Probabilistic regularization load reconstruction method based on iterative strategy

Abstract

In view of the poor solution accuracy of the traditional Green's function-based load reconstruction method, this paper proposes a load reconstruction method based on an iterative solution strategy. Using Green's function matrix as the gradient information of the load and dynamic response, the load history is continuously updated to minimize the residual difference between the measured response and the reference model response to obtain a reconstruction result closer to the real load history. In addition, this paper derives a Green's function matrix based on the acceleration response time series, which extends the application scope of the traditional Green's function-based load reconstruction method. Furthermore, considering the influence of uncertainty factors such as measurement noise and model error on the reconstruction results, this paper proposes a probabilistic regularized load reconstruction method based on an iterative strategy by using probability theory to describe the uncertainty. The influence of uncertainty factors is considered both in the selection of regularization parameters and in the load reconstruction process. The effectiveness of the proposed method is verified by an example of a 35-rod truss, and the effects of model error and measurement noise on the reconstruction results are discussed. Compared with the traditional method, the proposed method can achieve more accurate and robust load reconstruction results, and the effect of uncertainty on the load reconstruction results can be quantified in the framework of probability theory.