P_i = transpose(curvad3_xy(1:2,M-1)) - u*Sep_c;
r_P = norm(P_i-P_0);
b_P = atan((P_i(1)-Z_0)/(-P_i(2)+Y_0));
r_M = RaB*exp((BhB-b_P)*tan(Phi));
er_F = abs(r_P - r_M);
if r_P>r_M
Sep_s = Sep_c;
else
Sep_i = Sep_c;
end
end
curvad3_xy(1:2,M) = P_i;

% Logarithmic spiral from B

BhF = b_P;
140
 Appendices

RaF = r_P;
curvaB_rb = zeros(2,N);
curvaB_xy = zeros(2,N);
curvaB_rb(2,:) = linspace(BhF,BhB,N);
curvaB_rb(1,:) = RaB*exp((BhB-curvaB_rb(2,:))*tan(Phi));
curvaB_xy(1,:) = curvaB_rb(1,:).*sin(curvaB_rb(2,:))+Z_0;
curvaB_xy(2,:) = -curvaB_rb(1,:).*cos(curvaB_rb(2,:))+Y_0;

% Computation of Point G
Dz = -L*tan(pi/2-Phi)/(tan(pi/2-Phi)+tan(Bh1));
Dy = -Dz*tan(Bh1);
Fz = curvad3_xy(1,M);
Fy = curvad3_xy(2,M);
u1z = cos(Al2-Phi);
u1y = sin(Al2-Phi);
u2z = -cos(pi-Phi-Al2);
u2y = sin(pi-Phi-Al2);
Gy = (-u2z/u2y*Fy+Fz+u1z/u1y*Dy-Dz)/(u1z/u1y-u2z/u2y);
Gz = u1z/u1y*Gy-u1z/u1y*Dy+Dz;
if Gz<Dz
sig = 60000000;
return
end
if Fy >= tan(Ita)*Fz+tan(Ita)*2*L-0.01
sig = 60000000;
return
end
d5 = ((Gz-Dz)^2+(Gy-Dy)^2)^0.5;
d6 = ((Gz-Fz)^2+(Gy-Fy)^2)^0.5;
d1 = L*tan(pi/2-Phi)/(cos(Bh1)*(tan(pi/2-Phi)+tan(Bh1)));
d7 = L*tan(Bh1)/(cos(pi/2-Phi)*(tan(pi/2-Phi)+tan(Bh1)));
A1 = Airetri(d1,L,d7);

141
 Appendices

% Block 2
curvaT2_xy = zeros(2,N+M);
curvaT2_xy(1,:) = [curvad3_xy(1,:) curvaB_xy(1,2:N) curvad3_xy(1,1)];
curvaT2_xy(2,:) = [curvad3_xy(2,:) curvaB_xy(2,2:N) curvad3_xy(2,1)];
Area2 = polyarea(curvaT2_xy(1,:),curvaT2_xy(2,:));
A_cg2 = 0.0;
X_cg2 = 0.0;
Y_cg2 = 0.0;
for ii = 1:N+M-1
A_cg2 = A_cg2+(curvaT2_xy(1,ii)*curvaT2_xy(2,ii+1)-
curvaT2_xy(1,ii+1)*curvaT2_xy(2,ii));
X_cg2 =
X_cg2+(curvaT2_xy(1,ii)+curvaT2_xy(1,ii+1))*(curvaT2_xy(1,ii)*curvaT2_xy(2,ii+
1)-curvaT2_xy(1,ii+1)*curvaT2_xy(2,ii));
Y_cg2 = Y_cg2 +
(curvaT2_xy(2,ii)+curvaT2_xy(2,ii+1))*(curvaT2_xy(1,ii)*curvaT2_xy(2,ii+1)-
curvaT2_xy(1,ii+1)*curvaT2_xy(2,ii));
end
A_cg2 = 1/2*A_cg2;
X_cg2 = 1/6/A_cg2*X_cg2;
Y_cg2 = 1/6/A_cg2*Y_cg2;
Ra_cg2 = ((Y_0-Y_cg2)^2+(X_cg2-Z_0)^2)^0.5;
Bh_cg2 = atan((X_cg2-Z_0)/(Y_0-Y_cg2));

% Block 3
curvaT3_xy = zeros(2,M+3);
curvaT3_xy(1,:) = [curvad3_xy(1,:) Gz Dz 0];
curvaT3_xy(2,:) = [curvad3_xy(2,:) Gy Dy 0];
Area3 = polyarea(curvaT3_xy(1,:),curvaT3_xy(2,:));
A_cg3 = 0.0;
X_cg3 = 0.0;
Y_cg3 = 0.0;
for ii = 1:M+2

142
 Appendices

A_cg3 = A_cg3+(curvaT3_xy(1,ii)*curvaT3_xy(2,ii+1)-
curvaT3_xy(1,ii+1)*curvaT3_xy(2,ii));
X_cg3 =
X_cg3+(curvaT3_xy(1,ii)+curvaT3_xy(1,ii+1))*(curvaT3_xy(1,ii)*curvaT3_xy(2,ii+
1)-curvaT3_xy(1,ii+1)*curvaT3_xy(2,ii));
Y_cg3 = Y_cg3 +
(curvaT3_xy(2,ii)+curvaT3_xy(2,ii+1))*(curvaT3_xy(1,ii)*curvaT3_xy(2,ii+1)-
curvaT3_xy(1,ii+1)*curvaT3_xy(2,ii));
end
A_cg3 = 1/2*A_cg3;
X_cg3 = 1/6/A_cg3*X_cg3;
Y_cg3 = 1/6/A_cg3*Y_cg3;
Ra_cg3 = ((Y_0-Y_cg3)^2+(X_cg3-Z_0)^2)^0.5;
Bh_cg3 = atan((X_cg3-Z_0)/(Y_0-Y_cg3));

% Energy balance
EA_P1 = V1*Gam*A1;
EA_P2 = Omg*Ra_cg2*Gam*Area2*sin(Bh_cg2);
EA_P3 = V3*Gam*Area3*sin(Al2);
EA_P = EA_P1 + EA_P2 + EA_P3 ;
if Umb == 0
ED_d1 = V13*Coh*cos(Phi)*d1;
else
ED_d1 = 0;
end
ED_d5 = V3*Coh*cos(Phi)*d5;
ED_d6 = V3*Coh*cos(Phi)*d6;
ED_d7 = V1*Coh*cos(Phi)*d7;
ED_d2 = 0.0;
for ii = 2:N
ED_d2 = ED_d2 + curvaB_rb(1,ii)*norm(curvaB_xy(:,ii)-curvaB_xy(:,ii-1));
end
ED_d2 = Omg*Coh*cos(Phi)*ED_d2;
ED_d3 = 0.0;
143
 Appendices

for ii = 1:M-2
ED_d3 = ED_d3 + curvad3_xy(3,ii)*Sep;
end
ED_d3 = ED_d3 + curvad3_xy(3,M-1)*Sep_c;
ED_d3 = Coh*cos(Phi)*ED_d3;
ED = ED_d1 + ED_d2 + ED_d3 + ED_d5 + ED_d6 + ED_d7 ;
if Umb == 1
nu = 2*Mo/No/Delta;
ni = Vo/No;

% By moment
EA_S1 = -No*V1*(sin(Ita)*0+nu*cos(Ita));
EA_S3 = -No*V3*(cos(Al2-Ita)*0+nu*sin(Al2-Ita));

% By shear
% EA_S1 = -No*V1*(sin(Ita)^2*0+ni^2*cos(Ita)^2)^0.5;
% EA_S3 = -No*V3*(cos(Al2-Ita)^2*0+ni^2*sin(Al2-Ita)^2)^0.5;
EA_S = EA_S1 + EA_S3;
else
EA_S = 0;
End

% Collapse Pressure
Sigma = (EA_P + EA_S - ED) / (V1*L + RaE*Omg*D*cos(BhE));
sig = -Sigma;

144
 Appendices

Appendix B MATLAB-code Limit Analysis mechanism

considering the advance drainage

Airetri.m
function [sec] = Airetri(a,b,c)
aa = max([a, b, c]);
cc = min([a, b, c]);
if aa ~= b & cc ~= b
bb = b;
elseif aa ~= a & cc ~= a
bb = a;
else
bb = c;
end
if b + c ~= a
sec = 0.25 * ((aa + bb + cc) * (-aa + bb + cc) * (aa - bb + cc) * (aa + bb - cc)) ^ (1 / 2);
else
sec = 0;
end

