A robust method for optimization of semi-rigid steel frames subject to seismic loading

In this study, we develop a seismic optimization method to minimize the semi-rigid steel frame cost. In the proposed method, cross-sections of columns and beams and types of beam-to-column and base restraint semi-rigid joints are considered as the design variables of the optimization. The nonlinear seismic behaviors of the structure are carried out by using plastic-hinge beam-column elements for beams and columns, zero-length elements for semi-rigid connections, and time-history dynamic analysis. An effective implementation of harmony search technique (HS) is presented to find the global optimal solution of the optimization. In order to improve HS, a multi-comparison technique (MCT) is proposed that significantly reduces the useless time-consuming evaluations in the optimization. The robustness and efficiency of the proposed method are demonstrated through three optimization problems of semi-rigid steel frames. Compared with particle swarm optimization (PSO), micro-genetic algorithm (micro-GA), and genetic algorithm (GA), the proposed method is found to significantly reduce the number of structural analyses required and yield the better optimum frame designs.