ROBUST OPTIMIZATION OF TRUSS STRUCTURES TO MINIMIZE STRESS VARIATIONS CAUSED BY UNCERTAIN LOADS

Abstract

This paper discusses a multi-objective optimization problem with robustness in truss structures subjected to uncertain loads. Variations in characteristic values due to uncertain structural parameters are calculated by sensitivity analysis. The displacement, strain energy, and mean compliance in the structure are then often selected for the characteristic values. In this study, the stress variation caused by uncertain loads was chosen as the characteristic value to solve a robust optimization problem. In allowable stress calculations in structural design, it is necessary to ensure that the stresses are within the allowable stress. Therefore, it is reasonable to choose the variation of stress as the characteristic value. Since the objective function in this multi-objective optimization problem is set to the stress variation and the mass of the truss, the two objective functions are in a trade-off relationship, and it is necessary to find a set of Pareto optimal solutions. This set of Pareto optimal solutions is obtained using mPSO with a slight modification of the particle trajectories in Particle Swarm Optimization (PSO).