Research on Software Design of Monitoring System Based on BP Neural Network

A pier monitoring model based on GA-MIV-BP neural network is designed to solve the problem that the traditional pier monitoring method has low accuracy in predicting current scour depth, which leads to the deterioration of bridge safety. Firstly, genetic algorithm (GA) was used to optimize the BP neural network to construct the GA-BP neural network model. Then, mean impact value (MIV) was added based on the constructed model, and the input factors of the network model were screened by MIV. Finally, the pier scour depth was accurately predicted. The results show that the prediction correlation coefficient of the GA-MIV-BP model is 98.7%, which is 4.1% and 1.2% higher than that of the traditional BP model and the MIV-BP model, respectively. It shows that the model can realize BP neural network optimization after adding MIV factor screening and GA algorithm, and the prediction accuracy of current scour is significantly improved, which further proves that the model meets the software design requirements of the monitoring system.