Hybrid DoA–TDoA Method for Impact Localization on Thin-Walled Structures Using Sensor Clusters
Xu Zeng, Deshuang Deng, Hongjuan Yang, Zhengyan Yang, Lei Yang, Zhanjun Wu
Abstract
Impact monitoring technology plays a critical role in ensuring the structural integrity and safety of thin-walled engineering structures in service. This paper presents a novel hybrid DoA-TDoA method for impact localization on thin-walled engineering structures using sensor clusters. The methodology involves placing two sensor clusters on the structure to capture impact signals. Subsequently, narrowband Lamb wave signals at a specific frequency are extracted from impact signals using Continuous Wavelet Transform (CWT). The Normalized Variance Sequence (NVS) approach is then employed to determine the TDoA, and phase differences are calculated to estimate the Direction of Arrival (DoA). The DoA-based Spatial Beamforming Focusing (SBF) technique and TDoA-based hyperbolic locus imaging algorithm are used for impact imaging. An imaging fusion step is introduced to combine the results of the two imaging techniques, accurately determining the impact location. Experimental validation of the proposed method is conducted through impact tests on three distinct structures: a large-scale plate, a complex riveted stiffened plate, and a three-dimensional thin-walled cylindrical structure. A comparative analysis with two existing methods demonstrates the superior imaging resolution and localization accuracy of the proposed approach, which remains effective even in the presence of measurement noise. Additionally, the effects of sensor type, shape, and configuration on the localization results are discussed. This research contributes to the advancement of impact localization technology for thin-walled structures, with potential applications in structural health monitoring and safety assessment.