Grey clustering analysis for incipient fault diagnosis in oil-immersed transformers

This paper proposes a method for incipient fault diagnosis in oil-immersed transformers using grey clustering analysis (GCA). Incipient faults can produce hydrocarbon molecules and carbon oxides due to the thermal decomposition of oil, cellulose, and other solid insulation. The power transformers can be detected and monitor abnormal conditions with dissolved gas analysis (DGA). Various artificial intelligent (AI) techniques have been proposed for transformer fault diagnosis; however they have some limitations such as accuracy of diagnosis, requirement of inference rules, and determination of the detection architecture. IEC/Cigre standard and GCA are applied to diagnose internal faults including thermal faults, electrical faults, and faults with cellulosic insulation degrading. Compared with other diagnostic techniques, numerical tests with practical gas records were conducted to show the effectiveness of the proposed model, and are easy to implement with the portable device and hardware device.