Visual Localization and Quantitative Detection Method for Thermal Defects in Cable Terminals of High-Speed Trains Based on a Temperature Derivative

Abstract

Cable terminal defect detection plays an important role in ensuring the safe and stable operation of high-speed trains (HST). In this paper, a numerical model of the electromagnetic thermal field of overheating defects of cable terminal shielding grids and a method of detecting internal defects of cable terminals - even-order temperature derivative (EOTD) is proposed for quantitative detection of internal defective structures of vehicle-mounted cable terminals. Firstly, a numerical model of the electromagnetic thermal field of cable terminals under the condition of leakage current is constructed, through which the temperature field distribution characteristics of different defective structures are analyzed. Then, the intuitive location of the defective region is obtained by investigating the derivative characteristics of the temperature image, and the depth and intensity of the defects are quantitatively assessed by using the main side peak (MSP) distances and the MSP values extracted from the derivative curves. Finally, the simulation and experimental results achieve the identification of defect structures and the quantitative detection of defect depth and intensity, proving the effectiveness and accuracy of the proposed method.