Research On Disturbance Detection And Classification Method Of Power Quality

In recent years,with the rapid development of science and technology,a large number of advanced and sophisticated electrical equipment have been applied to people’s production and life.While bringing convenience to people’s production and life,the power grid has been seriously polluted and greatly reduced.Power quality,however,the use of these sophisticated electrical equipment and a large number of new energy equipment puts higher demands on the quality of electric energy,so it is imperative to improve the quality of power and improve its quality.To improve the quality of electric energy,it is first necessary to discriminate the occurrence and duration of power quality disturbances.However,the study of disturbance identification of power quality is of great significance for identifying the type of disturbance.In this paper,the transient power quality disturbance signal is taken as the research object.By means of mathematical morphology,S transform,classification rule tree and other algorithms,the occurrence and stop times of the power quality disturbance signal and the disturbance frequency are detected,and the classification of different disturbance signals is realized.First,the power quality disturbance signal is preprocessed based on the mathematical morphology method.According to the mathematical model of the power quality disturbance signal and the Matlab2012 simulation platform to realize the disturbance simulation waveform of the transient power quality.In order to make the simulation model more in line with the disturbance signal collected in the actual power grid,noise is added to the transient energy disturbance signal,and the multi-structural element composite morphology filter is used to denoise the disturbance signal of the noisy transient power quality.The experimental results show that the noise reduction effect of the multi-structure element composite filter is significantly better than that of the single structure element composite filter,and provides a more effective basis for the subsequent study of power quality disturbance detection and classification.After that,a transient power quality disturbance detection and localization method based on S transform is proposed.The S-transformation is used to process the transient power quality disturbance signal,and the S-transform complex matrix is obtained.The square matrix of the S-transformed matrix is calculated.The square and sum of themagnitudes of the S-transforms are extracted to realize the stop timing and frequency of the disturbance signal.The detection is performed,and compared with the simulation of the wavelet transform,the S transform is more effective in detecting the start and stop times and frequencies of the disturbance signal,especially the composite disturbance.Finally,a decision tree algorithm using CART is proposed to classify transient power quality disturbance signals.By extracting the S-transformed squared mean and its peak value as the characteristic parameters of the transient power quality disturbance signal,the thresholds of each node in the decision tree are set,which simplifies the classification rules and improves the classification efficiency.The experiment proves that the feature parameter selection based on S transform has correctness,and the classification method has the characteristics of high accuracy and good noise resistance.Figure [43] table [12] reference [63]...