Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.

269-292, 2008

Numerical Simulations of Dynamic Fracture in Thin Shell

Structures

C. Gato and Y. Shie1

Abstract: Numerical simulations of large deformation dynamic fracture in thin

shell structures using 3-D meshfree method is presented. Due to the smoothness of
the meshfree shape functions, they are well suited to simulate large deformation of
thin shell structures while avoiding ill-conditioning as well as stiffening in numer-
ical computations. Dynamic fracture is modeled by simple criterion, i.e. removing
connectivity between adjacent nodes once a fracture criterion is met. The main
advantage of such 3-D meshfree continuum approach is its simplicity in both for-
mulation and implementation as compared to shell theory approach, or degenerated
continuum approach. Moreover, it is believed that the accuracy of the computation
may increase because of using 3-D exact formulation.

Keyword: fracture, shell, meshfree

1 Introduction

The numerical simulation of thin shell structures has been a challenge in applied
mechanics and in many engineering branches for many years. Its engineering sig-
nificance as well as technical difficulties can be best measured by the seemingly
ever-lasting ’new formulations’ or ’new contributions’ in the literature in the past
decades. Its applications cover e.g. sheet metal forming, crash-worthiness test,
civil structure design, pressure vessel liability, shipbuilding, defense technology,
just to name a few. Regarding the strategy of numerical simulations of thin shell
structures, there are three major approaches:

1. numerical simulation based on shell theories

2. degenerated continuum, or continuum based approach

3. direct three-dimensional (3-D) continuum approach

1 Harbin Engineering University College of Mechanical Engineering Harbin, HLJ, P.R.China
270 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Among these three approaches, 3-D continuum direct approach is the simplest;
nonetheless it is the least popular one in practice. The major drawback, or dilemma
that prevents using 3-D direct simulation is that it is often required to deploy mul-
tiple elements in the thickness direction of the thin shell to acquire reasonable gra-
dient fields, which, on the other hand, leads to degrading the conditioning of the
discrete system and then the accuracy of the numerical solution; moreover, the di-
rect continuum approach is very expensive, which usually requires more elements
in the same simulation than shell theory approach, or degenerated approach does.
Most meshfree methods proposed so far have focused on a continuum-based ap-
proach or simply modeled the shell or plate as a standard continuum. A meshfree
thin shell formulation based on Kirchhoff-love theory and element-free Galerkin
(EFG) method Belytschko, Lu, and Gu (1994) has been developed by Krysl and Be-
lytschko (1996) in the context of small strain, linear elastic framework. Rabczuk,
Areias, and Belytschko (2007) extended this work with the consideration of finite
strain, non-linear elastic material and focused on fracture. In the following work,
Rabczuk and Areias (2006) have simplified the treatment of cracks in thin shell by
using an extrinsic basis. Donning and Liu (1998) noted the advantage of meshfree
approximations in addressing shear locking in Mindlin type of beams and plates and
have developed a meshfree formulation based on the reproducing kernel particle
method (RKPM) Liu, Jun, and Zhang (1995). This methodology is further extended
by Kanok-Nukulchai, Barry, Saran-Yasoontorn, and Bouillards (2001) with the use
of EFG. EFG has been employed by Noguchi, Kawashima, and Miyamura (2000)
for shell and membrane structures in which bi-cubic and quartic basis functions are
introduced in order to avoid shear and membrane locking. Leitao (2001) developed
a meshfree method based on radial basis functions (RBF) for modeling a Kirchhoff
type of plate. Extension of RBF approach to the Mindlin type of plate was presented
in Liew and Chen (2004b,a). The meshless local Petrov-Galerkin was proposed by
Atluri and Zhu (1998, 2000); Atluri and Shen (2002); Atluri (2002); Han and Atluri
(2003); Tang, Shen, and Atluri (2003); Liu, Han, Rajendran, and Atluri (2006) for
solving beam problems and application of this meshfree approach to plates and
shells can be found in Long and Atluri (2002); Atluri, Cho, and Kim (1999); Soric,
Li, Jarak, and Atluri (2004); Sladek, Sladek, Wen, and Aliabadi (2006); Jarak,
Soric, and Hostler (2007); Li, Soric, Jarak, and Atluri (2005); Andreaus, Batra,
and Porfiri (2005); Sladek, Sladek, and Solek (2008); Jiawei, Xuefeng, and Lianfa
(2008). Wang and Chen (2006) showed that the Kirchhoff mode in the Mindlin
plate can be reproduced using EFG or RKPM if second-order polynomial basis
is used in the moving least-squares approximation. By implementing this with a
nodal integration and stabilization scheme, they have shown that the formulation is
stable and free of shear locking. Yagawa and Miyamura (2005) developed a free
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 271

