Hardening Soil Model

Hardening Soil model with small strain stiffness
Andrzej Truty
ZACE Services
1.09.2008
Andrzej Truty ZACE Services Hardening Soil model with small strain stiffness
Introduction
Hardening Soil (HS) and Hardening Soil-small (HS-small)

models are designed to reproduce basic phenomena exhibited
by soils:
densification
stiffness stress dependency
plastic yielding
dilatancy
strong stiffness variation with growing shear strain amplitude
in the regime of small strains (γ = 10−6 to γ = 10−3 )
this phenomenon plays a crucial role for modeling deep
excavations and soil-structure interaction problems
NB. This model is limited to monotonic loads
Introduction
HS model was initially formulated by Schanz, Vermeer and

Bonnier (1998, 1999) and then enhanced by Benz (2006)
Current implementation is slightly modified with respect to
the theory given by Benz:
simplified treatment of dilatancy for the small strain version
(HS-small)
modified hardening law for preconsolidation pressure
This model seems to be one of the simplest in the class of
models designed to handle small strain stiffness
It consists of the two plastic mechanisms, shear and volumetric
Small strain stiffness is incorporated by means of nonlinear
elasticity which includes hysteretic effects
Triaxial test: illustration
Undrained triaxial test: video

Drained triaxial test : video
Annimations by P.Baran (University of Agriculture, Kraków,
Poland)
Notion of tangent and secant stiffness moduli
Initial stiffness modulus Eo
Unloading-reloading modulus Eur
Secant stiffness modulus at 50 % of the ultimate deviatoric
stress qf
250
Eo qf
200
E50 qun
150
1 1
q [kpa]
100
0.5 qf
q50 Eur
50
1 σ3=const
0
0 0.05 0.1 0.15 0.2 0.25
EPS-1 [-]
Remark: All classical soil models require specification of Eur

modulus (Cam-Clay, Cap etc..)
Stiffness-strain relation for soils (G /Go (γ))
G - current secant shear modulus

Go - shear modulus for very small strains
Atkinson 1991
Notion of treshold shear strain γ07
G
To describe the shape of (γ) curve an additional
Go
characteristic point is needed
It is common to specify the shear strain γ0.7 at which ratio
G
= 0.7
Go
0.7
γ07
Dynamic vs static modulus
Relation between ”static” Young modulus Es , obtained from

standard triaxial test at axial strain ε1 ≈ 10−3 , and ”dynamic”
Young modulus (the one at very small strains) Ed = Eo is
shown in diagram published by Alpan (1970) (after Benz)
100
Ed
Ro
Es ck
s
co h
es i
10 ve
so i
ls
gran
u l ar
soil
s
Es [kPa]
1
1000 10000 100000 1000000
HS model: general concept
Double hardening elasto-plastic model (Schanz, Vermeer,
Benz)
Nonlinear elasticity for stress paths penetrating the interior of
the elastic domain
600
500
Cap surface
400
q [kPa]
300
200
100
0
0 100 200 300 400 500
p [kPa]
Graphical representation of shear mechanism and cap surface

HS model: shear mechanism
Duncan-Chang model as the origin for shear mechanism

250
qf 200
M-C limit
150
q [kPa]
100
Eur
50 E50 ½ qf
1
1
0
0 0.01 0.02 0.03 0.04 0.05
eps-1
Stiffness stress dependency
m
σ3∗ + c cotφ

ref
Eur = Eur
σ ref + c cotφ
σ3∗ + c cotφ m

ref
E50 = E50
σ ref + c cotφ
Remarks
1 Stiffness degrades with decreasing σ 3 up to σ 3 = σL (by
default we assume σL =10 kPa)
Extension to small strain: new ingredients
To extend standard HS model to the range of small strain Benz

introduced few modifications:
1 Strain dependency is added to the stress-strain relation, for
stress paths penetrating the elastic domain
2 The modified Hardin-Drnevich relationship is used to relate
current secant shear modulus G and equivalent monotonic
shear strain γhist
3 Reversal points are detected with aid of deviatoric strain
history second order tensor Hij ; in addition the current
equivalent shear strain γhist is computed by using this tensor
How does it work ?
N
N-1 N+1
plot from paper by Ishihara 1986

