Combined Novel Approach of DWT and Feedforward MLP-RBF Network for the Classification of Power Signal Waveform Distortion

Power Quality (PQ) has become a major concern owing to its increased use of sensitive electronic equipment. In order to improve PQ problems, the detection and classification of PQ Disturbances (PQDs) must be carried out first. This paper presents a simple software based technique for detection and classification of PQDs by time-frequency analysis of Wavelet Transform (WT) as features extraction and Artificial Neural Network (ANN) as classifier. This approach detects and classifies the types of Waveform Distortion (WFD) problems of PQDs selecting suitable feature extraction with statistical parameters, as an input of feedforward Radial Basis Function (RBF) and Multilayer Perceptron (MLP). This methodology shows applicability, simplicity, and accuracy proving as promising tool for the automatic detection and classification of WFD of EPQ problems.