Study on the Detection Method of Winding Overheating Fault in Oil-Immersed Transformer Based on Infrared Thermal Imaging
During the operation of oil immersed transformers,the winding overheats due to various factors,which leads to a decrease in Kappa value and subsequently affects fault detection.Therefore,a method for detecting over-heating faults in oil immersed transformer windings based on infrared imaging is proposed.Using infrared thermal imaging technology to capture external thermal images of oil immersed transformers;To address the issue of uneven image quality,Gaussian filtering and histogram equalization techniques are used to preprocess infrared thermal imag-ing images,and image stitching techniques are employed to generate complete transformer information images;By combining image segmentation,enhanced residual network architecture,and relative temperature difference method,accurate identification and detection of transformer winding overheating fault areas can be achieved.The experimental results show that the Kappa value of the research method is significantly higher than that of the com-parative method,which can improve the detection performance of overheating faults in oil immersed transformer windings and has practical application value.
万方数据
维普
