SOFT COMPUTING BASED STRUCTURAL OPTIMIZATION FOR EARTHQUAKE

An efficient methodology is proposed to optimal design of structures for earthquake loading. In this methodology to reduce the optimization overall time, a serial integration of wavelet transforms, neural networks and evolutionary algorithms are employed. In order to reduce the computational work of the structural time history analysis, a discrete wavelet transform is used by means of which the number of points in the earthquake record is decreased. Also, an advanced metamodel, called self-organizing generalized regression is employed to predict the time history responses. The optimization task is achieved by an evolutionary algorithm called virtual sub population method. A 6-storey space steel frame structure is designed for optimal weight for the El Centro earthquake induced loads. The numerical results demonstrate the efficiency and computational advantages of the proposed methodology.