A power quality disturbance identification method based on discriminative dictionary learning under src framework

Abstract

The invention discloses a power quality disturbance recognition method based on discriminative dictionary learning under the SRC framework. First, feature dimensionality reduction is performed on a large number of different types of power quality data to ensure the irrelevance of the data and the completeness of the type. Secondly, under the sparse representation Construct the optimal basis, update and optimize the sparse representation matrix and establish redundant sub-dictionaries of different types of power quality disturbances, and cascade them into a discriminant dictionary, then obtain the sparse representation matrix of the power quality disturbance signal to be identified, and finally use The redundant sub-dictionary of different types of power quality disturbances reconstructs the signal in turn, calculates the redundant error with the original signal respectively, and determines the target class by the minimum value of the redundant error. The method of the present invention implements multi-classification disturbance recognition by training a universal and optimal discrimination dictionary and adopts a compressed sensing reconstruction algorithm under the SRC framework, and realizes a power quality disturbance recognition model of multi-classifiers without combining two classifiers.

Description

technical field [0001] The invention belongs to the research field of power quality analysis technology of power systems, in particular to a power quality disturbance identification method of discriminant dictionary learning under the framework of SRC (sparse representation based classification). Background technique [0002] With the development of non-linear, network integration, large-scale and other directions in the industrial field, the continuous increase of nonlinear power electronic devices such as rectifier equipment and variable frequency speed regulation equipment in the system and the multi-grid power supply, the problem of power quality pollution is becoming more and more serious. The issue has received extensive attention from all walks of life. In-depth study of various factors affecting power quality and accurate identification of power quality disturbances are the premise and basis for analysis and evaluation of power quality problems, rationally formulatin...

AI Technical Summary

Problems solved by technology

The traditional sampling method and identification process have brought a lot of waste of resources, so looking for a new identification method, using the discriminant dictionary directly from the sparse representation matrix of the compressed sampling signal, comparing the redundant errors of different redundant sub-dictionaries, and completing the analysis of power quality Disturbance identification has important theoretical and practical significance

Images