145
 Appendices

ComputeMC.m
Wp = 0; % Coefficient of weight
Wr = 0; % Coefficient of energy dissipation
Wu = 0; % Coefficient of pore water pressure

global PoreWdistribution
Grad=0; % The gradient of the pressure applied on the tunnel face
for j = 1:j_A-1
if j/2 == round(j/2)
depinf = Nteta/2-j/2;
depsup = Nteta/2-j/2;
fininf = Nteta/2+j/2-1;
finsup = Nteta/2+j/2;
else
depinf = Nteta/2-(j-1)/2;
depsup = Nteta/2-(j-1)/2-1;
fininf = Nteta/2+(j+1)/2-1;
finsup = Nteta/2+(j+1)/2-1;
end

for i = depinf:fininf
if j/2 == round(j/2)
x1 = Points(j,i,1);
x2 = Points(j,i+1,1);
x3 = Points(j+1,i+1,1);
y1 = Points(j,i,2);
y2 = Points(j,i+1,2);
y3 = Points(j+1,i+1,2);
z1 = Points(j,i,3);
z2 = Points(j,i+1,3);
z3 = Points(j+1,i+1,3);
else
x1 = Points(j,i,1);
x2 = Points(j,i+1,1);
146
 Appendices

x3 = Points(j+1,i,1);
y1 = Points(j,i,2);
y2 = Points(j,i+1,2);
y3 = Points(j+1,i,2);
z1 = Points(j,i,3);
z2 = Points(j,i+1,3);
z3 = Points(j+1,i,3);
end

% coefficients of force work rate

xc = (x1+x2+x3)/3;
yc = (y1+y2+y3)/3;
zc = (z1+z2+z3)/3; % The coordinates of Facet centre
pwp = PoreWdistribution(abs(xc),yc,zc);
r = sqrt((Yo-yc)^2+(Zo-zc)^2);
bet = atan(-(zc-Zo)/(yc-Yo));
a = sqrt((y1-y2)^2+(z1-z2)^2);
b = sqrt((y1-y3)^2+(z1-z3)^2);
c = sqrt((y3-y2)^2+(z3-z2)^2);
Wp = Wp+r*sin(bet)*Airetri(a,b,c)*abs(xc)*Gam;
a = sqrt((x1-x2)^2+(y1-y2)^2+(z1-z2)^2);
b = sqrt((x1-x3)^2+(y1-y3)^2+(z1-z3)^2);
c = sqrt((x3-x2)^2+(y3-y2)^2+(z3-z2)^2);
Wr = Wr+cos(Phi)*r*Airetri(a,b,c)*Coh;
Wu = Wu+sin(Phi)*r*Airetri(a,b,c)*pwp;
end
for i = depsup:finsup
if j/2 == round(j/2)
x1 = Points(j+1,i,1);
x2 = Points(j+1,i+1,1);
x3 = Points(j,i,1);
y1 = Points(j+1,i,2);
y2 = Points(j+1,i+1,2);
y3 = Points(j,i,2);
147
 Appendices

z1 = Points(j+1,i,3);
z2 = Points(j+1,i+1,3);
z3 = Points(j,i,3);
else
x1 = Points(j+1,i,1);
x2 = Points(j+1,i+1,1);
x3 = Points(j,i+1,1);
y1 = Points(j+1,i,2);
y2 = Points(j+1,i+1,2);
y3 = Points(j,i+1,2);
z1 = Points(j+1,i,3);
z2 = Points(j+1,i+1,3);
z3 = Points(j,i+1,3);
end
xc = (x1+x2+x3)/3;
yc = (y1+y2+y3)/3;
zc = (z1+z2+z3)/3;
pwp = PoreWdistribution(abs(xc),yc,zc);
r = sqrt((Yo-yc)^2+(Zo-zc)^2);
bet = atan(-(zc-Zo)/(yc-Yo));
a = sqrt((y1-y2)^2+(z1-z2)^2);
b = sqrt((y1-y3)^2+(z1-z3)^2);
c = sqrt((y3-y2)^2+(z3-z2)^2);
Wp = Wp+r*sin(bet)*Airetri(a,b,c)*abs(xc)*Gam;
a = sqrt((x1-x2)^2+(y1-y2)^2+(z1-z2)^2);
b = sqrt((x1-x3)^2+(y1-y3)^2+(z1-z3)^2);
c = sqrt((x3-x2)^2+(y3-y2)^2+(z3-z2)^2);
Wr = Wr+cos(Phi)*r*Airetri(a,b,c)*Coh;
Wu = Wu+sin(Phi)*r*Airetri(a,b,c)*pwp;
end
end
Points(j_A:size(Points,1),3*Nteta/4+1,:) = Points(j_A:size(Points,1),Nteta/4,:);
for j = j_A:size(Points,1)-1
dep = Nteta/4;
148
 Appendices

fin = 3*Nteta/4;
for i = dep:fin
if j/2 == round(j/2)
x1 = Points(j,i,1);
x2 = Points(j,i+1,1);
x3 = Points(j+1,i+1,1);
y1 = Points(j,i,2);
y2 = Points(j,i+1,2);
y3 = Points(j+1,i+1,2);
z1 = Points(j,i,3);
z2 = Points(j,i+1,3);
z3 = Points(j+1,i+1,3);
else
x1 = Points(j,i,1);
x2 = Points(j,i+1,1);
x3 = Points(j+1,i,1);
y1 = Points(j,i,2);
y2 = Points(j,i+1,2);
y3 = Points(j+1,i,2);
z1 = Points(j,i,3);
z2 = Points(j,i+1,3);
z3 = Points(j+1,i,3);
end
xc = (x1+x2+x3)/3;
yc = (y1+y2+y3)/3;
zc = (z1+z2+z3)/3;
pwp = PoreWdistribution(abs(xc),yc,zc);
r = sqrt((Yo-yc)^2+(Zo-zc)^2);
bet = atan(-(zc-Zo)/(yc-Yo));
if bet<0
bet = bet+pi;
end
a = sqrt((y1-y2)^2+(z1-z2)^2);
b = sqrt((y1-y3)^2+(z1-z3)^2);
149
 Appendices

c = sqrt((y3-y2)^2+(z3-z2)^2);
Wp = Wp+r*sin(bet)*Airetri(a,b,c)*abs(xc)*Gam;
a = sqrt((x1-x2)^2+(y1-y2)^2+(z1-z2)^2);
b = sqrt((x1-x3)^2+(y1-y3)^2+(z1-z3)^2);
c = sqrt((x3-x2)^2+(y3-y2)^2+(z3-z2)^2);
Wr = Wr+cos(Phi)*r*Airetri(a,b,c)*Coh;
Wu = Wu+sin(Phi)*r*Airetri(a,b,c)*pwp;
if j/2 == round(j/2)
x1 = Points(j+1,i,1);
x2 = Points(j+1,i+1,1);
x3 = Points(j,i,1);
y1 = Points(j+1,i,2);
y2 = Points(j+1,i+1,2);
y3 = Points(j,i,2);
z1 = Points(j+1,i,3);
z2 = Points(j+1,i+1,3);
z3 = Points(j,i,3);
else
x1 = Points(j+1,i,1);
x2 = Points(j+1,i+1,1);
x3 = Points(j,i+1,1);
y1 = Points(j+1,i,2);
y2 = Points(j+1,i+1,2);
y3 = Points(j,i+1,2);
z1 = Points(j+1,i,3);
z2 = Points(j+1,i+1,3);
z3 = Points(j,i+1,3);
end
xc = (x1+x2+x3)/3;
yc = (y1+y2+y3)/3;
zc = (z1+z2+z3)/3;
pwp = PoreWdistribution(abs(xc),yc,zc);
r = sqrt((Yo-yc)^2+(Zo-zc)^2);
bet = atan(-(zc-Zo)/(yc-Yo));
150
 Appendices

if bet<0
bet = bet+pi;
end
a = sqrt((y1-y2)^2+(z1-z2)^2);
b = sqrt((y1-y3)^2+(z1-z3)^2);
c = sqrt((y3-y2)^2+(z3-z2)^2);
Wp = Wp+r*sin(bet)*Airetri(a,b,c)*abs(xc)*Gam;
a = sqrt((x1-x2)^2+(y1-y2)^2+(z1-z2)^2);
b = sqrt((x1-x3)^2+(y1-y3)^2+(z1-z3)^2);
c = sqrt((x3-x2)^2+(y3-y2)^2+(z3-z2)^2);
Wr = Wr+cos(Phi)*r*Airetri(a,b,c)*Coh;
Wu = Wu+sin(Phi)*r*Airetri(a,b,c)*pwp;
end
end
Wt = 0;
for i = 1:Nteta-1
x1 = Contour(i,1);
y1 = Contour(i,2);
x2 = Contour(i+1,1);
y2 = Contour(i+1,2);
h = (y1+y2)/2;
r = sqrt((Yo-h)^2+Zo^2);
bet = atan(Zo/(h-Yo));
gr = 1+Grad*(h+Diam/2);
Wt = Wt+r*cos(bet)*gr*abs(x1+x2)/2*abs(y1-y2);
end
Wp = real(Wp);
Wr = real(Wr);
Wu = real(Wu);
WpG = Wp;

