| As the second category of inverse problem, the dynamic load identification technology gets more and more attention in recent years, and some effective identification methods have been studied out.In this paper, we focus on the technology of the distributed dynamic load identification for the one-dimentional and two-dimentional dirtribution based on the theory of modal coordinate transformation.This article obtains from the inverse problem to discuss the ill-posed problem and the regularization methods are introduced to improve the ill-posed problem. Taking the typical one-dimentional and two-dimentional stress structure(continuous beam structure and elastic thin plate stucture) as the research object, the concentrated and distributed dynamic load identification models are deduced in frequency domain. For the concentrated dynamic load, the method of modal coordinate transformationon is applied to rebuild it. For the one-dimentional and two-dimentional dirtribution of the dynamic load, we use the theory of modal coordinate transformation to descomposite it with the mode functions in the modal coordinate space. And the liner relationship bewteen the time function coefficients of the distributed load and the modal excitations is established. Then in frequency domain,the modal excitations can be identified from the known response dates of the measuring points based on the modal theory to rebulid the distributed dynamic load. To solve the noise problem, in the process of the modal coordinate reponses identificantion, the Tikhonov regularization algorithm is proposed to improve the recognition accuracy and the signal-to-noise ratio in the dynamic load identification problem.In this paper, a large number of computer simulation examples and the test of dynamic load identification on the elastic thin plate structure verify the reliability of this method and the precision which has some practical value in engineering. |