mesh method in which the discrete Kirchhoff theory is combined with the mixed
approach. In the case of 3D continuum models, Li, Hao, and Liu (2000) have
presented a formulation based on RKPM and have studied non-linear large defor-
mation of thin shells. Other relevant literature related to fracture or shell theory are
given e.g. in Hao and Liu (1999); Hao, Liu, and Chang (2000); Guz, Menshykov,
and Zozulya (2007); Ju and Liu (2007); Rabczuk and Belytschko (2006); Hagi-
hara, Tsunori, and Ikeda (2007); Nishioka, Kobayashi, and Fujimoto (2007); Hao,
Liu, and Weertman (2004); Liu, Liu, and Mahadevan (2007); Hao, Liu, Moran,
Vernerey, and Olson (2004); Guo and Nairn (2006); Zhao, Liu, and Wu (2008);
Rabczuk and Zi (2007); Ma, Lu, and Wang (2006); Gao, Liu, and Liu (2006); Hao,
Liu, Moran, and Olson (2003); Hao, Liu, Klein, and Rosakis (2004); Rong, Huang,
Liu, Song, and Wang (2008); Hao, Liu, and Belytschko (2004); Rabczuk, PMA,
and Belytschko (2007); Fujimoto and Nishioka (2006); Nishioka (2005); Hao and
Liu (2006); Nguyen-Van, N, and Tran-Cong (2008); Le, Mai-Duy, and Tran-Cong
(2008).
In this paper, we use meshfree 3-D continuum approach based on simple fracture
criterion. Once fracture criterion is met, connectivity between meshfree nodes are
removed. The advantage of this 3-D meshfree approach is its simplicity over more
complex methods.
The paper is structured as follows: We first describe the meshfree approximation
and the discrete equations. Then, we elaborate the fracture criterion. Two problems
involving large deformation dynamic fracture of thin shell structures are studied. At
the end, we conclude our paper and give future research directions.

2 General formulation

We have used a formulation within the total Lagrangian framework. Conservation

of linear momentum can be written as

∇X · P + ρ0 b = ρ0 ü, X ∈ Ω0 (1)

where b is the body force, ρ0 is initial density, u is displacement, P is the first

Piola-Kirchhoff stress tensor, ∇X denotes spatial derivatives with respect to mate-
rial coordinate and superimposed dots denote material time derivatives. The weak
formulation of linear momentum equation is: Find u ∈ U and δ u ∈ U0 such that

δ W = δ Wint − δ Wext + δ Wkin = 0 (2)

272 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

with

δ Wint = ∇X δ u : P dΩ0
Ω0
 
δ Wext = δ u · t̄0 dΓ0 + ρ0 δ u · b dΩ0
Γ0t Ω0

δ Wkin = ρ0 δ u · ü dΩ0 (3)
Ω0

with the approximation spaces U and U0 for the trial and test functions, respec-
tively,
 
U = u|u ∈ H 1 , u = ū on Γu
 
U0 = δ u|δ u ∈ H 1 , δ u = 0 on Γu (4)

The linear momentum equation is complemented with Dirichlet and von Neumann
boundary conditions:

u = ū, X ∈ ΓXu (5)

n0 · P = t0 = t̄0 , X ∈ Γ0t (6)

where the index t refers to traction boundaries, the index u refers to displacement
boundaries; n is the normal to the traction boundary and the subscript 0 refers to
quantities in the reference configuration; t is the traction.

3 Meshfree approximation
We use the elementfree Galerkin (EFG) method Belytschko, Lu, and Gu (1994)
that is based on moving least squares (MLS) Lancaster and Salkauskas (1981) ap-
proximation. The EFG approximation uh (X) of a given function u(X) can be posed
in terms of the shape functions and certain particle or nodal parameters uJ as
n
uh (X) = ∑ NJ (X)uJ = N u (7)
J=1

with n particles and shape functions

NT (X) = pT (X)A−1 (X)PW(X) (8)

where A is the moment matrix with moment matrix

A(X) = P(Y)W(X)PT (Y) (9)

Numerical Simulations of Dynamic Fracture in Thin Shell Structures 273

where the matrix PT (Y) contains the polynomial basis p and

W(X) = diag {WI (X − XI , h)VI } , I = 1, ..., n (10)

WI (X − XI , h) is a kernel function with compact support, its parameter h, usually

called smoothing length or dilation parameter is a certain characteristic measure of
the size of the support of the kernel function (e.g. the radius in circular supports).
We used the quartic spline function that is commonly used in the literature:

1 − 6s2 + 8s3 − 3s4 s ≤ 1
W (X − XI , h) = w(s) = (11)
0 s>1

with s = X−X 2h for circular support size. The size of the domain of influence is
I

connected to the nodal spacing. We consider only structured nodal arrangements

Rabczuk and Belytschko (2005). The support size of the domain of influence, i.e.
the radius of the support size is twice the particle spacing, figure 1. Such support
sizes are commonly used in meshfree methods ?. The matrix P contains the poly-
nomial basis p. Instead of the global polynomial basis p(Y), we use a scaled locally
defined polynomial basis p ((Y − X)/h), that leads to better conditioned moment
matrix A. In this work, we use a quadratic polynomial basis. We note that the
EFG approximation does not fulfil the Kronecker-delta property, i.e. u(XI ) = uI .
This requires specific attention when imposing displacement boundary conditions.
Also, if for example the displacement field needs to be plotted, then eq. (7) needs
to be used in the post-processing. More details on the EFG method can be found in
the literature Belytschko, Lu, and Gu (1994); ?.

support size

(a)
Figure 1: Radial support size
274 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