151
 Appendices

Createfigures.m
function Createfigures(xdata1, ydata1, zdata1)

% Create figure
figure1 = figure(...
'Color',[1 1 1],...
'PaperPosition',[0.6345 6.345 20.3 15.23],...
'PaperSize',[20.98 29.68],...
'PaperType','a4letter');
colormap gray

% Create axes
axes1 = axes(...
'CameraPosition',[0 50 -3],...
'CameraUpVector',[0 0 1],...
'PlotBoxAspectRatio',[2.5 1.5 3.5],...
'Position',[0.6223 0.03704 0.3763 0.9467],...
'Parent',figure1);
axis(axes1,[-5 5 0 6 -10 4]);
grid(axes1,'on');
hold(axes1,'all');

% Create surf1
surf1 = surf(xdata1,...
ydata1,zdata1,...
'FaceColor','interp',...
'LineStyle','none',...
'BackFaceLighting','lit',...
'AmbientStrength',0.35,...
'DiffuseStrength',0.85,...
'SpecularStrength',0.5,...
'SpecularExponent',15,...
'Parent',axes1,...
'ZDataSource','Z');
152
 Appendices

light1 = light('Position',[6 -2 2],'Parent',axes1);

axes2 = axes(...
'CameraPosition',[-120 2 -3],...
'CameraTarget',[0 2 -3],...
'CameraUpVector',[0 0 1],...
'CameraViewAngle',6.677,...
'PlotBoxAspectRatio',[2.5 1.5 3.5],...
'Position',[0.2818 -0.03 0.4678 1.082],...
'Parent',figure1);
axis(axes2,[-5 5 0 6 -10 4]);
grid(axes2,'on');
hold(axes2,'all');

% Create surf2
surf2 = surf(xdata1,...
ydata1,zdata1,...
'FaceColor','interp',...
'LineStyle','none',...
'BackFaceLighting','lit',...
'AmbientStrength',0.35,...
'DiffuseStrength',0.85,...
'SpecularStrength',0.5,...
'SpecularExponent',15,...
'Parent',axes2);
light2 = light('Position',[-12 6 -2],'Parent',axes2);
axes3 = axes(...
'CameraPosition',[-60 50 10],...
'CameraTarget',[0 5 -3],...
'PlotBoxAspectRatio',[2.5 1 3.5],...
'Position',[-0.0421 0.0163 0.3873 1.027],...
'Parent',figure1);
axis(axes3,[-5 5 0 6 -10 4]);
hold(axes3,'all');

153
 Appendices

% Create surf3
surf3 = surf(xdata1,...
ydata1,zdata1,...
'FaceColor','interp',...
'LineStyle','none',...
'BackFaceLighting','lit',...
'AmbientStrength',0.35,...
'DiffuseStrength',0.85,...
'SpecularExponent',15,...
'SpecularColorReflectance',0.5,...
'Parent',axes3,...
'ZDataSource','Z');
light3 = light('Position',[-6 8 -8],'Parent',axes3);

154
 Appendices

Generate.m
Signe = -1;
beta = Plans(1,7);% Angle Beta of the first plane
aax = 0;
aay = cos(Plans(1,7));
aaz = -sin(Plans(1,7));% Unit vector of the first plane from plane center to the rotation
center
xv = 0;
yv = Signe*sin(beta);
zv = Signe*cos(beta); % The coordinates of the velocity vector of the first plane
x1 = Points(1,Nteta/2,1);
y1 = Points(1,Nteta/2,2);
z1 = Points(1,Nteta/2,3);
Teta1 = Points(1,Nteta/2,4);
x2 = Points(1,Nteta/2+1,1);
y2 = Points(1,Nteta/2+1,2);
z2 = Points(1,Nteta/2+1,3);
Teta2 = Points(1,Nteta/2+1,4); % The coordinates and angle of points in Facet
xprime = Xc;
yprime = Yc-Diam/2;
zprime = 0;
Teta = (Teta1+Teta2)/2;
ax = -1;
ay = 0;
az = 0;
xc = Plans(2,3);
yc = Plans(2,4);
zc = Plans(2,5);
bx = xc-xprime;
by = yc-yprime;
bz = zc-zprime;
A = -zv/yv;
B = cos(pi/2+Phi)/yv;
C = -A*ay/ax-az/ax;
155
 Appendices

D = (-B*ay)/ax;
E = 1+C^2+A^2;
F = 2*C*D+2*A*B;
G = D^2+B^2-1;
Delta = F^2-4*E*G;
zn = (-F+sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
deltay = (zn*ax-xn*az);
deltaz = (xn*ay-yn*ax);
if Signe*(yv*deltay+zv*deltaz)>0
zn = (-F-sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
end
deltax = sin(Teta);
deltay = cos(Teta)*cos(Plans(2,7));
deltaz = -cos(Teta)*sin(Plans(2,7));
r = -(xn*bx+yn*by+zn*bz)/(deltax*xn+deltay*yn+deltaz*zn);
Points(2,Nteta/2,1) = xc+r*deltax;
Points(2,Nteta/2,2) = yc+r*deltay;
Points(2,Nteta/2,3) = zc+r*deltaz;
Points(2,Nteta/2,4) = Teta;
Points(2,Nteta/2,5) = deltax;
Points(2,Nteta/2,6) = deltay;
Points(2,Nteta/2,7) = deltaz;
Points(2,Nteta/2,8) = r;
Points(2,Nteta/2-1,1) = Contour(Nteta/2-1,1);
Points(2,Nteta/2-1,2) = Contour(Nteta/2-1,2);
bbx = Points(2,Nteta/2-1,1)-Plans(2,3);
bby = Points(2,Nteta/2-1,2)-Plans(2,4);
bbz = Points(2,Nteta/2-1,3)-Plans(2,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(2,Nteta/2-1,4) = acos((aax*bbx+aay*bby+aaz*bbz)/bb);
156
 Appendices