At step N : γhistN−1 = 8 × 10−5 γhistN = 10−4
At step N + 1 : γhistN = 0 γhistN+1 = 2 × 10−5
max
Primary loading: γhistN+1 > γhist
max
Unloading/reloading: γhistN+1 ≤ γhist
Go
Hardin-Drnevich law: G = γhist (secant modulus)
1+a
γ0.7
Shear tangent modulus cut-off
Gur
γc γ
r
γ0.7 Go
γc = −1
a Gur
Setting initial state variables: γoPS and pco
Given: σ o , OCR
Find: γoPS and pco
600
500
Cap surface
400
σSR
q [kPa]
300
200 Shear mechanism
100 σο
0
0 100 200 300 400 500
p [kPa]
Procedure:
Set effective stress state at the SR point
σySR = σyo OCR
σxSR = σzSR = σy KoSR
600
500
Cap surface
400
σSR
q [kPa] 300
200 Shear mechanism
100 σο
0
0 100 200 300 400 500
p [kPa]
Procedure:
For given σ SR state compute γoPS from plastic condition
f1 = 0
For given σ SR state compute pco from plastic condition f2 = 0
Remarks
1 KoSR = KoNC ≈ 1 − sin(φ) in the standard applications
(approximate Jaky’s formula)
2 KoSR = 1 for case of isotropic consolidation (used in triaxial
testing for instance)
3 For sands notion of preconsolidation pressure is not as
meaningful as for cohesive soils hence one may assume
OCR=1 and effect of density will be embedded in H and M
parameters
Setting M and H parameters based on oedometric test
σ
600
500
Eoed
400
q [kPa]
q* 1
300
200
100
p*
0
0 100 200 300 400 500
ε
p [kPa] σoed
ref
Material properties
Parameter Unit HS-standard HS-small
ref
Eur [kPa] yes yes
ref
E50 [kPa] yes yes
ref
σ [kPa] yes yes
m [—] yes yes
νur [—] yes yes
Rf [—] yes yes
c [kPa] yes yes
φ [o ] yes yes
ψ [o ] yes yes
emax [—] yes yes
ft [kPa] yes yes
D [—] yes yes
M [—] yes yes
H [kPa] yes yes
OCR/q POP [—/kPa] yes yes
Eoref [kPa] no yes
γ0.7 [—] no yes
Converting MC to HS model: indentation problem
Assumption: q = 0.5 qult

Eur E50
Given: E for MC model and = ..., = ...
E50 Eoed
1m
q = 0.5 qult
A
10m
10m
ref , M and H for standard HS model

Find: Eur
Example: triaxial test on dense Hostun sand
6
5.5
120000
5
100000
SIG-1 / SIG-3 [kPa]

4.5
4 80000
HS-std
G [kPa]
3.5
HS-small HS-std
60000
3 HS-small
2.5 40000
2
20000
1.5
1 0
0 0.02 0.04 0.06 0.08 0.1 0.00001 0.0001 0.001 0.01 0.1 1
-EPS-Y [-] EPS-X - EPS-Y [-]
σ
(a) 1 (ε1 ) (Z Soil) (b) G (γ) (Z Soil)
σ3
4
0 0.02 0.04 0.06 0.08 0.1

3.5
0.08
SIG-1 / SIG-3 [kPa]
0.07
3
0.06
HS-std 0.05
2.5
HS-small
-EPS-V [-]
0.04
HS-std
2 0.03
HS-small
0.02
1.5 0.01
0
1 -0.01
0 0.002 0.004 0.006 0.008 0.01 -0.02
-EPS-Y [-] -EPS-Y [-]
σ
(c) 1 (ε1 ) (zoom) (Z Soil) (d) εv (ε1 ) (Z Soil)
σ3
6
200000
5.5
180000
5
160000
SIG-1 / SIG-3 [kPa]

4.5
140000
4
120000
G [kPa]
HS-std
3.5 HS-std
HS-small 100000
3 HS-small
80000
2.5 60000
2 40000
1.5 20000
1 0
0 0.02 0.04 0.06 0.08 0.1 0.00001 0.0001 0.001 0.01 0.1 1
EPS-1 [-] EPS-1 - EPS-3 [-]
σ
σ3
4
0 0.02 0.04 0.06 0.08 0.1

3.5
0.08
SIG-1 / SIG-3 [kPa]
0.07
3
0.06
HS-std 0.05
2.5
HS-small
EPS-V [-]
0.04
HS-std
2 0.03
HS-small
0.02
1.5 0.01
0
1 -0.01
0 0.002 0.004 0.006 0.008 0.01 -0.02
EPS-1 [-] EPS-1 [-]
σ
σ3
6
300000
5.5
5 250000
SIG-1 / SIG-3 [kPa]