The test and trial functions have the structure of equation (7). Introducing them
into the weak formulation with a Bubnov Galerkin method yields
  
n n
∑ δ uI ∑− Ω0
∇X NI (X)P dΩ0 +
Ω0
NI (X)b dΩ0
I=1 J=1
  
+ NI (X)t̄0 dΓ0 + ρ0 NI (X)NJ (X)u dΩ0 =0 (12)
Γ0t Ω0

Thus, for each particle I, the following identity must hold

n  
∑ ∇X NI (X)P dΩ0 = NI (X)b dΩ0
J=1 Ω0 Ω0
 
+ NI (X)t̄0 dΓ0 + ρ0 NI (X)NJ (X)u dΩ0 = 0 (13)
Γ0t Ω0

These equations can be recast into a matrix form

MIJ üJ = −fext

I + fI
int
(14)

with

MIJ = ρ NI (X) NTJ (X) dΩ0
Ω0

 
fext
I = NTI (X) t̄0 dΓ0 + NTI (X) bdΩ0 (15)
Γ0t Ω0


fint
I = ∇X NTI X) PdΩ0 (16)
Ω0

The domain is subdivided into integration domains over which Gaussian quadrature
is performed. Background mesh is constructed such that nodes and integration
cell vertices coincide, see figure 2 for a two-dimensional description. We note
that the background mesh does not necessarily have to be conforming and hanging
nodes may easily be employed. Using Gaussian quadrature, the internal forces for
examples are given by Rabczuk, Belytschko, and Xiao (2004)

I =
fint ∑ n∇X N(X(ξJ ) − XI ) P(X(ξJ )) wQJ |J| (17)
J=1

where ξ are local coordinates of the background mesh, |J| is the determinant of the
Jacobian and wQJ are the quadrature weights.
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 275

(a)
Figure 2: EFG nodes plotted as circular black dots with background mesh; left side:
unstructured mesh and right side: structured mesh

4 Constitutive model and Failure criterion

4.1 Visoplasticity model
We use a J2 isotropic hardening viscoplastic model of the type

ε p 1/n
g(ε ) = σ0 1 + p
p
(18)
ε0
p
where σ0 is yield stress, ε p and ε0 are the total and reference plastic strains, re-
spectively and 1/n is hardening exponent.
Rate dependent behavior is modeled with power law of type

ε̇ p 1/m
σe f f = g(ε ) 1 + p
p
(19)
ε̇0
where 1/m is reference plastic strain rate and 1/m is strain rate sensitivity exponent.

4.2 Thermo-viscoplastic model

In this section, we outline the constitutive relation of the thermo-elasto-viscoplastic
solid adopted from Zhou, Ravichandran, and Rosakis (1996) in order to evaluate
the stress term in Equation 2. The rate form of the constitutive equation reads as
follows:


τ ∇ = C : D − Dvp − α Ṫ I (20)
where C is the first order elasticity tensor, D is the symmetric part of the velocity
gradient L, τ ∇ = τ̇ − W · τ − τ · W is the Jaumann rate of the Kirchhoff stress
where W is the antimetric part of the velocity gradient, α is the thermal expansion
coefficient and I is the second-order identity matrix. The viscoplastic overstress
model here is based on von Mises
 ¯
3ε̇
D =
vp
s̃ (21)
2σ̄
276 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

with

s̃ = s − a with s = τ − 1/3tr(τ ) I, σ̄ = 3/2s̃ : s̃ (22)

where a is the back stress, set to zero in our studies. The thermo viscoplastic flow
is governed by the following power law
 m
¯ε̇ = ε̇0 σ̄
(23)
g(ε̄ , T )

with
 n   
ε̄ T − T0
g(ε̄ , T ) = σ̄ 1 + 1 − δ exp −1 (24)
ε0 κ

In Equations (23) and (24), ε̇0 is a reference strain rate, m is the rate sensitivity
parameter, σ0 is the yield stress, ε0 = σ0 /E is the corresponding reference strain
and E is Young’s modulus, n is the strain hardening exponent, T0 is a reference
temperature and δ and κ are thermal softening parameters. The function g(ε̄ , T ) is
the stress-strain relation measured at quasi-static strain rate of ε̇ at temperature T .
The equivalent plastic strain ε̄ is defined as
 t  t
2 vp vp
ε̄ = ε̄ dt =
˙ D : D dt (25)
0 0 3

Softening in material due to temperature is accounted for by varying material pa-

rameters


E(T ) = E0 − 1.6 × 106 (T − T0 ) − 105 T − T) [Pa]
ν = v0 + 5 × 10−5 (T − T0 )
σ0 (T ) = σ0 − 1.5 × 103 (T − T0 )2 [Pa]
α (T ) = (2.2 + 0.0016 [T − T0 ]) × 10−5 [K −1 ] (26)

where E and ν are Young’s modulus and the Poisson ratio at temperature T . Fur-
thermore, for steel: E0 = 200GPa, ν0 =0.3 and σ0 =2.0GPa. The constitutive update
scheme for the thermo-elasto-viscoplastic model largely follows the rate tangent
modulus approach developed by Peirce, Shih, and Needleman (1984). The essence
of the rate tangent modulus method is to approximate any function of time in the
interval tn+θ ∈ [tn ,tn+1 ], θ ∈ [0, 1] as

f θ = (1 − θ ) fn + θ fn+1 (27)
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 277

n+1 = vn + Δt an it follows that

Thus, with the predicted velocity field vtrial

vθ = (1 − θ ) vn + θ vtrial
n+1 = vn + θ Δt an (28)
uθ = (1 − θ ) un + θ un+1 = un + θ Δt vn + θ 2 Δt 2 an (29)
Lθ = ∇X vθ · F−1
n+1 (30)

With θ = 0.5, we recovered central difference scheme. To update the Kirchhoff

stress, we proceed as follows:

τn+1 = τn + τ̇ Δt (31)
τ̇ ≈ τ ∇ + Wθ · τn + τn · WTθ (32)

Of course, one has to find first ε̄˙θ before τ ∇. Let

ε̄˙θ = (1 − θ ) ε̄˙n + θ ε̄˙n+1 (33)

where ε̄˙n+1 is approximated by a first-order Taylor series expansion in σ̄ , ε̄ and T


