Structural Optimization for Earthquake Loading with Nonlinear Responses by Surrogate Modeling Based Evolutionary Algorithms

An efficient methodology is proposed to design optimization of structures subjected to earthquake time history loading considering nonlinear structural response. It is clear that the structural optimization for transient time history loading is a computationally intensive task, especially when the nonlinear response is concerned. In the proposed hybrid methodology particle swarm optimization (PSG), genetic algorithm (GA), probabilistic neural network (PNN), radial basis function neural network (RBFNN) and wavelet transforms (WT) techniques are combined to achieve the optimization task. In order to investigate the efficiency of the proposed methodology, a 72-bar space steel tower is designed for optimal weight for EI Centro earthquake. The numerical results demonstrate the efficiency and computational advantages ofthe proposed methodology.