Modified PSO and wavelet transform-based fault classification on transmission systems

This paper presents the development of an algorithm based on discrete wavelet transform (DWT) and particle swarm optimisation (PSO) for classifying the power system faults. The proposed technique consists of a preprocessing unit based on DWT in combination with PSO. The DWT acts as extractor of distinctive features in the input current signal, which are collected at source end. The information is then fed into PSO for classifying the faults. It can also be used for offline processing of data stored in digital recorders. Extensive simulation studies carried out using MATLAB show that the proposed algorithm not only provides an acceptable degree of accuracy in fault classification of 400 kV transmission system under various fault conditions when compared to the results obtained using probabilistic neural network (PNN) method, but is reliable, fast and computationally efficient too.