Study On Damage Identification Of Simply Supported Beam Bridge Based On Kalman Filter And Statistical Analysis

The research of bridge damage identification is a very important part in the field of structural health monitoring.In the actual engineering inspection,the test signal will inevitably be affected by the measurement noise,which leads to the deterioration of the accuracy of damage identification.In order to reduce the impact of measurement noise on the identification results in bridge damage detection and improve the identification accuracy,this paper combined Kalman filtering and statistical analysis method to identify the damage location and damage degree of simply supported beam bridge structure while reducing the impact of noise on the response signal.The specific works are as follows:(1)A damage identification method for simply supported beam bridges based on Kalman filtering and principal component analysis under moving loads is proposed.After denoising the acceleration response of the structure with Kalman filter,the statistical characteristics of the principal components of the denoising signal were calculated by D-S evidence theory to get the final damage identification index.At the same time,a method to estimate the signal noise variance was proposed to improve the noise reduction accuracy of Kalman filter.The results of numerical example and model experiment showed that the Kalman filter algorithm had good noise reduction effect,and the proposed method could successfully identify the damage location of simply supported beam bridges on the basis of effectively reducing the influence of noise.(2)In order to denoise the signal under the condition of unknown noise intensity,a simple-supported beam bridge damage identification method based on factor analysis under moving load was proposed.The special factors of structural response were extracted to achieve the purpose of signal denoising,and the D-S evidence theory was used to calculate the statistical characteristics of special factors to construct the damage index.The influence of noise,sparse measuring points,vehicle speed and vehicle weight on damage identification was studied by a numerical example.It was found that only when the vehicle speed was too high or too low and the measuring points were sparse,the identification accuracy would be greatly affected.The influence law of vehicle speed and vehicle weight on damage identification effect was illustrated by plotting the interference rate curves of vehicle speed and vehicle weight.Finally,the effectiveness of this method was verified by experiment.(3)Based on the research of structural damage location identification,a damage degree identification method of simply supported beam bridge based on extended Kalman filter was proposed.The displacement response of the damage location was used as the system observation,and the damage factor of the element was used as the system state quantity.The damage factor of the structure was identified by the extended Kalman filter,and then the damage degree of the structure could be obtained.Numerical results showed that the method had good noise robustness,and the effectiveness of the method was verified by experiment.