Detection And Classification Of Power Quality Based On Adaptive Complementary LMD And RVM

Electricity is one of the most widely used energy sources in modern society.With the development of Economy and society,a large number of electric power and electronic equipment are continuously connected to the power grid,which leads to the deterioration of the power quality.Accurate and timely detection of the disturbance parameters and identification of their disturbance type is the key to solve the problem of power quality.On the basis of previous work,this paper studies the detection and classification of the problem of the power quality disturbance.The main content and results are as follows:Firstly,In this paper,the common methods of power quality disturbance detection and their advantages and disadvantages are introduced,and the method of using Hilbert-Huang(HHT)as the disturbance of power quality is considered in this paper.HHT algorithm has some advantages in dealing with nonlinear and unsteady signal processing and self-adaptation,but there are still some shortcomings,so the main work of this paper is to analyze and study the problems of HHT in application,the improvement method is put forward to apply it in the field of power quality inspection and classification.Secondly,the improved EMD algorithms EEMD,CEEMD and LMD are studied.These algorithms have some improvement to the traditional EMD algorithm,but EEMD and CEEMD have the shortcoming of total decomposition times and added noise,and LMD has certain modal mixing and false components.In this paper,a new algorithm based on adaptive complementary local mean decomposition(ACLMD)algorithm is proposed,and the improved method is applied to the analysis of power quality disturbance.The simulation results show that the proposed method can not only extract characteristic parameters of power quality disturbance effectively,but also accurately locate the time of the disturbance occurrence and stop,and it has good anti-noise performance.Finally,based on the classification of disturbance problems and the classification characteristics of the relevance vector machine(RVM),a new method of power quality disturbance classification based on adaptive complementary LMD and RVM is proposed.ACLMD method is used to deal with the disturbance signal of power quality,obtain the PF component of a series of different local feature time scales,take the first four components of most information,extract the characteristic value,and input PSO-RVM classifier after processing.Based on the Matlab platform,the accuracy of the algorithm is verified,and it is proved from the simulation results that the feature extraction method based on ACLMD can accurately obtain the feature information of each disturbance,PSO-RVM can correctly identify the various disturbance signals with different signal-to-noise ratio by the characteristic information of the disturbance,and it has some noise-resisting property.