4.5
200000
4
G [kPa]
HS-std
3.5 HS-std
HS-small 150000
HS-small
3
2.5 100000
2
50000
1.5
1 0
0 0.02 0.04 0.06 0.08 0.1 0.00001 0.0001 0.001 0.01 0.1 1
EPS-1 [-] EPS-1-EPS-3 [-]
σ
σ3
4
0 0.02 0.04 0.06 0.08 0.1
3.5 0.08
0.07
SIG-1 / SIG-3 [kPa]
3 0.06
0.05
HS-std
2.5
EPS-V [-]
HS-small 0.04
HS-std
0.03
2 HS-small
0.02
1.5 0.01
0
1 -0.01
0 0.002 0.004 0.006 0.008 0.01 -0.02
EPS-1 [-] EPS-1 [-]
σ
σ3
Estimation of material properties: input data
Given 3 drained triaxial test results for 3 confining pressures:

σ 3 = 100 kPa
σ 3 = 300 kPa
σ 3 = 600 kPa
Shear characteristics q − ε1
Dilatancy characteristics εv − ε1
Stress paths in p − q plane
Measurements of small strain stiffness moduli Eo (σ3 ) for the
assumed confining pressures (through direct measurement of
shear wave velocity in the sample)
Estimation of material properties: stress paths in p-q plane
Estimation of friction angle φ = φcs and cohesion c

q
Residual M-C envelope

6 sin φ
M* =
3 − sin φ
1
6c cos φ
c* =
3 − sin φ
p
If we know M ∗ and c ∗ then we can compute φ and c:

3 M∗ 3 − sin φ
φ = arcsin c = c∗
6 + M∗ 6 cos φ
Estimation of material properties: stress paths in p-q plane
Estimation of friction angle φ = φcs and cohesion c
3000
2500 2358/1386=1.7
2358 1
2000
q [kPa]
1500
1000
500
1386
0
0 300 600 900 1200 1500 1800
p [kPa]
3 ∗ 1.7
Here: φ = arcsin ≈ 42o c=0
6 + 1.7
Estimation of material properties: dilatancy angle
0.06
0.05
0.04
1
0.03
EPS-V [-]
d Dilatancy cut-off
0.02
0.01
-0.01
-0.02
0 0.02 0.04 0.06 0.08 0.1
EPS-1 = - EPS-3 [-]

d
ψ = arcsin
2+d
Estimation of material properties: dilatancy angle
0.06
0.05 1
0.04
d=0.75
εV 0.03
0.02
0.01
-0.01
-0.02
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
ε1

0.75
ψ = arcsin ≈ 16o
2 + 0.75
Estimation of material properties: Eoref and m
m
σ3∗ + c cotφ

Analytical formula: Eo = Eoref
σ ref + c cotφ
Measured: shear wave velocity vs at ε1 = 10−6 and at given
confining stress σ3
Compute : shear modulus Go = ρvs2
Compute : Young modulus Eo = 2 (1 + ν) Go
σ3 [kPa] Eo [kPa]
100 250000
300 460000
600 675000
m
σ3∗ + c cotφ

Analytical formula: Eo = Eoref
σ ref + c cotφ
Measured: shear wave velocity vs at ε1 = 10−6 and at given
confining stress σ3
Compute : shear modulus Go = ρvs2
Compute : Young modulus Eo = 2 (1 + ν) Go
σ3 [kPa] Eo [kPa]
100 250000
300 460000
600 675000
Reanalyze Eo vs σ3 in logarithmic scales
13.1 − 12.55
Averaged slope yields m; here m = = 0.55
1.0
Find intersection ofthe line with axis ln Eo at
σ3∗ + c cotφ
ln =0
σ ref + c cotφ
Here the intersection is at 12.43 hence
Eoref = e 12.43 ≈ 2.71812.43 = 250000 kPa
13.6
13.4
13.2
ln Eo 13 m
12.8
12.6 ⎛ σ + c cot φ ⎞
1 ln⎜⎜ ref3 ⎟⎟
12.4 12.43
⎝ σ + c cot φ ⎠
12.2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Estimation of Eoref from CPT testing
To estimate small strain modulus Go at a certain depth one

