Novel Diagnostic Method for GIS Mechanical Defects Based on Energy Entropy and Improved Probabilistic Neural Network

The mechanical defects of gas insulated switchgear (GIS) equipment are one of the important causes of equipment failures, and conducting research on mechanical defect diagnosis methods is crucial for ensuring the safe and stable operation of equipment. This paper proposes a novel diagnostic method for GIS mechanical defects based on energy entropy and improved probabilistic neural network. Firstly, based on the true GIS mechanical vibration simulation platform, the defect vibration signals were decomposed into multiple components using the variational mode decomposition algorithm; Then, based on the cross-correlation coefficient, the main components were selected to extract energy entropy feature values; Finally, the sine cosine algorithm was used to optimize the probabilistic neural network and constructed a GIS mechanical defect diagnosis model, and its effectiveness was verified. The results show that the diagnostic accuracy of this method can reach 98% or above, and the diagnostic accuracy is improved by more than 11% compared to the unoptimized model, which can achieve accurate classification of mechanical defects.