Reliability-Based Design Optimization of a Large-Scale Truss Structure Using Polynomial Chaos Expansion Metamodel

Optimal design of engineering systems considering uncertainties is broadly dealt in the area of reliability-based design optimization (RBDO). These RBDO problems involve a costly inner loop reliability estimate involving a computation-intensive true model solver. The current work is aimed at reducing this cost of computation in RBDO. To this end, a polynomial chaos expansion (PCE) metamodel is used. This PCE metamodel is later used to substitute the actual expensive true model in the reliability computation phase. A stochastic optimizer-particle swarm optimization (PSO) is used for the outer optimization loop. The effectiveness of PCE metamodel combined with PSO is demonstrated for a large-scale truss structural system in reducing the overall computation cost involved in RBDO.