Concurrent Identification of Impact Location and Force Magnitude on a Composite Panel

Hamed Kalhori,Mehrisadat Makki Alamdari,Bing Li,Ben Halkon,Seyedeh Marzieh Hosseini,Lin Ye,Zheng Li

DOI: https://doi.org/10.1142/s0219455420420043

IF: 2.9568

2020-01-01

International Journal of Structural Stability and Dynamics

Abstract:Simultaneous estimation of both the location and force history of an impact applied on a lattice truss core sandwich panel is inversely carried out utilizing velocity signals collected by means of a scanning laser Doppler vibrometer. The algorithm assumes that several impact forces are exerted concurrently on a number of specified locations on a panel, provided that the magnitude of all impact forces but one is actually equal to zero. This condition equates to a scenario where an impact occurs at only one location. The purpose is therefore to detect the actual impact location among all potential locations, together with its force history, through minimizing error functions. Two algorithms, the one-to-one (even-determined) approach and the superposition approach, are considered. The one-to-one approach solves the reconstruction problem independently for each pair of impact and measurement points. However, in the superposition approach, the impact forces at all potential locations are concurrently reconstructed through a single matrix equation. It is shown that the one-to-one approach fails to detect the true impact location while the superposition approach recognizes the actual impact location based on some qualitative evaluating criteria. Adopting the superposition approach, for a problem with four possible impact locations, two scenarios one with four and one with 12 measurement points, are investigated. It is observed that the additional measurement points do not necessarily enhance the efficiency and accuracy of the proposed method. It is found that different arrangements of measuring points lead to identification of the location and the magnitude of the impact force, though the use of four evenly distributed measurement points seems to be most effective in simultaneous identification of the location and magnitude of the impact force. Further, a quantitative index based on the concept of similarity search for time-series using wavelet transformation is proposed and it is demonstrated that the index can successfully identify the true impact force location in a fully automated way.