Research On The Load Identification Method Of Small And Medium-Sized Bridges Based On Dynamic Influence Line

In recent years, more attention has been paid to the security of bridges due to their frequent collapses. The bridge structure health monitoring (SHM) becomes a hot topic in the study of bridge engineering today, and now the bridge SHM is based on distributed sensing technique major in the damage identification of the bridge structure. And the moving load identification technique can improve the bridge SHM system. But most of the current load identification methods have different limitations and need to be proved by more practical engineering.In this master thesis, a novel moving load identification method is proposed, which is based on the influence line and load transverse distribution. For the multiple moving loads, the BP nervous network are also employed to identify the moving loads. A series of investigations have been carried out to confirm the feasibility of the proposed method, including the numerical simulation and car-bridge experiments.The main research contents and conclusions are as follows:(1) The influence line based moving load identification method is proposed and studied with the load transverse distribution under consideration. A series of numerical simulation cased were performed to compare the load identification with and without the load transverse distribution under consideration. The results showed that the method without considering the load transverse distribution is not suitable for solving the spatial problem, and the new proposed method with the transverse distribution under consideration can accurately identify the moving vehicle in any position on the deck. The proposed method is characterized by high accuracy and anti-noise performance.(2) Through the car-bridge experiments, the new proposed method with the transverse distribution under consideration was studied in detailed. The results of the speed identification showed that the relative error can be controlled within ±5% and the accuracy of the speed had great influence of the load identification. The result of the load identification showed that the relative error can be controlled within±10%, and in 93% of the samples the relative error is merely about ±5%.(3) It was unable to establish ideal mathematical model to do research on the two cars’ load identification, so this paper proposed a two cars’ load identification method based on the influence line and the BP neural network method. A series of car-bridge experiments were performed to confirm the feasibility of the proposed method. The results showed that this method can accurately identify the vehicles’ lane position and the relative error of the load identification can be controlled within ±10%, and in 97% of the samples the relative error is merely about ±5%.