Points(2,Nteta/2-1,8) = sqrt((Points(2,Nteta/2-1,3)-Zo)^2+(Points(2,Nteta/2-1,2)-
Yo)^2);
Points(2,Nteta/2+1,1) = Contour(Nteta/2+2,1);
Points(2,Nteta/2+1,2) = Contour(Nteta/2+2,2);
bbx = Points(2,Nteta/2+1,1)-Plans(2,3);
bby = Points(2,Nteta/2+1,2)-Plans(2,4);
bbz = Points(2,Nteta/2+1,3)-Plans(2,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(2,Nteta/2+1,4) = 2*pi-acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(2,Nteta/2+1,8) = sqrt((Points(2,Nteta/2+1,3)-Zo)^2+(Points(2,Nteta/2+1,2)-
Yo)^2);
beta = Plans(2,7);
aax = 0;
aay = cos(Plans(2,7));
aaz = -sin(Plans(2,7));
xv = 0;
yv = Signe*sin(beta);
zv = Signe*cos(beta);
x1 = Points(2,Nteta/2-1,1);
y1 = Points(2,Nteta/2-1,2);
z1 = Points(2,Nteta/2-1,3);
Teta1 = Points(2,Nteta/2-1,4);
x2 = Points(2,Nteta/2,1);
y2 = Points(2,Nteta/2,2);
z2 = Points(2,Nteta/2,3);
Teta2 = Points(2,Nteta/2,4);
x3 = Points(2,Nteta/2+1,1);
y3 = Points(2,Nteta/2+1,2);
z3 = Points(2,Nteta/2+1,3);
Teta3 = Points(2,Nteta/2+1,4);
Coefs = [Teta1^2,Teta1,1;Teta2^2,Teta2,1;Teta3^2,Teta3,1]^-1*[sqrt((xc-x1)^2+(yc-
y1)^2+(zc-z1)^2);sqrt((xc-x2)^2+(yc-y2)^2+(zc-z2)^2);sqrt((xc-x3)^2+(yc-
y3)^2+(zc-z3)^2)];
xc = Plans(2,3);
157
 Appendices

yc = Plans(2,4);
zc = Plans(2,5);
Teta = (Teta1+Teta2)/2;
Rprime = Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3);
xprime = xc+Rprime*sin(Teta);
yprime = yc+Rprime*cos(Teta)*cos(Plans(3,7));
zprime = zc-Rprime*cos(Teta)*sin(Plans(3,7));
deriv = ((2*Coefs(1)*Teta+Coefs(2))*cos(Teta)-
(Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3))*sin(Teta))/(-
(2*Coefs(1)*Teta+Coefs(2))*sin(Teta)-
(Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3))*cos(Teta));
ax = -1;
ay = -cos(Plans(2,7))*deriv;
az = sin(Plans(2,7))*deriv;
xc = Plans(3,3);
yc = Plans(3,4);
zc = Plans(3,5);
bx = xc-xprime;
by = yc-yprime;
bz = zc-zprime;
A = -zv/yv;
B = cos(pi/2+Phi)/yv;
C = -A*ay/ax-az/ax;
D = (-B*ay)/ax;
E = 1+C^2+A^2;
F = 2*C*D+2*A*B;
G = D^2+B^2-1;
Delta = F^2-4*E*G;
zn = (-F+sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
deltay = (zn*ax-xn*az);
deltaz = (xn*ay-yn*ax);
if Signe*(yv*deltay+zv*deltaz)>0
158
 Appendices

zn = (-F-sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
end
deltax = sin(Teta);
deltay = cos(Teta)*cos(Plans(3,7));
deltaz = -cos(Teta)*sin(Plans(3,7));
r = -(xn*bx+yn*by+zn*bz)/(deltax*xn+deltay*yn+deltaz*zn);
Points(3,Nteta/2,1) = xc+r*deltax;
Points(3,Nteta/2,2) = yc+r*deltay;
Points(3,Nteta/2,3) = zc+r*deltaz;
Points(3,Nteta/2,4) = Teta;
Points(3,Nteta/2,5) = deltax;
Points(3,Nteta/2,6) = deltay;
Points(3,Nteta/2,7) = deltaz;
Points(3,Nteta/2,8) = r;
xc = Plans(2,3);
yc = Plans(2,4);
zc = Plans(2,5);
Teta = (Teta2+Teta3)/2;
Rprime = Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3);
xprime = xc+Rprime*sin(Teta);
yprime = yc+Rprime*cos(Teta)*cos(Plans(3,7));
zprime = zc-Rprime*cos(Teta)*sin(Plans(3,7));
deriv = ((2*Coefs(1)*Teta+Coefs(2))*cos(Teta)-
(Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3))*sin(Teta))/(-
(2*Coefs(1)*Teta+Coefs(2))*sin(Teta)-
(Coefs(1)*Teta^2+Coefs(2)*Teta+Coefs(3))*cos(Teta));
ax = -1;
ay = -cos(Plans(2,7))*deriv;
az = sin(Plans(2,7))*deriv;
xc = Plans(3,3);
yc = Plans(3,4);
zc = Plans(3,5);
159
 Appendices

bx = xc-xprime;
by = yc-yprime;
bz = zc-zprime;
A = -zv/yv;
B = cos(pi/2+Phi)/yv;
C = -A*ay/ax-az/ax;
D = (-B*ay)/ax;
E = 1+C^2+A^2;
F = 2*C*D+2*A*B;
G = D^2+B^2-1;
Delta = F^2-4*E*G;
zn = (-F+sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
deltay = (zn*ax-xn*az);
deltaz = (xn*ay-yn*ax);
if Signe*(yv*deltay+zv*deltaz)>0
zn = (-F-sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
end
deltax = sin(Teta);
deltay = cos(Teta)*cos(Plans(3,7));
deltaz = -cos(Teta)*sin(Plans(3,7));
r = -(xn*bx+yn*by+zn*bz)/(deltax*xn+deltay*yn+deltaz*zn);
Points(3,Nteta/2+1,1) = xc+r*deltax;
Points(3,Nteta/2+1,2) = yc+r*deltay;
Points(3,Nteta/2+1,3) = zc+r*deltaz;
Points(3,Nteta/2+1,4) = Teta;
Points(3,Nteta/2+1,5) = deltax;
Points(3,Nteta/2+1,6) = deltay;
Points(3,Nteta/2+1,7) = deltaz;
Points(3,Nteta/2+1,8) = r;
Points(3,Nteta/2-1,1) = Contour(Nteta/2-2,1);
160
 Appendices

Points(3,Nteta/2-1,2) = Contour(Nteta/2-2,2);
bbx = Points(3,Nteta/2-1,1)-Plans(3,3);
bby = Points(3,Nteta/2-1,2)-Plans(3,4);
bbz = Points(3,Nteta/2-1,3)-Plans(3,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(3,Nteta/2-1,4) = acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(3,Nteta/2-1,8) = sqrt((Points(3,Nteta/2-1,3)-Zo)^2+(Points(3,Nteta/2-1,2)-
Yo)^2);
Points(3,Nteta/2+2,1) = Contour(Nteta/2+3,1);
Points(3,Nteta/2+2,2) = Contour(Nteta/2+3,2);
bbx = Points(3,Nteta/2+2,1)-Plans(3,3);
bby = Points(3,Nteta/2+2,2)-Plans(3,4);
bbz = Points(3,Nteta/2+2,3)-Plans(3,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(3,Nteta/2+2,4) = 2*pi-acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(3,Nteta/2+2,8) = sqrt((Points(3,Nteta/2+2,3)-Zo)^2+(Points(3,Nteta/2+2,2)-
Yo)^2);
for j = 3:j_A-1
beta = Plans(j,7);
sinbeta = sin(beta);
cosbeta = cos(beta);
betap = Plans(j+1,7);
sinbetap = sin(Plans(j+1,7));
cosbetap = cos(Plans(j+1,7));
xv = 0;
yv = Signe*sinbeta;
zv = Signe*cosbeta;
aax = 0;
aay = cosbeta;
aaz = -sinbeta;
if j/2 == round(j/2)
dep = Nteta/2-j/2;
fin = Nteta/2+(j-2)/2;
else
161
 Appendices

dep = Nteta/2-(j-1)/2;
fin = Nteta/2+(j-1)/2;
end
for i = dep:fin+1
bbx = Points(j,i,1)-xc;
bby = Points(j,i,2)-yc;
bbz = Points(j,i,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(j,i,4) = real(acos((aax*bbx+aay*bby+aaz*bbz)/bb));
if Points(j,i,1)<0
Points(j,i,4) = 2*pi-Points(j,i,4);
end
end
A = -zv/yv;
for i = dep:fin
xc = Plans(j,3);
yc = Plans(j,4);
zc = Plans(j,5);
if i == dep
x1 = Points(j,i,1);
x2 = Points(j,i+1,1);
x3 = Points(j,i+2,1);
x4 = Points(j,i+3,1);
y1 = Points(j,i,2);
y2 = Points(j,i+1,2);
y3 = Points(j,i+2,2);
y4 = Points(j,i+3,2);
z1 = Points(j,i,3);
z2 = Points(j,i+1,3);
z3 = Points(j,i+2,3);
z4 = Points(j,i+3,3);
Teta1 = Points(j,i,4);
Teta2 = Points(j,i+1,4);
Teta3 = Points(j,i+2,4);
162
 Appendices

