This example verifies the application of a standard
rate-independent plasticity theory for metals and assesses the accuracy of the
integration of the plasticity equations, especially in the case of
nonproportional stressing.
Integration of elastic-plastic
material models is a potential source of error in numerical structural
analysis. See, for example, the discussions by Krieg and Krieg (1977) and
Schreyer et al. (1979). Usually the error is most severe when kinematic
hardening is used in plane stress with nonproportional stressing (perhaps
because of the complexity of the motion of the stress point and yield surface
in stress space in this theory). This example contains two such problems for
which the exact solutions are available (Foster Wheeler report, 1972).
Experience with a number of other computer programs has suggested that the
second example, in particular, is a severe test of the numerical implementation
of the plasticity theory. Both problems involve states of uniform plane stress
and, hence, are done here by using a single plane stress element.
The material models for the unixially and biaxially loaded cases are
described below.
Case 1—Uniaxial loading
Figure 1
shows the material model for this case. The elastic modulus is 68.94 GPa (10.0
× 106 lb/in2), the yield stress is 68.9 MPa (10.0 ×
103 lb/in2), and the work hardening slope is 68.9 GPa
(10.0 × 106 lb/in2). This is specified by giving a yield
stress of 34.57 GPa (5.01 × 106 lb/in2) at a plastic
strain of 0.5. The total force and the total moment on the loaded face of the
model are output to the results file.
Case 2—Biaxial loading
Figure 1
shows the material model for this case. The elastic modulus is 207 GPa (30.0 ×
106 lb/in2), the yield stress is 207 MPa (30.0 ×
103 lb/in2), and the work hardening slope is 11 GPa (1.59
× 106 lb/in2). This is specified by giving a yield stress
of 10.62 GPa (1.53 × 106 lb/in2) at a plastic strain of
0.95.
Model and loading
The geometries and loading distributions for the unixial and biaxial cases
are described below.
Case 1—Uniaxial loading
Figure 1
shows the geometry for this case. Two types of meshes are provided: a
single-element mesh using higher-order plane stress and shell elements (CPS8R, S8R5, S9R5, and STRI65) and a mesh using linear shell and continuum shell elements (S4R and SC8R). Two edges have simple support. The load history is shown in
Figure 2
and is prescribed with an amplitude curve (Amplitude Curves).
The load distribution is a uniform, direct stress on the element edge. Since
the strain should be uniform, the edge nodes are constrained using an equation
constraint (Linear Constraint Equations)
to move together in the direction normal to the edge. Then the total load on
the edge is simply given on one of the edge nodes.
Case 2—Biaxial loading
The case is set up with the same geometric model (Figure 1).
However, the loading is more complex (see
Figure 2).
First, the plate is loaded into the plastic range in uniaxial tension in the
x-direction, unloaded slightly, and reloaded. Biaxial
loading then follows, with
and
prescribed, as shown in
Figure 2,
so that the quantity
remains constant at 276 MPa (40000 lb/in2). This loading is defined
by an amplitude curve by reading in a file of values previously calculated in
the small program AMP (see
elasticplasticplate_amplitude.f).
Results and discussion
Exact solutions for these two problems have been developed by Chern in a
Foster Wheeler report (1972), where they are documented as Problems 8 and 9.
These solutions provide a basis for the comparison of the
Abaqus
results.
Case 1—Uniaxial loading
The plastic strains are the basic solution in these cases (since stress is
prescribed). The results for this case are summarized in
Table 1.
The
Abaqus
results agree with the exact solution.
Table 1
also records the number of iterations required to achieve equilibrium.
Case 2—Biaxial loading
The results in this case are best represented by the
versus
plot shown in
Figure 3.
The agreement with the exact solution is again very close.
Foster Wheeler
Corporation, “Intermediate Heat Exchanger for Fast Flux
Test Facility: Evaluation of the Inelastic Computer
Programs,” report prepared for
Westinghouse ARD, Foster Wheeler Corporation, Livingston, NJ,
1972.
Krieg, R.D., and D. B. Krieg, “Accuracies
of Numerical Solution Methods for the Elastic-Perfectly Plastic
Model,” ASME Journal of Pressure Vessel
Technology, vol. 99, no. 4, pp. 510–515, 1977.
Schreyer, H.L., R. F. Kulak, and J. M. Kramer, “Accurate
Numerical Solutions for Elastic-Plastic
Models,” ASME Journal of Pressure Vessel
Technology, vol. 101, no. 3, pp. 226–234, 1979.