Transformer Fault Diagnosis Method Based on Hunter Hunt Algorithm Optimization

The method of dissolved gas analysis (DGA) in oil is one of the important methods for transformer fault detection. However, due to the shortcomings of traditional diagnostic methods, such as incomplete coding and too absolute coding, they can not meet the actual fault diagnosis needs. In recent years, intelligent algorithms have been widely used. Aiming at the shortcomings of Extreme Learning Machine (ELM), such as poor training stability and low training accuracy, this paper proposes a transformer fault diagnosis method through Hunter Prey Optimizer (HPO) to optimize ELM. At the same time, the noncoding ratio method is used to increase the information dimension, which can obtain more comprehensive fault information. Experimental results show that the HPO-ELM algorithm has fast training speed, high accuracy, strong stability, and better fault diagnosis performance.