Application of Adaptive Neuro-Fuzzy Inference System Based on IEC Method for Transformer Fault Diagnosis

摘要

Power transformer is one of the most important components in a power system. It experiences thermal and electrical stresses during its operation. The insulation system consisting of mineral oil and the insulation paper used in transformer undergoes chemical changes under these stresses and gases are generated. These gases dissolve in oil. The dissolved gases are extracted in the laboratory using gas chromatograph. The dissolved gases are used for fault identification. The fault identifications in a transformer are based on certain key-gas ratios. International standards such as IEEE and ASTM are used for fault identification. However, these standards are not able to diagnose the faults under certain conditions. Hence, there is a need to improve the diagnostic accuracy. This study attempts to diagnose the faults in a power transformer using adaptive Neuro-Fuz2y Inference System. Simulation model is developed using MATLAB™ Software and trained using the IEC TC 10 database of faulty equipments inspected in service. The outputs of the adaptive Neuro-Fuzzy Inference System based model are compared with the Roger's Ratio Method. The comparison shows that the condition assessments offered by the adaptive Neuro-Fuzzy Inference System based model is capable of predicting the transformer faults with higher accuracy.