Teta4 = Points(j,i+3,4);
xprime = (x1+x2)/2;
yprime = (y1+y2)/2;
zprime = (z1+z2)/2;
ax = x2-x1;
ay = y2-y1;
az = z2-z1;
Teta = (Teta2+Teta1)/2;
costeta = cos(Teta);
sinteta = sin(Teta);
elseif i == fin
x1 = Points(j,i-2,1);
x2 = Points(j,i-1,1);
x3 = Points(j,i,1);
x4 = Points(j,i+1,1);
y1 = Points(j,i-2,2);
y2 = Points(j,i-1,2);
y3 = Points(j,i,2);
y4 = Points(j,i+1,2);
z1 = Points(j,i-2,3);
z2 = Points(j,i-1,3);
z3 = Points(j,i,3);
z4 = Points(j,i+1,3);
Teta1 = Points(j,i-2,4);
Teta2 = Points(j,i-1,4);
Teta3 = Points(j,i,4);
Teta4 = Points(j,i+1,4);
xprime = (x3+x4)/2;
yprime = (y3+y4)/2;
zprime = (z3+z4)/2;
ax = x4-x3;
ay = y4-y3;
az = z4-z3;
Teta = (Teta4+Teta3)/2;
163
 Appendices

costeta = cos(Teta);
sinteta = sin(Teta);
else
x1 = Points(j,i-1,1);
x2 = Points(j,i,1);
x3 = Points(j,i+1,1);
x4 = Points(j,i+2,1);
y1 = Points(j,i-1,2);
y2 = Points(j,i,2);
y3 = Points(j,i+1,2);
y4 = Points(j,i+2,2);
z1 = Points(j,i-1,3);
z2 = Points(j,i,3);
z3 = Points(j,i+1,3);
z4 = Points(j,i+2,3);
Teta1 = Points(j,i-1,4);
Teta2 = Points(j,i,4);
Teta3 = Points(j,i+1,4);
Teta4 = Points(j,i+2,4);
xprime = (x3+x2)/2;
yprime = (y3+y2)/2;
zprime = (z3+z2)/2;
ax = x3-x2;
ay = y3-y2;
az = z3-z2;
Teta = (Teta2+Teta3)/2;
costeta = cos(Teta);
sinteta = sin(Teta);
end
xc = 0;
yc = Plans(j+1,4);
zc = Plans(j+1,5);
bx = xc-xprime;
by = yc-yprime;
164
 Appendices

bz = zc-zprime;
B = cos(pi/2+Phi)/yv;
C = -A*ay/ax-az/ax;
D = (-B*ay)/ax;
E = 1+C^2+A^2;
F = 2*C*D+2*A*B;
G = D^2+B^2-1;
Delta = F^2-4*E*G;
zn = (-F+sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
deltay = (zn*ax-xn*az);
deltaz = (xn*ay-yn*ax);
if Signe*(yv*deltay+zv*deltaz)>0
zn = (-F-sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
end
deltax = sinteta;
deltay = costeta*cosbetap;
deltaz = -costeta*sinbetap;
r = -(xn*bx+yn*by+zn*bz)/(deltax*xn+deltay*yn+deltaz*zn);
if j/2 == round(j/2)
Points(j+1,i+1,1) = xc+r*deltax;
Points(j+1,i+1,2) = yc+r*deltay;
Points(j+1,i+1,3) = zc+r*deltaz;
Points(j+1,i+1,4) = Teta;
Points(j+1,i+1,5) = deltax;
Points(j+1,i+1,6) = deltay;
Points(j+1,i+1,7) = deltaz;
Points(j+1,i+1,8) = r;
else
Points(j+1,i,1) = xc+r*deltax;
Points(j+1,i,2) = yc+r*deltay;
165
 Appendices

Points(j+1,i,3) = zc+r*deltaz;
Points(j+1,i,4) = Teta;
Points(j+1,i,5) = xprime;
Points(j+1,i,6) = yprime;
Points(j+1,i,7) = zprime;
Points(j+1,i,8) = r;
end
end
if j/2 == round(j/2)
Points(j+1,Nteta/2-j/2,1) = Contour(Nteta/2-j,1);
Points(j+1,Nteta/2-j/2,2) = Contour(Nteta/2-j,2);
bbx = Points(j+1,Nteta/2-j/2,1)-xc;
bby = Points(j+1,Nteta/2-j/2,2)-yc;
bbz = Points(j+1,Nteta/2-j/2,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(j+1,Nteta/2-j/2,4) = acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(j+1,Nteta/2-j/2,8) = sqrt((Points(j+1,Nteta/2-j/2,3)-
Zo)^2+(Points(j+1,Nteta/2-j/2,2)-Yo)^2);
Points(j+1,Nteta/2+j/2+1,1) = Contour(Nteta/2+j+1,1);
Points(j+1,Nteta/2+j/2+1,2) = Contour(Nteta/2+j+1,2);
bbx = Points(j+1,Nteta/2+j/2+1,1)-xc;
bby = Points(j+1,Nteta/2+j/2+1,2)-yc;
bbz = Points(j+1,Nteta/2+j/2+1,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(j+1,Nteta/2+j/2+1,4) = 2*pi-acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(j+1,Nteta/2+j/2+1,8) = sqrt((Points(j+1,Nteta/2+j/2+1,3)-
Zo)^2+(Points(j+1,Nteta/2+j/2+1,2)-Yo)^2);
else
Points(j+1,Nteta/2-(j+1)/2,1) = Contour(Nteta/2-j,1);
Points(j+1,Nteta/2-(j+1)/2,2) = Contour(Nteta/2-j,2);
bbx = Points(j+1,Nteta/2-(j+1)/2,1)-xc;
bby = Points(j+1,Nteta/2-(j+1)/2,2)-yc;
bbz = Points(j+1,Nteta/2-(j+1)/2,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
166
 Appendices

Points(j+1,Nteta/2-(j+1)/2,4) = acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(j+1,Nteta/2-(j+1)/2,8) = sqrt((Points(j+1,Nteta/2-(j+1)/2,3)-Zo)^2+(Points(j+
1,Nteta/2-(j+1)/2,2)-Yo)^2);
Points(j+1,Nteta/2+(j+1)/2,1) = Contour(Nteta/2+j+1,1);
Points(j+1,Nteta/2+(j+1)/2,2) = Contour(Nteta/2+j+1,2);
bbx = Points(j+1,Nteta/2+(j+1)/2,1)-xc;
bby = Points(j+1,Nteta/2+(j+1)/2,2)-yc;
bbz = Points(j+1,Nteta/2+(j+1)/2,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(j+1,Nteta/2+(j+1)/2,4) = 2*pi-acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(j+1,Nteta/2+(j+1)/2,8) = sqrt((Points(j+1,Nteta/2+(j+1)/2,3)-
Zo)^2+(Points(j+ 1,Nteta/2+(j+1)/2,2)-Yo)^2);
end
end
summ = 10^10;
for j = jA:jF-1
beta = Plans(j,7);
sinbeta = sin(beta);
cosbeta = cos(beta);
betap = Plans(j+1,7);
sinbetap = sin(Plans(j+1,7));
cosbetap = cos(Plans(j+1,7));
xv = 0;
yv = Signe*sinbeta;
zv = Signe*cosbeta;
aax = 0;
aay = cosbeta;
aaz = -sinbeta;
dep = Nteta/4;
fin = 3*Nteta/4;
Trapezes = zeros(1,8);
for i = dep:fin
X1 = Points(j,i,1);
Y1 = Points(j,i,2);
167
 Appendices