may use empirical formula by Mayne and Rix:
q 0.695
Go = 49.4 t1.13 [MPa]
e
qt is a corrected tip resistance expressed in MPa
e is the void ratio
ref
Estimation of material properties: E50
Lets us find E50 for each confining stress
2500
E50 (σ 3 = 300 kPa )
1
2000
E50 (σ 3 = 600 kPa ) E50 (σ 3 = 100 kPa )
1 1
1500
q50f (σ 3 = 100)
1000
q50f (σ 3 = 100)
500
q50f (σ 3 = 100)
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
ref
Estimation of material properties: E50
Reanalyze E50 vs σ3 in logarithmic scales
Here we can fix m to the one obtained for small strain moduli
Find intersection ofthe line with axis ln E50 at
σ3∗ + c cotφ
ln =0
σ ref + c cotφ
Here the intersection is at ≈ 10.30 hence
ref ≈ e 10.30 ≈ 2.71810.30 ≈ 30000 kPa
E50
11.4
11.2
11
10.8
ln E50
10.6 ⎛ σ + c cot φ ⎞
ln⎜⎜ ref3 ⎟⎟
10.4
10.30 ⎝ σ + c cot φ ⎠
10.2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Estimation of material properties: Eurref
The unloading reloading modulus as well as oedometric

moduli are usually not accessible
We can use Alpans diagram to deduce Eur ref once we know
E ref
Eoref (default is ur = 3); for cohesive soils like tertiary clays
Eoref
this value is larger
For oedometric modulus at the reference stress σ ref = 100
ref = E ref
kPa we can assume Eoed 50
γ0.7 = 0.0001...0.0002 for sands and γ0.7 = 0.00005...0.0001
for clays
Smaller γ0.7 values yield softer soil behavior
Excavation in Berlin Sand: engineering draft
Excavation in Berlin Sand: FE discretization
Excavation in Berlin Sand: Bending moments
-500 -400 -300 -200 -100 0 100 200 300 400 500
0
-5
-10
-15 HS
Y [m[]
HS-small
-20 MC
-25
-30
-35
M [kNm/m]
Excavation in Berlin Sand: Wall deflections
-0.015 -0.01 -0.005 0 0.005 0.01

0
-5
-10
-15 HS-small
Y [m]
HS
-20 MC
-25
-30
-35
Ux [m]
Excavation in Berlin Sand: Soil deformation in cross
section x =20m
0 0.01 0.02 0.03 0.04

0
-10
-20
-30
-40 HS
Y [m]
-50 HS-small
-60 MC
-70
-80
-90
-100
Uy [m]
Vertical heaving of subsoil at last stage of excavation, relative to

the step when dewatering was finished (t = 2)
Excavation in Berlin Sand: Soil deformation in cross
section x =20m
-0.004 -0.003 -0.002 -0.001 0

0
-10
-20
-30
-40 HS
Y [m]
-50 HS-small
-60 MC
-70
-80
-90
-100
Ux [m]
Horizontal movement in cross section x=20m at last stage of

excavation, relative to the step when dewatering was finished
(t = 2)
Conclusions
Model properly reproduces strong stiffness variation with shear

strain
It can be used in simulations of soil-structure interaction
problems
Implementation is ”rather heavy”
It should properly predict deformations near the excavations
Model reduces excessive heavings at the bottom of the
excavation

Hardening Soil Model

Uploaded by

Copyright:

Available Formats

Hardening Soil Model

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hardening Soil Model

Uploaded by

Copyright:

Available Formats

Hardening Soil model with small strain stiffness

Hardening Soil (HS) and Hardening Soil-small (HS-small)

HS model was initially formulated by Schanz, Vermeer and

Undrained triaxial test: video

Remark: All classical soil models require specification of Eur

G - current secant shear modulus

Relation between ”static” Young modulus Es , obtained from

Graphical representation of shear mechanism and cap surface

Duncan-Chang model as the origin for shear mechanism

To extend standard HS model to the range of small strain Benz

plot from paper by Ishihara 1986

200 Shear mechanism

200 Shear mechanism

Assumption: q = 0.5 qult

ref , M and H for standard HS model

SIG-1 / SIG-3 [kPa]

0 0.02 0.04 0.06 0.08 0.1

SIG-1 / SIG-3 [kPa]

0 0.02 0.04 0.06 0.08 0.1

SIG-1 / SIG-3 [kPa]

Given 3 drained triaxial test results for 3 confining pressures:

Estimation of friction angle φ = φcs and cohesion c

Residual M-C envelope

If we know M ∗ and c ∗ then we can compute φ and c:

Estimation of friction angle φ = φcs and cohesion c

To estimate small strain modulus Go at a certain depth one

Lets us find E50 for each confining stress

The unloading reloading modulus as well as oedometric

-0.015 -0.01 -0.005 0 0.005 0.01

0 0.01 0.02 0.03 0.04

Vertical heaving of subsoil at last stage of excavation, relative to

-0.004 -0.003 -0.002 -0.001 0

Horizontal movement in cross section x=20m at last stage of

Model properly reproduces strong stiffness variation with shear

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.