Identifying Prestress Loss of Long Span Bridge Based on Genetic Algorithm Combined with Neural Network

In recent years, computational intelligence methods are widely used to solve problems in engineering structural field by more and more researchers. In this paper, a method based on combining artificial neural networks (ANN) and genetic algorithm (GA) in identifying the prestress loss of the long span bridge was proposed. A model with some parameters was constructed to access the prestress loss. Based on the differences of the bridge deck elevation of the long span bridge in service between damaged structure and intact structure, the model parameters were identified by the genetic algorithm, and the prestress loss values of different time were obtained. At the same time, the differences of responses were constructed within BP neural networks. In fact, the trained BP neural networks were used as a subroutine in the process of identification using GA. It is demonstrated that the method combining BP neural networks with GA of identification for prestress loss is efficient and the identified results fit well with the actual value.