Application Of BP Neural Network Based On Genetic Algorithm Optimization In Bridge Health Monitoring And Safety Evaluation

With the rapid development of our country’s economy and the continuous acceleration of the country’s urbanization process,the number and level of bridge construction has reached the world’s first-class level.However,with the increase in the number of vehicles per capita and the inadequate implementation of bridge inspection and maintenance work,bridge safety accidents have occurred.The safety status of the bridge structure is usually monitored through the bridge health monitoring system,traditional bridge health monitoring system is expensive to build,can only be installed on large bridges,as a result,a large number of small and medium-sized bridges cannot be monitored in time.Therefore,a simple and effective method is needed to assess the safety of bridges,early warning of the safety of the bridge structure,to provide reference and suggestions for the maintenance of the bridge.This article is based on the genetic algorithm optimizing the BP neural network algorithm,data mining on the massive data of bridge health monitoring,and carry out the bridge structure safety warning,carry out safety assessment of the bridge structure.Taking a concrete-filled steel tube arch bridge as a background project,establish a bridge health monitoring safety assessment method based on genetic algorithm to optimize BP neural network.This article summarizes the development and core ideas of genetic algorithm and BP neural network respectively,according to the massive data of the health monitoring system,determine the appropriate genetic algorithm model and BP neural network training network structure,realize the combination of the two use Matlab and internal toolbox functions.Secondly,establish a finite element model based on the relevant engineering data of a concrete-filled steel tube arch bridge,and designed and completed the health monitoring system of the bridge,determine the displacement,stress,and cable force monitoring subsystems of the health monitoring system.After that,the arch bridge safety assessment model based on the analytic hierarchy process and the comprehensive principle of variable weights was established,according to the static index,dynamic index and load index,the initial weight of the bottom monitoring index is allocated.The safety evaluation model is divided into four levels of evaluation standards,and each evaluation index is processed in a dimensionless manner.Finally,excavate and analyze the massive data obtained by the concrete-filled steel tube health monitoring system,the genetic algorithm is used to optimize the BP neural network prediction method to predict the bridge and compare and analyze it with the measured data,verify the feasibility of genetic algorithm to optimize the BP neural network prediction method,the results show that this method can effectively predict the changes of various indicators during the operation of the bridge.Use the measured data of the monitoring system and the numerical results of the finite element model to establish a three-level evaluation standard for safety assessment,the genetic algorithm optimizes the prediction results of the BP neural network for bridge safety assessment.