Research On Power Quality Signal Analysis Based On Wavelet Transform And Neural Network

With the intelligent factory becoming more and more common,precision technology production and high-precision production operations bring a large number of nonlinear loads and time-varying power loads to the power grid,resulting in complex power quality disturbance problems.Serious power quality disturbance will bring huge security risks and irreparable economic losses.Therefore,it is particularly important to detect and classify power quality disturbances,which can accurately find out the problems of power quality disturbances and quickly solve them.Based on the research of wavelet transform and neural network,the causes and solutions of power quality disturbances are analyzed.The main contents of this paper are as follows.Firstly,this thesis analyzes and summarizes two kinds of common problems of power quality,and expounds the detection of power quality and the classification of power quality disturbance at home and abroad.Five kinds of common transient power quality disturbances are introduced in detail,and the mathematical model is established.This paper introduces the modulus maximum detection method based on wavelet transform,and selects the most suitable wavelet function to detect the power quality disturbance signal,and analyzes the five transient disturbances.After that,the noisy disturbance signal is denoised.Secondly,by combining the classification advantages of BP neural network.According to the learning algorithm of BP neural network,according to the different energy distribution of power quality disturbance,combined with the feature vector extracted by wavelet transform,the feature vector is input into the neural network to achieve the classification of power quality disturbance.Finally,combining wavelet transform with artificial neural network,a wavelet neural network for harmonic detection is proposed.By setting the wavelet function as the excitation transfer function of neural network,a new neural network structure is designed to meet the requirements of harmonic detection.The fundamental frequency,amplitude,phase and other parameters can be obtained at one time.Finally,through the simulation analysis,the experimental results show that the method is effective and accurate.