Moving force identification by using discrete wavelet basis and L1-norm regularization
As one of most important issues in the field of structural health monitoring(SHM),moving force identification(MFI)on bridges has been the hotspot in bridge engineering.MFI is a typical inverse problem,but sometimes the existing traditional methods,such as time domain method(TDM),method of moments(MOM)etc,cannot identify moving vehicle loads precisely under environmental noises.For the shortcomings of TDM,a new MFI method is proposed based on the truth that natural signals are usually sparse in some domains and moving forces can be sparsely represented by some bases.In which,the wavelet basis is firstly used to sparsely represent time-varying components of moving force.Sparsity of moving force coefficients in wavelet domain is then considered and l1 regularization is adopted to obtain a sparse solution to the MFI problem and to improve ill-posedness of the MFI problem.In order to assess the feasibility of the proposed method in this study,a simply supported beam is taken as an example of numerical simulations,and MFI results are obtained from bending moment and acceleration responses polluted by different noise levels.The illustrated MFI results of numerical simulations show that the proposed method is a good MFI method with reasonable MFI results and good robustness.
Moving force identification (MFI) time domain method (TDM) l1 regularization discrete wavelet basis
Huanlin Liu Ling Yu
School of Mechanics and Construction Engineering,Jinan University,Guangzhou 510632,China School of Mechanics and Construction Engineering,Jinan University,Guangzhou 510632,China;MOE Key Lab
国际会议
The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)
青岛
英文
2774-2782
2018-07-22（万方平台首次上网日期，不代表论文的发表时间）