Transformer fault diagnose intelligent system based on DGA methods

Abstract

Power transformers have great importance in power system networks. Any malfunctions in the power transformers cause a system disconnection, which leads to lost profits for the electricity utilities. Transformer malfunctions can result from various stresses like electrical, thermal, or mechanical pressures acting on the insulation system, typically composed of insulating oil and paper. Dissolved Gas Analysis (DGA) is widely adopted to identify transformer faults. While traditional DGA methods such as the IEC Code, Rogers Ratio, and Duval triangle exist, their diagnostic accuracies are often lacking. So, optimization techniques are applied to augment the artificial intelligence of conventional DGA, aiming to significantly enhance the accuracy in diagnosing faults in power transformers. Still, it individually does not give high diagnostic accuracy. Therefore, a transformer fault diagnosis intelligent system (TFDIS) was developed in this work to increase the high analytical accuracy of recent DGA methods based on comparing the output of four DGA methods such as code tree 2020, modified IEC, and Rogers' ratio method, and Neural pattern recognition. The intelligent system developed a diagnostic accuracy (89.12%), higher than the highest diagnostic accuracy created by neural pattern recognition (86.01%).