To find out the incipient power transformer fault symptom diagnosis,  a successful adaptation of the neural based algorithm using artificial neural network (ANN) is presented in this paper. A neuron based encoding technique is applied to improve the accuracy of classification, which removed redundant input features that may be confusing the classifier. Experiments using actual data demonstrated the effectiveness and high efficiency of the proposed approach, which makes operation faster and also increases the accuracy of the classification. Five gases namely hydrogen, ethane, methane, acetylene and ethylene are chosen as inputs. Nine output codes for the different type of faults such as partial discharge of low energy, partial discharge of high energy, low energy discharge, high energy discharge, thermal fault  thermal fault of  thermal fault of  thermal fault  and no fault condition are considered. The neural network is implemented using the back propagation algorithm and the hidden layer neurons are chosen from hit and trail method.