Products: ABAQUS/Standard ABAQUS/CAE ABAQUS/AMS
Benefits: The AMS eigensolver offers significant performance advantages over the default Lanczos method, particularly when a large number of eigenmodes is required for a system with many degrees of freedom. Benchmark tests at ABAQUS, Inc., indicate that natural frequency extraction problems can be solved anywhere between 10 times and 25 times faster with the new AMS eigensolver, depending on the complexity of the model.
Description: When you are performing natural frequency extraction, you can now select the automatic multi-level substructuring (AMS) eigensolver. The AMS eigensolver is provided with ABAQUS/AMS, an add-on analysis capability for ABAQUS/Standard.
The Lanczos method remains the default eigenvalue extraction method because it has more general capabilities; however, the AMS method is generally faster than the Lanczos method, particularly when you extract a large number of eigenmodes for a system with many degrees of freedom. The AMS method currently has some limitations, which are listed in detail in “Natural frequency extraction,” Section 6.3.5 of the ABAQUS Analysis User's Manual. For example,
Only a single step using the AMS method can be used in an analysis.
If an analysis includes a step that uses the AMS method, you cannot include any other natural frequency extraction steps in the same analysis.
The AMS eigensolver provides only nodal output; you cannot request element output variables, such as stress.
As an example of the performance gains of the AMS eigensolver, the NVH truck body problem shown in Figure 6–10 was tested with the AMS eigensolver and the Lanczos method. The AMS eigensolver solved the problem in 600 seconds; the Lanczos method solved it in 14000 seconds. The problem has the following characteristics:
127000 shell elements
Approximately 800000 equations
Approximately 1000 natural frequencies are extracted
Step module: StepCreate: Frequency: Basic: Eigensolver: AMS
