Wavelet-based neural network for power quality recognition

Summary form only given. Power quality has become an important concern both to utilities and their customers with wide spread use of electronic and power electronic equipment. Power quality embraces problems caused by harmonics, over or undervoltages, or supply discontinuities. To improve the electric power quality, sources of disturbances must be known and controlled. This paper reports a new method, which does not have the limitations as mentioned previously. The new method is based on wavelets. Current waveforms of typical loads on the power system are sampled and converted into a sequence of digital values. A discrete wavelet transform is then applied to these values. In this way, the authors have been able to find out the different types of load that contributes electric power harmonics to the power system. Encouraging results have been obtained and are presented in the paper.