Artificial intelligent based technique for partial discharge detection / Rosainy Baharom

This project presents an artificial intelligent based technique for partial discharge (PD) detection. The study involves an artificial neural network (ANN) development which will be used to train and test a series of data set in order to predict the presence of PD. The PD data are generated using a PD circuit, controlled by the various parameters in the circuit. The generated data will be grouped into training and testing data. Two programmed are developed namely the training programme and testing programme. The developed ANN model is able to predict the occurrence of partial discharge under various scenarios.