Moving Force Identification Of Bridge Under Constraint Of Stable Mean Value Based On L₁ Regularization Method

Bridge is a kind of important infrastructure in the transportation network.It is of great social and economic significance to ensure the safe operation of the bridge.In actual operation,bridges will face many potential threats,such as environmental erosion,material aging and heavy load.Among them,vehicle moving load is the main live load of bridge,which plays an important role in the safe operation of bridge structure.Moving load identification is an important means to pick up the load on the bridge,which is of great significance to the health assessment of the bridge.At present,moving load identification has attracted much attention,and many methods such as sparse regularization identification method have been applied.However,the influence of the existing research on the calibration error of the sensor is not fully discussed.Based on this,this paper considers the influence of sensor calibration error,and combined with L₁regularization method,carries on the related research on bridge moving load identification:（1）Firstly,this paper introduces the research background and significance of moving load identification,and systematically introduces the research status of moving load identification at home and abroad.At the same time,the problems of L₁regularization method are introduced in detail,and the main research content of this paper is determined.（2）The principle of L₁regularization method and the theoretical basis of mobile load identification are introduced in detail.The unknown moving load is expanded by using load dictionary and initial conditions of structure are described by modal space.Based on L₁ regularization method,the identification equation of the moving load of simply supported beam is established and the numerical simulation is carried out.The results show that L₁ regularization method can be used to identify the moving load and the recognition effect is better than Tikhonov method.However,for the simulation conditions with calibration error at the measuring point,the moving load curve identified by L₁ regularization method has obvious wave lines,and the recognition effect is poor.（3）Aiming at the problem of poor load identification effect caused by sensor calibration error,this paper proposes a sparse regularization method for bridge moving load identification based on stable mean constraint.Firstly,the inaccuracy of the relationship between the moving load and the structural response caused by the calibration error is analyzed qualitatively;then,the moving load trend line is described quantitatively,and the identification equation is optimized by adding the mean stability constraint,which overcomes the problem of large fluctuation of the load identification curve.Furthermore,the improved algorithm is used to simulate the simply supported beam model.The results show that the improved algorithm has a strong ability of load identification under the same conditions,which verifies the superior performance of the method.（4）In order to verify the feasibility and effectiveness of the proposed method,a simply supported beam model of Alec plate is set up in the laboratory.The first four order frequency,vehicle moving speed and bending moment response diagrams of the simply supported beam under various working conditions are collected by using instruments and equipment,and the collected data are imported into the improved algorithm for moving load identification.The experimental results show that the L₁regularization identification method based on the stable mean constraint can accurately identify the moving load on the test beam,which has good practical significance.