Tomographic Reconstruction of Defects in Composite Plates Using Genetic Algorithms with Cluster Analysis

Abstract

Tomographic reconstruction of defects in composite plates is a complex procedure, as the elastic properties of the composites are direction-dependent also leading to the directional dependence of the wave velocities. Furthermore, at the interface of inhomogeneities a wave undergoes refraction, mode conversion, and beam skewing. A tomographic method for composite materials will have to deal with multiple elastic properties and ray bending. In the present approach, the genetic algorithm (GA), being an efficient optimization method, is used as a tool for reconstruction. The objective function of GA consists of three terms, viz. time-of-flight difference between the projection data and the data generated by the GA, the number of clusters in that data set, and the number of violation of bounds on variables. The laminate is considered thin, and Lamb waves are used as the source of radiation. A ray-tracing algorithm is used to determine the ray-paths, and an edge-detection algorithm is used to find out the boundary of the insert from the data created by the GA as it proceeds. This approach is successfully applied to simulated data sets, and the results are found to be quite encouraging.

Keywords:

• cluster analysis
• composite plate defects
• genetic algorithm
• nondestructive testing
• tomography

ACKNOWLEDGMENT

This work was supported by a grant from the German Ministry of Education and Research within the framework of Indo-German collaboration, International Office, DLR Cologne, Germany, and the Department of Science and Technology, Ministry of Education, Government of India, New Delhi.