Localisation of unknown impact loads on a steel plate using a pattern recognition method combined with the similarity metric via structural stress responses in the time domain

Abstract

Abstract The localisation and reconstruction of impact loads are the most basic issues in structural health monitoring. Further, when identifying random unknown impact loads, obtaining the position and time history of the load using classical inversion techniques is both time-consuming and infeasible. This study proposes a novel scheme based on the pattern recognition method combined with the similarity metric in the time domain to determine the position of unknown impact loads. The basic idea of this approach is to apply two groups of impact loads to the structure, compare the structural responses of the two groups, and then determine the position of random impact loads. Considering a rectangular steel plate as the research object, impact forces are applied at 180 uniformly distributed reference points (reference group) and 23 randomly generated validation points (validation group). Based on the structural stress responses, the construction of the feature vectors is completed, and the region localisation and load localisation of the random impact load in validation group are then realised to determine the nearest reference point. Thereafter, the time history of the unknown impact load is reconstructed using the Tikhonov regularisation method and generalised cross-validation. In this study, the validity and anti-interference ability of the proposed method have been verified, and the analysis results indicate that the accuracies of the load localisation and impact load reconstruction at different noise levels satisfy the engineering requirements.