Z1 = Points(j,i,3);
if i<fin
X2 = Points(j,i+1,1);
Y2 = Points(j,i+1,2);
Z2 = Points(j,i+1,3);
else
X2 = Points(j,dep,1);
Y2 = Points(j,dep,2);
Z2 = Points(j,dep,3);
end
if abs(X2)>abs(X1)
X3 = X1;Y3 = Y1;Z3 = Z1;
X1 = X2;Y1 = Y2;Z1 = Z2;
X2 = X3;Y2 = Y3;Z2 = Z3;
end
Trapezes(size(Trapezes,1)+1,1) = abs(X2)*sqrt((Y1-Y2)^2+(Z1-Z2)^2);
Trapezes(size(Trapezes,1),2) = X2/2;
Trapezes(size(Trapezes,1),3) = Y1+(Y2-Y1)/2;
Trapezes(size(Trapezes,1),4) = Z1+(Z2-Z1)/2;
Trapezes(size(Trapezes,1),5) = abs(X1-X2)*sqrt((Y1-Y2)^2+(Z1-Z2)^2)/2;
Trapezes(size(Trapezes,1),6) = X2+(X1-X2)/3;
Trapezes(size(Trapezes,1),7) = Y1+(Y2-Y1)/3;
Trapezes(size(Trapezes,1),8) = Z1+(Z2-Z1)/3;
end
suma = 0;
for i = 2:size(Trapezes,1)
suma = suma+Trapezes(i,1)+Trapezes(i,5);
end
xc = 0;
yc = Plans(j,4);
zc = Plans(j,5);
for i = dep:fin
bbx = Points(j,i,1)-xc;
bby = Points(j,i,2)-yc;
168
 Appendices

bbz = Points(j,i,3)-zc;
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(j,i,4) = real(acos((aax*bbx+aay*bby+aaz*bbz)/bb));
if Points(j,i,1)<0 & i ~= dep
Points(j,i,4) = 2*pi-Points(j,i,4);
end
end
for i = dep:fin
if j/2 == round(j/2)
if i == dep
x1 = Points(j,fin-1,1);
x2 = Points(j,fin,1);
x3 = Points(j,i,1);
x4 = Points(j,i+1,1);
y1 = Points(j,fin-1,2);
y2 = Points(j,fin,2);
y3 = Points(j,i,2);
y4 = Points(j,i+1,2);
z1 = Points(j,fin-1,3);
z2 = Points(j,fin,3);
z3 = Points(j,i,3);
z4 = Points(j,i+1,3);
Teta1 = Points(j,fin-1,4)-2*pi;
Teta2 = Points(j,fin,4)-2*pi;
Teta3 = Points(j,i,4);
Teta4 = Points(j,i+1,4);
elseif i == dep+1
x1 = Points(j,fin,1);
x2 = Points(j,i-1,1);
x3 = Points(j,i,1);
x4 = Points(j,i+1,1);
y1 = Points(j,fin,2);
y2 = Points(j,i-1,2);
y3 = Points(j,i,2);
169
 Appendices

y4 = Points(j,i+1,2);
z1 = Points(j,fin,3);
z2 = Points(j,i-1,3);
z3 = Points(j,i,3);
z4 = Points(j,i+1,3);
Teta1 = Points(j,fin,4)-2*pi;
Teta2 = Points(j,i-1,4);
Teta3 = Points(j,i,4);
Teta4 = Points(j,i+1,4);
elseif i == fin
x1 = Points(j,i-2,1);
x2 = Points(j,i-1,1);
x3 = Points(j,i,1);
x4 = Points(j,dep,1);
y1 = Points(j,i-2,2);
y2 = Points(j,i-1,2);
y3 = Points(j,i,2);
y4 = Points(j,dep,2);
z1 = Points(j,i-2,3);
z2 = Points(j,i-1,3);
z3 = Points(j,i,3);
z4 = Points(j,dep,3);
Teta1 = Points(j,i-2,4);
Teta2 = Points(j,i-1,4);
Teta3 = Points(j,i,4);
Teta4 = Points(j,dep,4)+2*pi;
else
x1 = Points(j,i-2,1);
x2 = Points(j,i-1,1);
x3 = Points(j,i,1);
x4 = Points(j,i+1,1);
y1 = Points(j,i-2,2);
y2 = Points(j,i-1,2);
y3 = Points(j,i,2);
170
 Appendices

y4 = Points(j,i+1,2);
z1 = Points(j,i-2,3);
z2 = Points(j,i-1,3);
z3 = Points(j,i,3);
z4 = Points(j,i+1,3);
Teta1 = Points(j,i-2,4);
Teta2 = Points(j,i-1,4);
Teta3 = Points(j,i,4);
Teta4 = Points(j,i+1,4);
end
else
if i == dep
x1 = Points(j,fin,1);
x2 = Points(j,i,1);
x3 = Points(j,i+1,1);
x4 = Points(j,i+2,1);
y1 = Points(j,fin,2);
y2 = Points(j,i,2);
y3 = Points(j,i+1,2);
y4 = Points(j,i+2,2);
z1 = Points(j,fin,3);
z2 = Points(j,i,3);
z3 = Points(j,i+1,3);
z4 = Points(j,i+2,3);
Teta1 = Points(j,fin,4)-2*pi;
Teta2 = Points(j,i,4);
Teta3 = Points(j,i+1,4);
Teta4 = Points(j,i+2,4);
elseif i == fin-1
x1 = Points(j,i-1,1);
x2 = Points(j,i,1);
x3 = Points(j,i+1,1);
x4 = Points(j,dep,1);
y1 = Points(j,i-1,2);
171
 Appendices

y2 = Points(j,i,2);
y3 = Points(j,i+1,2);
y4 = Points(j,dep,2);
z1 = Points(j,i-1,3);
z2 = Points(j,i,3);
z3 = Points(j,i+1,3);
z4 = Points(j,dep,3);
Teta1 = Points(j,i-1,4);
Teta2 = Points(j,i,4);
Teta3 = Points(j,i+1,4);
Teta4 = Points(j,dep,4)+2*pi;
elseif i == fin
x1 = Points(j,i-1,1);
x2 = Points(j,i,1);
x3 = Points(j,dep,1);
x4 = Points(j,dep+1,1);
y1 = Points(j,i-1,2);
y2 = Points(j,i,2);
y3 = Points(j,dep,2);
y4 = Points(j,dep+1,2);
z1 = Points(j,i-1,3);
z2 = Points(j,i,3);
z3 = Points(j,dep,3);
z4 = Points(j,dep+1,3);
Teta1 = Points(j,i-1,4);
Teta2 = Points(j,i,4);
Teta3 = Points(j,dep,4)+2*pi;
Teta4 = Points(j,dep+1,4)+2*pi;
else
x1 = Points(j,i-1,1);
x2 = Points(j,i,1);
x3 = Points(j,i+1,1);
x4 = Points(j,i+2,1);
y1 = Points(j,i-1,2);
172
 Appendices

y2 = Points(j,i,2);
y3 = Points(j,i+1,2);
y4 = Points(j,i+2,2);
z1 = Points(j,i-1,3);
z2 = Points(j,i,3);
z3 = Points(j,i+1,3);
z4 = Points(j,i+2,3);
Teta1 = Points(j,i-1,4);
Teta2 = Points(j,i,4);
Teta3 = Points(j,i+1,4);
Teta4 = Points(j,i+2,4);
end
end
Teta = (Teta2+Teta3)/2;
sinteta = sin(Teta);
costeta = cos(Teta);
xprime = (x2+x3)/2;
yprime = (y2+y3)/2;
zprime = (z2+z3)/2;
if Phi == 0
DepContr = 1;
return
end
ax = x3-x2;
ay = y3-y2;
az = z3-z2;
xc = 0;
yc = Plans(j+1,4);
zc = Plans(j+1,5);
bx = xc-xprime;
by = yc-yprime;
bz = zc-zprime;
A = -zv/yv;
B = cos(pi/2+Phi)/yv;
173
 Appendices

C = -A*ay/ax-az/ax;
D = (-B*ay)/ax;
E = 1+C^2+A^2;
F = 2*C*D+2*A*B;
G = D^2+B^2-1;
Delta = F^2-4*E*G;
zn = (-F+sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
if Signe*(yv*(zn*ax-xn*az)+zv*(xn*ay-yn*ax))>0
zn = (-F-sqrt(Delta))/(2*E);
yn = A*zn+B;
xn = C*zn+D;
end
deltax = sinteta;
deltay = costeta*cosbeta;
deltaz = -costeta*sinbeta;
r = -(xn*bx+yn*by+zn*bz)/(deltax*xn+deltay*yn+deltaz*zn);
Points(j+1,i,1) = xc+r*deltax;
Points(j+1,i,2) = yc+r*deltay;
Points(j+1,i,3) = zc+r*deltaz;
Points(j+1,i,4) = Teta;
Points(j+1,i,5) = xprime;
Points(j+1,i,6) = yprime;
Points(j+1,i,7) = zprime;
Points(j+1,i,8) = r;
end
if j>j_A & (suma-(summ-suma)<0 | suma<0.1 |
min(Points(j+1,Nteta/4:3*Nteta/4,8))<0 | suma>summ)
break
end
summ = suma;
end
Points = cat(2,Points,zeros(size(Points,1),1,8));
174
 Appendices

