On the Application of Artificial Neural Network for Classification of Incipient Faults in Dissolved Gas Analysis of Power Transformers

Oil-submerged transformer is one of the inherent instruments in South African power system. Transformer malfunction or impairment may interpose operation electric distribution and transmission system, coupled with liability for high overhaul costs. Hence, recognition inchoate faults an oil-submerged indispensable it has turned into intriguing subject interest by utility owners manufacturers. This work proposes a hybrid implementation multi-layer artificial neural network (MLANN) IEC 60599:2022 gas ratio method identifying mineral oil-based submerged transformers employing dissolved analysis (DGA) method. DGA staunch practice to discover as furnishes comprehensive information examining state. In current work, MLANN was established pigeonhole seven fault types states predicated on three International Electrotechnical Commission (IEC) combustible ratios. The designs enmesh development numerous algorithms picking networks optimum performance. ratios are accordance standard whilst empirical databank comprised 100 datasets used training testing activities. designated design produces overall correlation coefficient 0.998 categorization state reference produced.