Implementation of Power Quality Disturbance Classifier in FPGA Employing Wavelet Transform, ANN and Fuzzy Logic

Intelligent power quality monitoring systems have become an essential component of high technology and high availability-oriented industries. Most power quality disturbances are non-stationary and transitory and the detection and classification have proved to be very demanding. The first step towards any solution for a disturbance is to recognize the presence of a particular type of disturbance. This research automates this process where the design and implementation of a power quality disturbance classifier in FPGA is presented. The disturbances of interest include sag, swell, transient, fluctuation, interruption and normal waveform. The approach combines new intelligent system technologies using wavelet transform, artificial neural networks and fuzzy logic and provides some unique advantages regarding fault analysis. The approach developed in VHDL obtained a classification accuracy of 98.17%. The implementation of the project on Altera APEX EP20K200EBC652-1X FPGA utilized 1209 logic cells and achieved a maximum frequency of 263.71 MHz. Experimental results are included to demonstrated the performance of the classifier. The results obtained from the hardware exactly match the results obtained from software simulations when tested with software-generated signals and utility sampled disturbance events. This result validates the utility of the proposed approach.