An artificial neural network based genetic algorithm for estimating the reliability of long span suspension bridges

An accurate and efficient artificial neural network (ANN) based genetic algorithm (GA) is presented for estimating the reliability of long span suspension bridges. In this method, the training datasets for establishing an ANN model are generated by uniform design method and are distributed uniformly over the entire design space. The explicit formulation of the approximate limit state function is then derived by using the parameters of the developed ANN model. Once the explicit limit state function is obtained, the failure probability can be easily estimated by using an improved GA that introduces new approach for the penalty function and coded method and GA operators. A numerical example involving a detailed computational model of a long span suspension bridge with a main span of 1108 m is presented to demonstrate the applicability and merits of the present method. Finally, several important parameters in the present method are discussed.