Points(:,Nteta+1,:) = Points(:,1,:);

175
 Appendices

SigmaC.m
global PoreWdistribution

% Import pore water pressure distribution obtained numerically from FLAC3D and
% Saved in EXCEL with name ''ppvalues''
[Matriz_Poro] = xlsread ('ppvalues_1',1,'F2:I31040');

% Interpolate the pore water pressure by discretization function ''scatteredInterpolant''

% To localize the required pore pressure in points of Limit Analysis mechanism
PoreWdistribution=scatteredInterpolant(Matriz_Poro(:,1),Matriz_Poro(:,2),Matriz_Po
ro(:,3),Matriz_Poro(:,4));

% Globalize a series of variables and assign values to each variable global Phi Coh
%Param Params_Geom sig Diam Gam Nteta Delta_Beta
Diam = 10; % Tunnel diameter
Phi = 35/180*pi; %Friction angle
Coh = 0; % Cohesion
Gam = 15.6; % Saturate unit weight
Params_Geom = [20,10]; % Initial values of mechanism geometry parameters
%Optimization

Nteta = 100; % Tunnel face discretization points number

Delta_Beta = 1; % Mechanism Section 2 angle increment of discretized faces
Delta_Beta = Delta_Beta/180*pi;

% Params_Geom optimization through calling for Function ''SigMC''

options = optimset('TolX',0.01,'Tolfun',0.01);
[Params_Geom,sig,exitflag,output] = fminsearch(@SigMC,Params_Geom,options);
%Minimize Function ''SigMC'' to get the optimal values

% Assign smalles values again to get the more precise results

Nteta = 400;
Delta_Beta = 0.2;
Delta_Beta = Delta_Beta/180*pi;
176
 Appendices

Parameters = cat (2,Params_Geom,[0,-Diam/2]);

% Call all the functions to generate the mechanism and show the collapse pressure
ValueMC
Generate
ComputeMC
Params_Geom
Critical_Collapse_Pressure = Signe*(WpG/Wt+Signe*Wr/Wt+Wu/Wt)
Trace

177
 Appendices

SigMC.m
function [sig] = SigMC(Para)
global Gam Phi Coh Params_Geom Diam Nteta Delta_Beta
Parameters = cat(2,Para,[0,-Diam/2]); % Geometry parameters for ValueMC
Para
DepContr = 0;
ValueMC
if DepContr == 1 % The mechanism generates wrongly
sig = 100000+err*100000
return
end
if Para(1)<5 % The mechanism coordinate BetaC is negative
sig = 100000-100000*Para(1)
return
end
if Para(2)<6 % The mechanism cordinate RC is negative
sig = 100000-100000*Para(2)
return
end
if Yo>0 % The rotation center lies above the tunnel crown
DepContr = 0;% The mechanism generates correctly
Generate
ComputeMC
sig = Signe*(WpG/Wt+Signe*Wr/Wt+Wu/Wt)% Signe is defined -1 in ''Generate''
else
sig = 100000-100000*Yo
end

178
 Appendices

Trace.m
Points_dessin = zeros(10,10,3);
for j = 1:size(Points,1)
if j/2 ~= round(j/2)
for i = 1:Nteta/2+1
Points_dessin(j,2*i-1,:) = Points(j,Nteta/4-1+i,1:3);
Points_dessin(j,2*i,:) = Points(j,Nteta/4-1+i,1:3);
end
else
Points_dessin(j,1,:) = Points(j,Nteta/4,1:3);
Points_dessin(j,Nteta+2,:) = Points(j,3*Nteta/4,1:3);
for i = 1:Nteta/2
Points_dessin(j,2*i,:) = Points(j,Nteta/4-1+i,1:3);
Points_dessin(j,2*i+1,:) = Points(j,Nteta/4-1+i,1:3);
end
end
for i = 1:Nteta/2
if Points_dessin(j,Nteta/2+2+i,:) == zeros(1,1,3)
Points_dessin(j,Nteta/2+2+i,:) = Points_dessin(j,Nteta/2+1+i,:);
end
if Points_dessin(j,Nteta/2+1-i,:) == zeros(1,1,3)
Points_dessin(j,Nteta/2+1-i,:) = Points_dessin(j,Nteta/2+2-i,:);
end
end
end
Points_dessin(1:j_A,Nteta+3,:) = Points_dessin(1:j_A,Nteta+2,:);
Points_dessin(1:j_A,Nteta+4,:) = Points_dessin(1:j_A,Nteta+2,:);
Points_dessin(j_A+1:size(Points_dessin,1),Nteta+3:Nteta+4,:)=Points_dessin(j_A+1:
size(Points_dessin,1),1:2,:);
Points_dessin=cat(1,Points_dessin,flipdim(Points_dessin(size(Points_dessin,1),:,:),2);
Createfigures(Points_dessin(:,:,1),Points_dessin(:,:,3),Points_dessin(:,:,2))

179
 Appendices

ValueMC.m
global x_p0 x_p1 x_p2 x_p3 y_p0 y_p1 y_p2 y_p3 x_c y_c z_c
BetaC = Parameters(1)/180*pi;
RC = Parameters(2);
Xc = Parameters(3);
Yc = Parameters(4);
Yo = RC*cos(BetaC)+Yc;
Zo = -RC*sin(BetaC);
BetaA = atan(-Zo/(Yo-Yc-Diam/2));
RA = sqrt(Zo^2+(Yo-Yc-Diam/2)^2);
BetaB = atan(-Zo/(Yo-Yc+Diam/2));
RB = sqrt(Zo^2+(Yo-Yc+Diam/2)^2);

% Calculate the curve starting from point B

j_A = Nteta/2; % The number of discretized plans in Section 1

% Address the intersection plane of Section 1 and 2

dbeta = (BetaA-BetaB)/200;
zd = 0;
yd = -Diam;
be = BetaB;
for i = 1:200
be = be+dbeta;
zdn = ((Yo+Zo/tan(be))-(yd-zd*tan(Phi+be)))/(tan(Phi+be)+1/tan(be));
ydn = -zdn/tan(be)+Yo+Zo/tan(be);
zd = zdn;
yd = ydn;
end
Rc = sqrt((Yo-yd)^2+(Zo-zd)^2); % Radius of the curve starting from B in Section 1
if Rc <= RA
DepContr = 1;
err = RA-Rc;
return
end
180
 Appendices

dbeta = Delta_Beta;
be = BetaA;
zda = 0;
yda = 0;
zdb = Zo+Rc*sin(be);
ydb = Yo-Rc*cos(be);
ra = sqrt((Yo-yda)^2+(Zo-zda)^2);
rb = sqrt((Yo-ydb)^2+(Zo-zdb)^2);

% Calculate the mechanism stop position

j_F = j_A;
while rb>ra
j_F = j_F+1;
be = be+dbeta;

% Calculate the curva starting from point A

zdna = ((Yo+Zo/tan(be))-(yda-zda*tan(-Phi+be)))/(tan(-Phi+be)+1/tan(be));
ydna = -zdna/tan(be)+Yo+Zo/tan(be);
zda = zdna;
yda = ydna;

% Calculate the curva starting from point B of Section 2

zdnb = ((Yo+Zo/tan(be))-(ydb-zdb*tan(Phi+be)))/(tan(Phi+be)+1/tan(be));
ydnb = -zdnb/tan(be)+Yo+Zo/tan(be);
zdb = zdnb;
ydb = ydnb;
ra = sqrt((Yo-yda)^2+(Zo-zda)^2);
rb = sqrt((Yo-ydb)^2+(Zo-zdb)^2);
end
RE = (rb+ra)/2;%Radius of the final point

