A non-intrusive load identification method and system based on wavelet transform

Abstract

This application relates to the field of load identification technology, specifically to a non-intrusive load identification method and system based on wavelet transform. The method includes: performing wavelet decomposition on the collected current time-series signal of the household main inlet to obtain multi-layer wavelet coefficients; analyzing and removing harmonic features in each layer of wavelet coefficients, and constructing a current feature matrix of the current time-series signal; inputting the current feature matrix into a pre-trained label discrimination model, and outputting labels for the types of appliances currently connected to the household main inlet. This application aims to select and correct the wavelet coefficients used in the wavelet decomposition to remove the influence of harmonic currents, thereby improving the accuracy of non-intrusive load identification using a multi-dimensional feature matrix of the current time-series signal constructed from wavelet coefficients.

Images