Diagnostic of Defects with Inductive Thermography Imaging System by Using PSO Algorithm

Background/Objectives: This paper proposes an unsupervised method for diagnosing and monitoring defects in inductive coils. The objective of the present paper is to locate the exact crack location and diagnose easily. Methods/statistical analysis: The core of the method is based on the signal processing-based pattern extraction algorithm using Sparse Greedy based Principal Analysis(SGPCA).The proposed method is fully automated and diagnosis of defects is done by Particle Swarm Optimization algorithm(PSO).By implementing PSO algorithm, the cracks on the complex surface will be detected easily. Findings: It is shown that the SGPCA analysis for detecting the defects is not suitable for all type of metals. An internal functionality is built in the proposed algorithm to control the sparsity of SGPCA. Experimental tests with other methods have been conducted to verify efficiency of proposed method. Application/ Improvements: Improvement in efficiency of inductive coils.