Transformer fault diagnosis using fuzzy logic and neural network

For a good quality of power system there is a great need for well designed protection for power system devices. Among all the devices, power transformer plays a pivotal role and requires high capital cost. During the last few years, there has been a trend of continuous increase in transformer failures. It is therefore vital to correctly diagnose their incipient faults for safety and reliability of electrical network. Various faults could occur in a transformer such as disc to disc, winding to ground and winding to winding. This paper presents fuzzy logic tool and neural network technique that are used to diagnose multiple faults in a transformer and monitor the trend.