Method for identifying damage position and damage degree of truss bridge based on neural network

Abstract

The invention discloses a method for identifying damage position and damage degree of truss bridge based on neural network. The method comprises the steps of constructing the bridge sample data, usinga finite element method to establish a bridge model, and obtaining the simulated frequency data of a bridge in good condition and under different damage conditions; calculating the frequency change ratio, frequency square change ratio and frequency change rate according to the simulated frequency data of the bridge in good condition and under different damage conditions; determining the topologyof the network, using a BP neural network to determine the number of neurons in each layer and initialize the weights and thresholds of the neural network at the same time; training the neural networkby the elastic gradient method combined with a genetic algorithm and testing by test samples, and obtaining the trained neural network; inputting the real-time frequency data of the bridge into the trained neural network to realize the identification of the damage location and damage degree of the bridge. The method of the invention has the advantages of fast calculation speed and great engineering application value for the maintenance and reinforcement of the actual bridges.

Description

technical field [0001] The invention relates to bridge structure health monitoring technology, in particular to a neural network-based identification method for damage location and damage degree of a truss bridge. Background technique [0002] In recent years, with the rapid development of the national economy, the output and quality of steel have been continuously improved, and steel structures have become more and more widely used in large-scale engineering structures in my country. Steel truss bridges have become one of the most widely used bridge types in my country due to their advantages of large span, short construction period and adaptability to heavy-load traffic. However, with the aging and corrosion of steel structures, the reduction of fatigue strength and the increase of heavy traffic, the safety status of steel truss bridges in service is not optimistic. Therefore, it is particularly important to monitor the health of steel truss bridges and quickly and effici...

AI Technical Summary

Problems solved by technology

The method of testing stress has high accuracy, but it is sensitive to local damage, and the test workload is too large; the method of mode shape and flexibility is theoretically feasible, and the mode shape, especially the high-order mode shape, is sensitive to local damage, but for truss A large structure such as a bridge structure is affected by various factors, and the error of the measured mode shape is relatively large, which leads to a decrease in the accuracy of damage identification using the mode shape and its derivatives

Images