Reliability-Based Seismic Optimization of Steel Frames by Metaheuristics and Neural Networks

The main aim of the present study is to propose an efficient methodology for tackling reliability-based seismic design optimization problems of steel moment-resisting frames incorporating the concepts of performance-based design. A serial integration of particle swarm optimization (PSO) and bat algorithm (BA), termed as PSO-BA metaheuristic, is proposed as the optimizer of this study. The Monte Carlo simulation (MCS) method is employed to evaluate the reliability constraints during the optimization process. As the reliability analysis by the means of MCS requires a long computational time, wavelet back-propagation (WBP) neural networks are trained to predict the required deterministic and probabilistic structural nonlinear seismic responses at performance levels. In order to investigate the computational merits of the proposed methodology, two numerical examples of steel moment frames are presented and optimal probabilistic designs found by metaheuristics are compared. The numerical results indicate that the proposed PSO-BA has better computational performance in comparison with both PSO and BA metaheuristics. (C) 2016 American Society of Civil Engineers.