The Power Quality Disturbance S Transform Classification And Identification

Along with the expanding power system scale and a great deal of nonlinear, impact and unsymmetrical load increased significantly in the power system. The power system is more and more seriously polluted, these pollutions greatly worsened system of electric energy quality, brought more and more economic loss. So to improve and improve power quality become the research hot spot, and how to detect the power quality disturbance and the power quality disturbance to effectively identify seems particularly important.This paper first introduces the basic definition of the power quality disturbance and classification, detailed analysis of the power quality disturbance recognition feature extraction and classification method, the power quality disturbance feature extraction are mainly include time domain, frequency domain and transform domain method, this paper mainly introduces S transform, this paper introduces S transform algorithm principle, the power quality disturbance signal become a complex matrix through S transformation, feature extracting from module time-frequency matrix, the row vector of module matrix represents a frequency amplitude changing with time distribution, column vector represents a sampling time of amplitude with frequency variation distribution. Based on S transformation’s advantages this paper choose S transformation to extract the electric power quality disturbance signal feature.In this paper, using Matlab construct a perfect power quality disturbance signal, there are five kinds of single disturbances and four kinds of compound disturbances, single disturbances include voltage swell, voltage sag and voltage interrupt, voltage flicker, and harmonic Compound disturbance have harmonic and voltage swell, harmonic and voltage sag, have harmonic and voltage interrupt, have harmonic and voltage flicker. Use of S transform and the module time-frequency matrix, the extracted feature vector, and then classify power quality disturbance is adopted in this paper through the probabilistic neural network to classify power quality disturb ance. According to the simulation results shows that this method classification accuracy is high, the anti-noise ability is good, classification time is short, This Method can be effectively used in power quality disturbance identification.