Structural optimization with frequency constraints by genetic algorithm using wavelet radial basis function neural network

In this study, a combination of genetic algorithm (GA) and neural networks (NN) is proposed to find the optimal weight of structures subject to multiple natural frequency constraints. The optimization is carried out by an evolutionary algorithm using discrete design variables. The evolutionary algorithm employed in this investigation is virtual sub-population (VSP) method. To reduce the computational time of optimization process, the natural frequencies of structures are evaluated using properly trained radial basis function (RBF) and wavelet radial basis function (WRBF) neural networks. In the WRBF neural network, the activation function of hidden layer neurons is substituted with a type of wavelet functions. In this new network, the position and dilation of the wavelet are fixed and only the weights are optimized. The numerical results demonstrate the robustness and high performance of the suggested methods for structural optimization with frequency constraints. It is found that the best results are obtained by VSP method using WRBF network.