∂ ε̄
˙ ∂ ε̄
˙ ∂ ε̄
˙
ε̄˙n+1 = ε̄˙n + Δtn |n σ̄˙ θ + |n ε̄˙θ + |n Ṫθ (34)
∂ σ̄ ∂ ε̄ ∂T

Assume that the temperature update proceeds first, and Ṫθ comes in handy; based
on plastic consistency condition and constitutive relations, on may find that
 
ε̄˙n ζθ ∂ ε̄˙ /∂ T
ε̄˙θ ≈ + Pθ : Dθ + Ṫθ (35)
1 + ζθ Hθ (1 + ζθ ) ∂ ε̄˙ /∂ σ̄ n

where

Pθ = C : pn (36)
3s̃
pn = (37)
2σ̄
∂ ε̄˙ /∂ ε̄
Hθ ≈ |n + (p : L : p)n (38)
∂ ε̄˙ /∂ σ̄

∂ ε̄˙
ζθ ≈ θ Δt Hθ (39)
∂ σ̄ n
∂ ε̄˙ mε̄˙
= (40)
∂ σ̄ σ̄
∂ ε̄˙ /∂ T σ̄ ∂g
|n = − (41)
∂ ε̄˙ /∂ σ̄ g(ε̄ , T ) ∂ T
278 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Following the rate tangent modulus approach Peirce, Shih, and Needleman (1984),
the objective rate of the Kirchhoff stress can be given as

ε̄˙n
τ ∇ = Ctan
θ : Dθ − Pθ −
1 + ζθ
ζθ ∂ ε̄˙ /∂ T
|n Ṫθ Pθ − α Ṫθ C : I (42)
(1 + ζθ )Hθ ∂ ε̄˙ /∂ σ̄

where

ζ
Ctan
θ = C− Pθ : Pθ (43)
(1 + ζ )H θ

Once the objective rate is obtained, the Kirchhoff stress can then be updated ac-
cording to Equation(31). The corresponding first Piola Kirchhoff stress tensor is
then given as P = F−1 · τ .

4.3 Failure model

Two types of failure models are used

• Stress-based fracture criterion

• Strain-based fracture criterion

For stress-based fracture criterion, fracture is introduced once the maximum prin-
cipal tensile stress exceed 3 times the tensile strength σ0 of the material.
For strain-based fracture criterion, fracture is introduced once the effective plastic
strain rate exceeds a given threshold. This threshold is 0.2 in our studies.
Both criteria required a non-local averaging of stress and strain field.
Fracture is modeled in a very simple manner. We broke the link between neigh-
boring particles once fracture criterion is met. More sophisticated models will be
studied in the future.

5 Results
5.1 Pinched cylinder
We first show the efficiency and accuracy of the method with simple example: Lin-
ear elastic pinched cylinder with rigid end diaphragm. This problem resembles one
of bench mark problems in the so-called standard problem set testing finite ele-
ment accuracy and was tested for example by Sze, Lo, and Yao (2002); Weissmann
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 279

0.9

0.8

Normalized displacement
0.7

0.6
our method
0.5 Sze 2002
Simo 1989
0.4 Weissmann 1996

0.3

0.2

0.1

0
0 5 10 15 20 25 30
Number of elements (background cells for EFG) in each side

(a)
Figure 3: Normalized displacement versus number of elements in each side,
pinched cylinder example

(1996); Bucalem and Bathe (1993); Parish (1991); Simo, Fox, and Rifai (1989).
Only one eights of the cylinder is usually modeled. We model the entire cylinder
since we cannot take advantage of symmetry in the following examples. In order
to compare results with the literature, we have scaled the number of background
cells to the number of elements of the 1/8 cylinder. This normalized displacement
plot is depicted in figure 3. The results are of similar accuracy as results with finite
elements that use the same number of elements though this method is much simpler
and better suited for modeling fracture.

5.2 Cylinder under internal pressure

The configuration of the problem is shown in Figure 4. The cylinder considered
has an axial length of 1.2m. The circular cross-section has a mean radius of 22.575
cm and thickness of 0.15 cm. The mesh for approximately 30,000 nodes is shown
in figure 5a. The finer mesh with approximately 150,000 nodes is shown in figure
5b. The background cells built by 4 nodes are illustrated in this figure as well. We
used standard commercial software to obtain the nodal points and the background
mesh. We also used a mesh with 100,000 nodes and 120,000 nodes.
The initial crack is modeled with the visibility criterion developed Belytschko and
Lu (1995); Belytschko, Lu, and Gu (1995). Thermo-viscoplastic model described
in section 4.2 is used with material parameters: ε̇0 = 0.001s−1, m = 70, n = 0.01,
T0 = 293K, δ = 0.5, κ = 1000K, ρ = 7839kg/m3, c p = 448J/(kgK), χ = 0.9.
Stress-based fracture criterion is used as described in section 4.3. The internal
280 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

1.2m

0.2m

thickness 1.5mm
0.225m

Figure 4: Dimension of the pressurized cylinder problem

(a) 30,000 nodes (b) 150,000 nodes

Figure 5: Structured nodal arrangement and background mesh of the pressurized
cylinder problem; initial configuration
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 281

pressure is increased linearly. We test 2 scenarios:

• slow loading rate

• fast loading rate

For slow loading rate, we perform quasi static analysis and the loading rate is
15.2Pa/s. For fast loading rate, the maximum load of 15.2MPA is reached after
1ms leading to loading rate of 15.2kPa/s.

