Power Quality Disturbance Classification Based on Using Root Sum Square

Abstract

Voltage and current data obtained by power quality disturbances are used to protect the loads connected to the power system. In analyzing failure patterns based on the data, the accurate detection and classification of the power quality disturbance is important to solve the fundamental problem of disturbance occurrence. There are 8 types of general disturbance waveforms, which are modeled mathematically in the paper. 4 types preprocessing methods are applied to these waveforms to compare their performance. At this time, Root Sum Square (RSS) technology is proposed to increase the efficiency of preprocessing. The extracted feature data is used to determine the degree of classification using the dispersion and cohesion of distribution. The preprocessing technique that are suitable for the detection and classification of power quality disturbance are proposed.