Ultrasonic Detection and Diagnostic Method for the Internal Defects in Epoxy Composite

Abstract

Insulation components are prone to form internal defects during the production process, which can have adverse effects on their stable operation. Thus, effective detection and diagnosis hold immense importance. In this article, an ultrasonic testing platform is designed to bubbles/cracks detect and verify the effectiveness and feasibility of this method. To address the issue of resolving weak acoustic signals to determine and distinguish defect types, a diagnosis method is proposed. The time and frequency domain information of defect echoes are obtained by discrete wavelet decomposition, and the defect features are reduced by linear discriminant analysis (LDA). Moreover, a step-by-step diagnosis model of typical defects is established by combining a sparrow optimization algorithm and a backpropagation neural network. This method enables the identification of various defect types and morphological characteristics, enhancing the sensitivity of the ultrasonic testing method, which can provide effective support for defect detection of insulators.