Research On Bridge Risk Assessment Algorithms And Health Monitoring Software

With the rapid growth of the construction speed of basic transportation facilities in China,China has become a large traffic country,but its informationization degree is far from keeping up with the growth of infrastructure construction speed,which leads to the low degree of automation of bridge operation and maintenance,and the difficulty of bridge state evaluation,real-time monitoring of bridges and integrated information management.The traditional method of regular bridge inspection and monitoring can not meet the needs of supervision and maintenance,therefore,the development of bridge risk assessment algorithm and bridge health monitoring software in one of the bridge health monitoring system is imminent.Based on bridge information and hardware data returned by sensors,the paper establishes the bridge risk assessment model and evaluation algorithm by bridge risk source identification,Grey Relation Analysis,Fuzzy Analytic Hierarchy Analysis,Wolf Group Algorithm and BP neural network,designs and implements bridge monitoring software based on Android application,web platform,server and database,and completes the Bridge Health Monitoring system,which has very important practical value for realizing the dynamic real-time supervision of the bridge.In the process of establishing the Bridge Health Monitoring System,firstly,the risk sources affecting the operation of the bridge are identified,the bridge risk index system is established,the comprehensive correlation degree between the risk indexes is determined by Grey Relation Analysis,the final index weight is determined by combining the Fuzzy Analytic Hierarchy Analysis method and the weighted summation method,the bridge composite index is obtained in each month,the safety degree of the bridge is evaluated,and the accuracy of the evaluation results of bridge risk evaluation model is verified by comparing with the actual bridge state.Secondly,through the comparison between BP neural network to predict bridge state and actual bridge state,and the analysis of BP neural network processing performance,the BP neural network is optimized by using Wolf Group Algorithm(WPA),and the structure of wolf swarm algorithm is improved.The improved Wolf Group Algorithm is established to optimize the bridge risk assessment algorithm of BP Neural network,andthrough the comparison between the WPA-BP algorithm and the improved WPA-BP algorithm,the good learning and prediction ability of the improved WPA-BP algorithm is verified,and compared with the results of the bridge risk evaluation model,the accuracy of the algorithm processing results is verified again.Finally,aiming at the feasibility analysis,functional requirement analysis and non-functional requirement analysis results of developing bridge health monitoring software,the technical structure and network structure of bridge health monitoring software are established,the design and code realization of the functions of each module of the software are completed,and the final test of the software is carried out,and the function and performance of the software are verified.