Optimization is a capability of MSC.NASTRAN that enables
you to synthesize and optimize your designs. Sensitivity
analysis is also available in Optimization.
Analysis
Types: Optimization may be applied to the following
analysis types
Basic Analysis
Linear statics analysis
Normal modes analysis
Buckling analysis
Dynamic Response
Modal frequency response
Direct frequency response
Modal transient response
Acoustic analysis
Design
Variable Parameters:
Shape variables are tied to the geometry
Sizing variables are tied to finite element properties
Shape and sizing variables can both be used in the same
run
Large
models: Efficient optimization algorithms permit the use
of hundreds of design variables and responses for large models.
Efficiencies include:
Design variable linking - multiple design variables can
be linked
Approximation methods - three methods are provided
Robust optimization algorithms - three methods are provided
Constraint deletion and regionalization - only the critical
constraints retained
Restarts - optimization restarts from a completed design
cycle and continued
Adjustable convergence criteria and move limits - for
faster convergence
Sparse matrix solver - faster speed and minimal disk
space
Adjoint Sensitivity Analysis
Mode Tracking
Response
Types: Either design objective or the performance constraints