Industrial user power equipment classification method based on harmonic emission level

Abstract

The invention relates to an industrial user power equipment classification method based on a harmonic emission level. The method comprises the following steps: collecting the load feature data of industrial user power equipment according to a preset sampling interval; preprocessing the acquired data; adopting a bilateral CUSUM algorithm to detect a switching event of the industrial user power equipment; according to the detected switching event, extracting a harmonic distortion rate and a harmonic value before and after the occurrence of the switching event from the preprocessed load characteristic data so as to construct a harmonic characteristic matrix of the industrial user equipment; screening and extracting high-contribution-rate features from the harmonic feature matrix of the industrial user equipment to construct a contribution feature matrix of the industrial user equipment; and performing clustering calculation on the contribution characteristic matrix by adopting a k-means algorithm, thereby determining different equipment categories in the industrial user power equipment. Compared with the prior art, the method enables the industrial user power equipment to be efficiently and accurately classified.

Description

technical field [0001] The invention relates to the technical field of electric equipment classification, in particular to a method for classifying electric equipment for industrial users based on harmonic emission levels. Background technique [0002] In the operation and regulation of the power grid, it is often necessary to clarify the characteristics of each load in the power grid and the operating status of the equipment in advance to ensure the accuracy of subsequent power regulation. At present, the loads in the power grid are mainly divided into residential loads and industrial loads. Among them, residential loads It includes power equipment for household users, while industrial load includes power equipment for industrial users. Compared with residential loads, industrial loads occupy a larger capacity and proportion in the power grid. Therefore, research on the classification of power equipment for industrial users will have important practice for power grid regulat...

AI Technical Summary

Problems solved by technology

[0003] Different from household power equipment, industrial power equipment contains more and more complex types. In order to accurately classify industrial power equipment, traditional methods mainly rely on deep learning and image recognition. This method relies on a large number of training Sample data, its efficiency is low, once new industrial user power equipment appears, it is necessary to re-train the classification model, otherwise the accuracy of the classification results cannot be guaranteed

Images