Non-intrusive residential user load decomposition method based on residual convolution module

Abstract

The invention relates to the technical field of power systems, in particular to a non-intrusive residential user load decomposition method based on a residual convolution module. The method comprises the following steps: acquiring training data and preprocessing the data; constructing and training a load decomposition model: inputting a total active power sequence in training data into a residual convolution module, learning active power features by taking a CNN model as a basis in the residual convolution module, adding original input data and feature data learned by the CNN through cross-layer connection, further inputting the obtained data into a GRU network to learn time sequence features, and outputting a predicted value of the active power of the target electric appliance; comparing the predicted value of the active power of the target electric appliance with a true value, and continuously adjusting network parameters of the load decomposition model to obtain a trained load decomposition model; and decomposing the total active power of the user to be decomposed through the trained load decomposition model to obtain an active power decomposition result of the target electric appliance. The method is high in decomposition precision.

Description

technical field [0001] The invention relates to the technical field of power systems, in particular to a non-intrusive residential user load decomposition method based on a residual convolution module. Background technique [0002] The large-scale deployment of electricity meters and the supporting communication network and data system together constitute an advanced measurement system, and the measured data can be used to obtain information such as user behavior habits and electricity consumption of various electrical appliances through data analysis techniques such as data mining; One process is load decomposition. [0003] Load monitoring is divided into two types: intrusive load monitoring and non-intrusive load monitoring. Intrusive load monitoring is to install detection equipment on each household appliance of the user in order to obtain relevant electrical appliance operation information; Difficult, and it is difficult to protect the privacy of users, and it is diff...

AI Technical Summary

Problems solved by technology

[0003] Load monitoring is divided into two types: intrusive load monitoring and non-intrusive load monitoring. Intrusive load monitoring is to install detection equipment on each household appliance of the user in order to obtain relevant electrical appliance operation information; Difficult, and it is difficult to protect the privacy of users, and it is difficult to realize large-scale promotion; non-intrusive load monitoring is also called load decomposition. Only by analyzing the data of the user's general meter, the type, switch status, and active power consumption of the user's household appliances can be obtained. Information, to more fully tap the information value of the meter data; this method is low in cost, easy to implement, easy to obtain user approval, and is the main direction of future load monitoring

[0004] The non-intrusive load decomposition method based on deep learning is one of the commonly used methods for load decomposition; there are many network variants in deep learning models, such as denoising autoencoder (DAE), RNN, seq2point conventional convolution, etc., but now Some decomposition methods have the problems of low decomposition accuracy and failure to screen electrical appliances

Images