Damage identification for bridge structures based on correlation of the bridge dynamic responses under vehicle load

Abstract

In order to realize the damage identification of the bridge, a method based on the correlation of vehicle-induced response of the bridge is proposed. Firstly, using the measured vibration responses of the bridge under vehicle load, damage indexes based on Pearson correlation coefficient, Spearman correlation coefficient and Kendall correlation coefficient are defined, respectively. Then, the proposed method is verified by numerical simulation and compared with the method based on Correlation function amplitude Vector (CorV). The effectiveness of the method for single damage and multiple damage of the bridge is examined. Numerical results demonstrate that the proposed method can effectively identify and locate the damage, and the effectiveness of the index based on Pearson correlation coefficient is the best. Finally, the proposed damage identification method is verified by experiments, and the results manifests that the comprehensive decision-making based on different damage indexes is helpful to improve the accuracy of damage location. Besides, the influence of road-surface roughness on damage identification is discussed. Results indicate that the damage indexes are affected by road-surface roughness, and the statistical mean value is beneficial to identify and locate the damage.