(a) temperature distribution (b) von Mises stress

(c) Slow load rate; temperautre (d) Slow load rate; temperautre

(e) Fast load rate; von Mises stress

Figure 6: Displaced configuration of the cylinder under internal pressure

The displaced cylinder for slow loading rate is illustrated in figure 6. The crack
propagates straight. For fast loading rate, the crack propagates first straight but
then branches. This was also observed by other authors Zhou, Ravichandran,
and Rosakis (1996); Ravi-Chandar (1998); Sharon, Gross, and Fineberg (1995);
282 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

80

70
30,000 nodes
60 100,000 nodes
120,000 nodes
150,000 nodes

Expansion [mm]
50

40

30

20

10

0
0 0.25 0.5 0.75 1
Time [ms]

Figure 7: Maximum radial displacements for different meshes

Rabczuk and Belytschko (2004, 2007). The maximum radial displacements mea-
sured from the simulation data for different meshes are shown in figure 7. The
results are almost indistinguishable after exceeding 100,000 nodes. This indicates
mesh independent results. The displaced cylinder and the crack patterns are also
indistinguishable and therefore we presented only the displaced cylinder for the
simulation with 100,000 nodes in figure 6.

5.3 Detonation driven fracture of cylinder

This problem was experimentally studied by Chao (2004). The test-setup consists
of a detonation tube of 152cm length to which a thin-walled aluminium tube is at-
tached. The lengths of aluminium tube range from 45.7cm to 89.6cm. The inner
tube radius is 1.975cm and thickness of shell is 0.89mm. While the lower end of
the device is closed, a thin diaphragm seals up the other end. The aluminium tube
contains notches of various lengths midspan of the aluminium tube. The entire ap-
paratus is filled with a combustible mixture of ethylene and oxygen. Initial pressure
varies from 80kPa to 180kPa. The mixture is thermally ignited at the closed end
and the combustion transitions quickly to a detonation. When it enters the test spec-
imen, the detonation is close to the Chapman-Jouguet (CJ) limit of quasi-stationary
self-sustained propagation. Its velocity is between 2300m/s and 2400m/s and the
pressure values in the fully recreated CJ state range from 2.6MPa to 6.1MPa (de-
pending on the initial pressure).
We model this problem with von Mises type J2 visco-plasticity model, section
4.1, and strain-based failure criterion. The material data is: E = 69GPa, ρ =
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 283

5
coarse
fine
4

Pressure [MPa]
3

0
0 0.1 0.2 0.3 0.4 0.5 0.6
Time [ms]

Figure 8: Pressure over the time of a particle close to the notch from a pure fluid
simulation of gas-detonation in rigid specimen

2719kg/m3, ν = 0.33, σ0 = 275GPa, ε0p = 0.001, 1/n = 0.07, 1/m = 0.01. Since
it is complicated to model the gas explosion and fracture of the cylinder at the same
time, we proceed as follows:

• We first simulate the gas explosion at rigid cylinder.

• Then, we prescribe the pressure time history of the previous simulation to

our fracture model.

The gas explosion simulation was done by using equation of state model. The
pressure time-history at the notch location of one particle is shown in figure 8. We
now focus on the fracture of the shell. More details on the gas explosion simulation
is given in Gato (2007).
The cylindrical shells are modeled with up to 280,000 nodes. Using half the number
of nodes resulted in similar numerical data. We considered 2 notch lengths.
For notch length of 1”, the results are shown in figure 9. Cracks propagate shortly
from the crack tips and then bifurcate in circumferential direction. Similar obser-
vations were made in the experiment. The cracks are arrested after propagating
approximately 3/4 of the circumference. In the experiment, the side opposed to
the source of explosions failed around the entire circumference while the other side
did not. The simulation cannot capture this behavior due to the relatively regularly
applied pressure load. A coupled simulation might resolve this issue.
For notch length of 2”, result is shown in figure 10. Since pressure differences
284 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

(a) (b)

(c)
Figure 9: Displaced configuration and effective stress of the detonation-driven frac-
ture of cylinder at different times; notch length is 1”

(a)
Figure 10: Displaced configuration and effective stress of the detonation-driven
fracture of cylinder; notch length is 2”
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 285

around and ahead of the crack tips were more pronounced from the gas explosion
simulations, we captured the basic behavior of the experiment. While crack prop-
agates straight on the RHS, crack curves in 45 degree angle before propagating in
circumferential direction and getting arrested. This behavior was also seen in the
experiment.

6 Conclusions
With the developments of a non-linear meshfree method and a thermo-visco-plastic
material model, we studied the failure of cylinders. The advantages of using a
meshfree method over finite element method are:

• Meshfree method can deal with large deformations more accurately than fi-
nite element method.

• Arbitrary crack growth is represented naturally and independent of the nodal

arrangement because of absence of mesh.

• No complicated shell theory needs to be used that is difficult to implement.

Nevertheless, the results are of the same accuracy for similar amount of
nodes.

We predicted the failure mode of cylinders under internal pressure independent

of the mesh size (after exceeding 100,000 nodes). We modeled detonation-driven
fracture of cylindrical tubes with different notch lengthes. Therefore, internal pres-
sure time histories from pure gas-detonation fluid simulation in rigid tube is used.
For notch length of 2”, experimental fracture pattern can be predicted accurately
while for notch length of 1”, our simulated fracture pattern is different from the
experimental fracture pattern though we were able to predict the basic features, i.e.
crack bifurcation in circumferential direction and short straight crack propagation.
These differences might occur due to neglecting fluid-structure interaction effect.
For short notch, spatial pressure distribution around and ahead the notch is similar
in simulation of pure fluid simulation while spatial differences in pressure distribu-
tion occur for longer notch length. A more sophisticated fracture model might be
another source of inaccuracies. Discrete crack cohesive zone model will improve
accuracy. These aspects will be studied in future research.

