A Study on the Application of GA-BP Neural Network in the Bridge Reliability Assessment

In the design of the bridge reliability assessment proposal, the application of the BP neural network model can help overcome some shortcomings in the traditional bridge reliability assessment, such as the poor model adaptability, the low calculation efficiency and so on. However, there are also some problems in the BP neural network model, for example, the uneasily determinable initial weights and easily local optimum. The BP neural network model optimized by the application of GA global optimum characteristics can get the optimum relation quickly. Therefore, this paper puts forward a GA-BP neural network model based on the real number coding system to analyze the bridge reliability assessment and applies it to the reliability assessment in Masangxi Yangtze River Bridge. The results of the application show that the GA-BP neural network model has a higher accuracy, higher reliability and application value in the assessment of the engineering works than the traditional BP neural network model.