Research On Damage Identification Method Of Environmental Vibration Bridge Based On CPSO-BP

With the rapid development of China's transportation industry,the number of new and old bridges continues to increase.Understand the health status of the bridge under working conditions,and quickly and effectively identify the damage degree of the bridge structure.It not only can guarantee the health and safety of people's lives,but also has great significance for the regulation of transportation hubs.Today's bridge health monitoring and damage identification has become a hot issue in the field of bridge engineering research.Current bridge damage detection methods are visual or local experimental methods.These experimental techniques require prior inspection of damage near the bridge structure and can only inspect structural parts that the engineer can access.Subject to these limitations,these experimental methods can only detect damage on or near the surface of the bridge structure.Based on a large number of researches on the damage identification of bridge structures at home and abroad,this paper establishes a finite element bridge model,combines the theory of vibration modal analysis to establish a neural network training bridge damage identification model and simultaneously determines the damage location and damage degree of the bridge structure,and establishes an environment-based excitation.Vibration modal analysis theory and neural network bridge damage identification methods to monitor the health of bridges.In this paper,the environnental incentive data of bridge health monitoring is the main research object.According to the current situation of bridge structure health monitoring system,according to the new form of bridge monitoring index requirements,the intelligent algorithm is used to solve the structural dynamic response data under natural environment vibration with small amplitude and randomness.Problems such as strong sex,large amount of data,and disordered noise data in the process of transmission.The existing environmental vibration test data processing method can only obtain the basic modal parameters of the structure,and can not effectively support the maintenance management and security decision of the structure.This paper adopts the method of identifying the deep bridge damage from the environmental vibration test data..The vibration data of bridge monitoring is very important for structural safety diagnosis and evaluation.The compressed sensing theory is applied to the collected data to remove the clutter noise,so that the processed data can be applied in the damage identification of the bridge.The damage identification model of environmental vibration data is mainly established by BP neural network.However,BP neural network has the disadvantages of slow convergence speed,initial weight sensitivity,easy to fall into local extremum,and disordered environmental vibration data.Only the BP neural network can not establish an effective damage recognition model.Therefore,we introduce the chaotic particle swarm optimization algorithm to optimize the environmental vibration data.Combined with the characteristics of chaos theory,we can effectively make up for the shortcomings of BP neural network.In summary,the BP neural network algorithm based on chaotic particle swarm optimization is used to analyze the damage of the model.The simulation results show that the proposed method can improve the monitoring efficiency.The model has practicality in bridge structure monitoring and has a theory for large bridge structural damage research.Meaning and practical application value.