References
Andreaus, U.; Batra, B.; Porfiri, M. (2005): Vibrations of cracked euler-
bernoulli beams using meshless local petrov-galerkin (mlpg) method. CMES:
Computer Modeling in Engineering & Sciences, vol. 9, pp. 111–131.
286 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Atluri, S. (2002): The Meshless Local Petrov-Galerkin (MLPG) Method. Tech

Science Press.

Atluri, S.; Cho, J.; Kim, H. (1999): Analysis of thin beams, using the mesh-
less local petrov-galerkin method, with generalized moving least-squares interpo-
lations. Computational Mechanics, vol. 24, pp. 334–347.

Atluri, S.; Shen, S. (2002): The meshless local petrov-galerkin method: a simple
and less-costly alternative to the finite element and boundary element methods.
CMES: Computer Modelling in Engineering and Sciences, vol. 3, pp. 11–51.

Atluri, S.; Zhu, T. (1998): A new meshless local petrov-galerkin (mlpg) approach
in computational mechanics. Computational Mechanics, vol. 22, pp. 117–127.

Atluri, S.; Zhu, T. (2000): The meshless local petrov-galerkin (mlpg) approach
for solving problems in elasto-statics. Computational Mechanics, vol. 25, pp.
169–179.

Belytschko, T.; Lu, Y. (1995): Element-free galerkin methods for static and
dynamic fracture. International Journal of Solids and Structures, vol. 32, pp.
2547–2570.

Belytschko, T.; Lu, Y.; Gu, L. (1994): Element-free galerkin methods. Interna-
tional Journal for Numerical Methods in Engineering, vol. 37, pp. 229–256.

Belytschko, T.; Lu, Y.; Gu, L. (1995): Crack propagation by element-free

galerkin methods. Engineering Fracture Mechanics, vol. 51, no. 2, pp. 295–315.

Bucalem, M.; Bathe, K.-J. (1993): Higher-order mitc general shell elements.
International Journal for Numerical Methods in Engineering, vol. 36, pp. 3729–
3754.

Chao, T. (2004): Gaseous detonation-driven fracture of tubes. PhD thesis,

California Institute of Technology.

Donning, B.; Liu, W. (1998): Meshless methods for shear-deformable beams and
plates. Computer Methods in Applied Mechanics and Engineering, vol. 152, no.
1-2, pp. 47–71.

Fujimoto, T.; Nishioka, T. (2006): Numerical simulation of dynamic elasto

visco-plastic fracture using moving finite element method. CMES: Computer Mod-
eling in Engineering & Sciences, vol. 11, no. 2, pp. 91–101.
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 287

Gao, L.; Liu, K.; Liu, Y. (2006): Applications of mlpg method in dynamic
fracture problems. CMES: Computer Modeling in Engineering & Sciences, vol.
12, no. 3, pp. 181–195.

Gato, C. (2007): Numerical analysis of cylindrical shell fracture. PhD thesis,

Harbin Engineering University.

Guo, Y.; Nairn, J. (2006): Three-dimensional dynamic fracture analysis using the
material point method. CMES: Computer Modeling in Engineering & Sciences,
vol. 16, no. 3, pp. 141–155.

Guz, A.; Menshykov, O.; Zozulya, V. (2007): Contact problem for the flat
elliptical crack under normally incident shear wave. CMES: Computer Modeling
in Engineering & Sciences, vol. 17, no. 3, pp. 205–214.

Hagihara, S.; Tsunori, M.; Ikeda, T. (2007): Application of meshfree method to

elastic-plastic fracture mechanics parameter analysis. CMES: Computer Modeling
in Engineering & Sciences, vol. 17, no. 2, pp. 63–72.

Han, Z.; Atluri, S. (2003): Truly meshless local petrov-galerkin (mlpg) solutions
of traction and displacement bies. CMES: Computer Modelling in Engineering
and Sciences, vol. 4, pp. 665–678.

Hao, S.; Liu, W. (1999): Bimaterial interfacial crack growth with strain gradient
theory. Journal of Engineering Materials and Technology-Transactions of the
ASME, vol. 121, pp. 413–421.

Hao, S.; Liu, W. (2006): Moving particle finite element method with supercon-
vergence: Nodal integration formulation and applications. Computer Methods in
Applied Mechanics and Engineering, vol. 195, pp. 6059–6072.

Hao, S.; Liu, W.; Belytschko, T. (2004): Moving particle finite element method
with global smoothness. International Journal for Numerical Methods in Engi-
neering, vol. 59, pp. 1007–1020.

Hao, S.; Liu, W.; Chang, C. (2000): Computer implementation of damage

models by finite element and meshfree methods. Computer Methods in Applied
Mechanics and Engineering, vol. 187, pp. 401–440.

Hao, s.; Liu, W.; Klein, P.; Rosakis, A. (2004): Modeling and simulation of
intersonic crack growth. International Journal of Solids and Structures, vol. 41,
no. 7, pp. 1773–1799.
288 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Hao, S.; Liu, W.; Moran, B.; Olson, G. (2003): A hierarchical multi-physics
model for design of high toughness steels. Journal of Computer-Aided Materials
Design, vol. 10, pp. 99–142.

