Design and implementation of a power quality disturbance classifier: An AI approach

This paper presents a new intelligent system incorporating wavelet transform, artificial neural network and fuzzy logic to automate the classification of power quality disturbance. This novel and efficient method in hardware, based on FPGA technology showed improved performance over existing approaches for power quality disturbance detection and classification on six types of disturbances including sag, swell, transient, fluctuation, interruption and normal waveform. The approach obtained an average classification accuracy of 98.19%. The design was successfully implemented, tested and validated on Altera APEX EP20K200EBC652-1X FPGA utilizing 1209 logic cells and achieved a maximum frequency of 263.71 MHz.