An improved two-dimensional multiple signal classification approach for impact localization on a composite structure

摘要

Efficient impact monitoring techniques are important to be researched to ensure the safety of in-service composite structures. Multiple signal classification-based method is a new promising method. This method uses small aperture sensor array arrangement instead of sparse sensor arrangement. It has the advantage of easy sensor array arrangement. By processing the whole sensor array outputs using the multiple signal classification algorithm, transfer function and material property of the structure are not needed. Although some researches have shown the potential of this new method, most of the current researches have been performed based on the far-field assumption on aluminum structures. In engineering applications, composite structures are the main structures which need impact monitoring. Deep research needs to be addressed. Aiming at these problems, an improved two-dimensional near-field multiple signal classification processing algorithm is presented for impact localization on composite structure in this article. The proposed method can simultaneously estimate the impact occurring direction and distance. A uniform carbon fiber composite plate and a real aircraft oil tank made of carbon composite structure with varied thickness are adopted to evaluate the presented method. To deal with the anisotropy performance of the complex composite, the two-dimensional multiple signal classification processing algorithm is also improved by taking into account the existing anisotropy performance.