Hao, S.; Liu, W.; Moran, B.; Vernerey, F.; Olson, G. (2004): Multi-scale con-
stitutive model and computational framework for the design of ultra-high strength,
high toughness steels. Computer Methods in Applied Mechanics and Engineering,
vol. 193, pp. 1865–1908.

Hao, S.; Liu, W.; Weertman, J. (2004): Cohesive solutions of intersonic moving
dislocations. Philisophical Magazine, vol. 84, pp. 1067–1104.

Jarak, T.; Soric, J.; Hostler, H. (2007): Analysis of shell deformation repsonses
by the meshless local petrov-galerkin (mlpg) approach. CMES: Computer Model-
ing in Engineering & Sciences, vol. 18, no. 3, pp. 235–246.

Jiawei, X.; Xuefeng, C.; Lianfa, Y. (2008): A class of wavelet-based flat shell
elements using b-spline wavelet on the interval and its applications. CMES: Com-
puter Modeling in Engineering & Sciences, vol. 23, no. 1, pp. 1–12.

Ju, S.; Liu, S. (2007): Determining stress intensity factors of composites using
crack opening displacement. Composite Structures, vol. 81, pp. 614–621.

Kanok-Nukulchai, W.; Barry, W.; Saran-Yasoontorn, K.; Bouillards, P.

(2001): On elimination of shear locking in the element-free galerkin method.
International Journal for Numerical Methods in Engineering, vol. 52, pp. 705–
725.

Krysl, P.; Belytschko, T. (1996): Analysis of thin shells by the element-free

galerkin method. International Journal for Numerical Methods in ENgineering,
vol. 33, pp. 3057–3078.

Lancaster, P.; Salkauskas, K. (1981): Surfaces generated by moving least

squares methods. Mathematics and Computation, vol. 37, pp. 141–158.

Le, P.; Mai-Duy, N.; Tran-Cong, T. (2008): A meshless modeling of dynamic

strain localization in quasi-brittle materials using radial basis function networks.
CMES: Computer Modeling in Engineering & Sciences, vol. 25, no. 1, pp. 43–67.

Leitao, V. (2001): A meshless method for kirchhoff plate bending problems.

International Journal for Numerical Methods in Engineering, vol. 52, pp. 1197–
1130.
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 289

Li, Q.; Soric, J.; Jarak, T.; Atluri, S. (2005): A locking-free meshless local
petrov-galerkin (mlpg) formulation for thick and thin plates. Journal of Computa-
tional Physics, vol. 208, no. 1, pp. 116–133.

Li, S.; Hao, W.; Liu, W. (2000): Numerical simulations of large deformation of
thin shell structures using meshfree methods. Computational Mechanics, vol. 25,
pp. 102–116.

Liew, K.; Chen, X. (2004): Buckling of rectangular mindlin plates subjected to

partial in-plane edge loads using the radial point interpolation method. Interna-
tional Journal of Solids and Structures, vol. 41, pp. 1677–1695.

Liew, K.; Chen, X. (2004): Mesh-free radial point interpolation method for the
buckling analysis of mindlin plates. International Journal for Numerical Methods
in Engineering, vol. 60, pp. 1961–1877.

Liu, H.; Han, Z.; Rajendran, A.; Atluri, S. (2006): Computational modeling of
impact response with the rg damage model and the meshless local petrov-galerkin
(mlpg) approaches. CMC: Computers Materials and Continua, vol. 4, pp. 43–53.

Liu, W.; Jun, S.; Zhang, Y. (1995): Reproducing kernel particle methods. Inter-
national Journal for Numerical Methods in Engineering, vol. 20, pp. 1081–1106.

Liu, Y.; Liu, L.; Mahadevan, S. (2007): Analysis of subsurface crack propa-
gation under rolling contact loading in railroad wheels using fem. Engineering
Fracture Mechanics, vol. 17, pp. 2659–2674.

Long, S.; Atluri, S. (2002): A meshless local petrov-galerkin method for solving
the bending problem of a thin plate. CMES: Computer Modeling in Engineering
and Sciences, vol. 3, pp. 53–63.

Ma, J.; Lu, H.; Wang, B. (2006): Multiscale simulation using generalized in-
terpolation material point (gimp) method and molecular dynamics (md). CMES:
Computer Modeling in Engineering & Sciences, vol. 14, no. 2, pp. 101–117.

Nguyen-Van, H.; N, N. M.-D.; Tran-Cong, T. (2008): A smoothed four-node

piezoelectric element for analysis of two-dimensional smart structures. CMES:
Computer Modeling in Engineering & Sciences, vol. 23, no. 3, pp. 209–222.

Nishioka, T. (2005): Recent advances in numerical simulation technologies for

various dynamic fracture phenomena. CMES: Computer Modeling in Engineering
& Sciences, vol. 10, no. 3, pp. 209–215.
290 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Nishioka, T.; Kobayashi, Y.; Fujimoto, T. (2007): The moving finite element
method based on delaunay automatic triangulation for fracture path prediction sim-
ulations in nonlinear elastic-plastic materials. CMES: Computer Modeling in En-
gineering & Sciences, vol. 17, no. 3, pp. 231–238.

Noguchi, H.; Kawashima, T.; Miyamura, T. (2000): Element free analysis

of shell and spatial structures. International Journal for Numerical Methods in
Engineering, vol. 47, pp. 1215–1240.

Parish, H. (1991): An investigation of a finite rotation four node assumed strain

shell element. International Journal for Numerical Methods in Engineering, vol.
31, pp. 127–150.

