Identification and Diagnosis of Bridge Structural Damage Based on Static Test Data

Large bridge structures are pivotal projects in the transportation system and play a crucial role in social life. With the frequent occurrence of bridge accidents, people are paying more and more attention to the safety of bridge structures. However, existing bridge structure damage identification methods have problems such as low recognition accuracy, high damage localization error rate, and poor recognition effect. In response to the appeal issue, this article used data processing methods based on static test data to denoise and clean the experimental data re-collected from static test data, and obtained effective bridge structural damage data. With the help of these data and backpropagation (BP) neural network, a bridge structural damage identification pattern was constructed. Using this pattern to identify bridge structural damage can effectively address the issues of low identification accuracy and high damage localization error rate. Through experiments, it can be found that the recognition pattern based on BP neural network had an accuracy of over 92.16%, 93.44%, and 94.13% for extracting displacement, strain, and deflection of bridges, respectively. The average recognition accuracy was 95.038%, 94.696%, and 95.27%, respectively. Using static test data and BP neural network to construct a bridge structural damage identification pattern can effectively improve the accuracy of bridge structural damage identification and reduce the error rate of identification and positioning.