% Assign matrix to include the information of the tunnel contour

Angles = zeros(1,Nteta);
for i = 1:Nteta
181
 Appendices

Angles(1,i) = 2*pi/Nteta*(i-0.5);
end
Contour = zeros(Nteta,5);
for i = 1:Nteta
Contour(i,3) = sin(Angles(1,i));
Contour(i,4) = cos(Angles(1,i));
Contour(i,5)=sqrt((Diam^2*Diam^2)/(Diam^2*cos(Angles(1,i))^2+Diam^2*sin(Angl
es(1,i))^2))/2;
Contour(i,1) = Xc+Contour(i,3)*Contour(i,5);
Contour(i,2) = Yc+Contour(i,4)*Contour(i,5);
end

% Assign matrix to include the information of generated planes by discretization

Plans = zeros(j_F,12);
for j = 1:j_A
Plans(j,2) = Contour(j_A+j,2);
Plans(j,1) = (Yo-Plans(j,2))/Zo;
Plans(j,7) = atan(Plans(j,1))+pi/2;
Plans(j,4) = Yo-RE*cos(Plans(j,7));
Plans(j,5) = Zo+RE*sin(Plans(j,7));
end
for j = j_A+1:j_F
Plans(j,7) = Plans(j-1,7)+Delta_Beta;
Plans(j,4) = Yo-RE*cos(Plans(j,7));
Plans(j,5) = Zo+RE*sin(Plans(j,7));
Plans(j,1) = tan(Plans(j,7)-pi/2);
Plans(j,2) = Yo-Zo*Plans(j,1);
end

% Assign matrix to include the information of generated points by discretization

Points = zeros(1,Nteta,8);
Points(1,Nteta/2,1) = Contour(Nteta/2,1);
Points(1,Nteta/2,2) = Contour(Nteta/2,2);
aax = 0;
182
 Appendices

aay = cos(Plans(1,7));
aaz = -sin(Plans(1,7));
bbx = Points(1,Nteta/2,1)-Plans(1,3);
bby = Points(1,Nteta/2,2)-Plans(1,4);
bbz = Points(1,Nteta/2,3)-Plans(1,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(1,Nteta/2,4) = acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(1,Nteta/2,8) = sqrt((Points(1,Nteta/2,3)-Zo)^2+(Points(1,Nteta/2,2)-Yo)^2);
Points(1,Nteta/2+1,1) = Contour(Nteta/2+1,1);
Points(1,Nteta/2+1,2) = Contour(Nteta/2+1,2);
bbx = Points(1,Nteta/2+1,1)-Plans(1,3);
bby = Points(1,Nteta/2+1,2)-Plans(1,4);
bbz = Points(1,Nteta/2+1,3)-Plans(1,5);
bb = sqrt(bbx^2+bby^2+bbz^2);
Points(1,Nteta/2+1,4) = 2*pi-acos((aax*bbx+aay*bby+aaz*bbz)/bb);
Points(1,Nteta/2+1,8)=sqrt((Points(1,Nteta/2+1,3)-Zo)^2+(Points(1,Nteta/2+1,2)-
Yo)^2);

183
184
 Appendices

Appendix C FLAC3D-code Numerical Simulation considering the

advance drainage

Modelo_V01.dat
rest Malla_Fina.sav
config zgroup 2
config fluid
mo mec mohr
mo fluid fl_iso
ini den 0.00152
prop bulk 417. sh 192.
prop fric 35. coh 0. dil 30.
prop perm 1e-7
prop poros 0.42
prop u_thc 1.0
ini fmod 30000.0
ini sat 1.0
ini ftens -1e-3
ini fdens 0.001
set grav 10
fix x ran x -0.001 0.001
fix x ran x 54.999 55.001
fix y ran y -20.001 -19.999
fix y ran y 49.999 50.001
fix z ran z -25.001 -24.999
ini pp 0.250 grad 0 0 -0.01
fix pp ran x 54.999 55.001
fix pp ran y -20.001 -19.999
fix pp ran y 49.999 50.001
fix pp ran z -25.001 -24.999
def ini tens
pnt = zone_head

185
 Appendices

loop while pnt#null

z_szz(pnt)=-(25-z_zcen(pnt))*0.0020*10
z_sxx(pnt)=k0x*(z_szz(pnt)+(25-z_zcen(pnt))*0.001*10)-(25-z_zcen(pnt))*0.001*10
z_syy(pnt)=k0y*(z_szz(pnt)+(25-z_zcen(pnt))*0.001*10)-(25-z_zcen(pnt))*0.001*10
pnt = z_next(pnt)
end_loop
end
def variables
k0x = 0.5000
k0y = 0.5000
end
@variables
@ini_tens
range name Contorno_tunel cyl end1 0.0 -20.01 0.0 end2 0.0 50.01 0.0 rad 5.001
hist unbal
hist ratio
hist add gp yvel id 475
save Malla_Final.sav
rest Malla_Final.sav
set mech off
set fluid on
save Fase_1.sav
rest Fase_1.sav
fix pp
set mech on
set fluid off
solve ratio 1.0e-7
save Fase_2.sav
rest Fase_2.sav
ini xd 0 yd 0 zd 0
ini xv 0 yv 0 zv 0
ini xs 0 ys 0 zs 0
ini state 0
free pp
186
 Appendices

fix pp ran x 54.999 55.001

fix pp ran y -20.001 -19.999
fix pp ran y 49.999 50.001
fix pp ran z -25.001 -24.999
mo null ran group TUNNEL y -20.0 0.0;
mo null ran group DRAINAGE y -20.0 30.0
fix x,z range Contorno_tunel y -20.0001 0.001
fix x,z range group DRAINAGE y -0.001 30.001
free x,z range cyl end1 0.0 -0.001 0.0 end2 0.0 0.001 0.0 rad 4.999
fix x ran x -0.001 0.001
fix pp range cyl end1 0.0 -0.001 0.0 end2 0.0 0.001 0.0 rad 5.001
ini pp 0 range cyl end1 0.0 -0.001 0.0 end2 0.0 0.001 0.0 rad 5.001
fix pp range group DRAINAGE y -0.001 30.001
ini pp 0 range group DRAINAGE y -0.001 30.001
set mech off
set fluid on
step 1000000
save Fase_3.sav
call Modelo_V02.dat

187
 Appendices

Modelo_V02.dat
rest Fase_3.sav
call Intervalo.dat
@limites
def acotacion
P_cal = 1./2. * (P_sup+P_inf)
if abs(P_sup-P_inf) > 0.1e-3 then
command
app syy @P_cal range Contorno_tunel y -0.1 0.1
print @P_sup
print @P_inf
print @P_cal
@estabilidad
end_command
if estable = 1 then
P_sup_archivo = P_cal
P_inf_archivo = P_inf
else
P_sup_archivo = P_sup
P_inf_archivo = P_cal
end_if
fname = 'Calculo_Presion_'+string(k)+'.sav'
if float(k)/1. = k/1 then
command
save @fname
end_command
end_if
array bbbb(5)
bbbb(1) = 'def limites'
bbbb(2) = 'P_sup = ' + string(P_sup_archivo) + ' * 1.'
bbbb(3) = 'P_inf = ' + string(P_inf_archivo) + ' * 1.'
bbbb(4) = 'k = ' + string(k+1)
bbbb(5) = 'end'
dummy = open('Intervalo.dat',1,1)
188
 Appendices

dummy = write(bbbb,5)
dummy = close
command
call Modelo_V02.dat
end_command
end_if
end
@acotacion

189
 Appendices

Intervalo.dat
def limites
P_sup = -1000.
P_inf = 0.
k=9
end

190
 Appendices

Mixboundary.dat
def mixedboundary
while_stepping
gp_pnt = gp_head
loop while gp_pnt # null
x0 = gp_xpos(gp_pnt)
y0 = gp_ypos(gp_pnt)
z0 = gp_zpos(gp_pnt)
if sqrt(x0^2+z0^2) < 5.05
if y0 > -0.05
if y0 < 0.05
x1 = gp_xpos(gp_pnt)-0.05
x2 = gp_xpos(gp_pnt)+0.05
z1 = gp_zpos(gp_pnt)-0.05
z2 = gp_zpos(gp_pnt)+0.05
if gp_pp(gp_pnt) >= 0 then
command
apply pp 0 range x @x1 @x2 z @z1 @z2 y -0.05 0.05
end_command
end_if
end_if
end_if
end_if
gp_pnt = gp_next(gp_pnt)
end_loop
end
@mixedboundary

191