Peirce, D.; Shih, C.; Needleman, A. (1984): A tangent modulus method for rate
dependent solids. Computers and Structures, vol. 18, pp. 875–887.

Rabczuk, T.; Areias, P. (2006): A meshfree thin shell for arbitrary evolving
cracks based on an extrinsic basis. CMES: Computer Modeling in Engineering &
Sciences, vol. 16, no. 2, pp. 115–130.

Rabczuk, T.; Areias, P.; Belytschko, T. (2007): A meshfree thin shell method
for non-linear dynamic fracture. International Journal for Numerical Methods in
Engineering, vol. 72, no. 5, pp. 524–548.

Rabczuk, T.; Belytschko, T. (2004): Cracking particles: A simplified mesh-

free method for arbitrary evolving cracks. International Journal for Numerical
Methods in Engineering, vol. 61, no. 13, pp. 2316–2343.

Rabczuk, T.; Belytschko, T. (2005): Adaptivity for structured meshfree particle

methods in 2d and 3d. International Journal for Numerical Methods in Engineer-
ing, vol. 63, no. 11, pp. 1559–1582.

Rabczuk, T.; Belytschko, T. (2006): Application of particle methods to static

fracture of reinforced concrete structures. International Journal of Fracture, vol.
137, no. 1-4, pp. 19–49.

Rabczuk, T.; Belytschko, T. (2007): A three-dimensional large deformation

meshfree method for arbitrary evolving cracks. Computer Methods in Applied
Mechanics and Engineering, vol. 196, no. 29-30, pp. 2777–2799.

Rabczuk, T.; Belytschko, T.; Xiao, S. (2004): Stable particle methods based on
lagrangian kernels. Computer Methods in Applied Mechanics and Engineering,
vol. 193, pp. 1035–1063.
Numerical Simulations of Dynamic Fracture in Thin Shell Structures 291

Rabczuk, T.; PMA, P. A.; Belytschko, T. (2007): A simplified mesh-free method

for shear bands with cohesive surfaces. International Journal for Numerical Meth-
ods in Engineering, vol. 69, no. 5, pp. 993–1021.

Rabczuk, T.; Zi, G. (2007): A meshfree method based on the local partition of
unity for cohesive cracks. Computational Mechanics, vol. 39, no. 6, pp. 743–760.

Ravi-Chandar, K. (1998): Dynamic fracture of nominally brittle materials. In-

ternational Journal of Fracture, vol. 90, pp. 83–102.

Rong, F.; Huang, C.; Liu, Z.; Song, D.; Wang, Q. (2008): Microstructure
changes in the catalyst layers of pem fuel cells induced by load cycling: Part i.
mechanical model. Journal of Power Sources, vol. 175, pp. 699–711.

Sharon, E.; Gross, P.; Fineberg, J. (1995): Local crack branching as a mech-
anism for instability in dynamic fracture. Physical Review Letters, vol. 74, pp.
5096–5099.

Simo, J.; Fox, D.; Rifai, M. (1989): On a stress resultant geometrically exact
shell model. part ii: The linear theory; computational aspects. Computer Methods
in Applied Mechanics and Engineering, vol. 73, pp. 53–92.

Sladek, J.; Sladek, V.; Solek, P. (2008): Thermal bending of reissner-mindlin

plates by the mlpg. CMES: Computer Modeling in Engineering & Sciences, vol.
28, no. 1, pp. 57–76.

Sladek, J.; Sladek, V.; Wen, P.; Aliabadi, M. (2006): Meshless local petrov-
galerkin method for shear deformable shell analysis. CMES: Computer Modeling
in Engineering & Sciences, vol. 13, no. 2, pp. 103–117.

Soric, J.; Li, Q.; Jarak, T.; Atluri, S. (2004): Meshless local petrov-galerkin
(mlpg) formulation for analysis of thick plates. CMES: Computer Modeling in
Engineering & Sciences, vol. 6, no. 4, pp. 349–357.

Sze, K.; Lo, S.; Yao, L. (2002): Hybrid-stress solid elements for shell structures
based upon a modified variational functional. International Journal for Numerical
Methods in Engineering, vol. 53, pp. 2617–2642.

Tang, Z.; Shen, S.; Atluri, N. (2003): Analysis of materials with strain-gradient
effects: A meshless local petrov-galerkin(mlpg) approach, with nodal displace-
ments only. CMES: Computer Modelling in Engineering and Sciences, vol. 4, pp.
177–196.
292 Copyright © 2008 Tech Science Press CMES, vol.33, no.3, pp.269-292, 2008

Wang, D.; Chen, J. (2006): A locking-free meshfree curved beam formulation

with the stabilized conforming nodal integration. Computational Mechanics, vol.
39, no. 1, pp. 83–90.

Weissmann, S. (1996): High-accuracy low-order three-dimensional brick ele-

ment. International Journal for Numerical Methods in Engineering, vol. 39, pp.
2337–2361.

Yagawa, G.; Miyamura, T. (2005): 3-node triangular shell element using mixed
formulation and its implementation by free mesh method. Computers and Struc-
tures, vol. 83, pp. 2066–2076.

Zhao, J.; Liu, R.; Wu, X. (2008): Effects of partial crack-face contact for the
bending of thin shell structures. Theoretical and Applied Fracture Mechanics, vol.
49, pp. 128–150.

Zhou, M.; Ravichandran, G.; Rosakis, A. (1996): Dynamically propagating

shear bands in impact-loaded prenotched plates, 2. numerical simulations. Journal
of the Mechanics and Physics of Solids, vol. 44, no. 6, pp. 1007–1032.