Fault diagnosis of a transformer based on polynomial neural networks

In view of the low accuracy of transformer fault diagnosis with traditional method, a novel multi-input and multi-output polynomial neural network (PNN) is proposed and used for transformer fault diagnosis. Firstly, single output PNN I classification model is trained and constructed according to the five kinds of characteristic gas corresponding four fault types (high-energy discharge, low-energy discharge, superheat and normal state) sample data, the transformer states are divided into normal state and abnormal state, then a transformer fault diagnosis model based on multiple output PNN II is built to aim at the three fault types such as high-energy discharge, low-energy discharge and thermal heating. Simulation and test results show that accuracy can reach 100% by using the presented model, which has excellent anti-interference performance.