PRIMARY FAULT DETECTION OF TRANSFORMER USING NEURAL NETWORK

The most widely recognized determination technique for power transformer faults is the dissolved gas analysis (DGA) of transformer oil. Different strategies have been produced to define DGA results such as key gas method and roger’s ratio method. The present methodology uses IEC 60599 ratio method to distinguish fault in transformers, which is having the benefit of using three gas proportions rather than four gas proportions. In some cases, the DGA results cannot be coordinated by the current codes, making the diagnosis unsuccessful in multiple faults. To overcome this issue, we have proposed the utilization of neural networks to demonstrate their capability to recognize the primary faults in transformers.