Line Loss Prediction of Low Voltage Distributions Considering Mass PV and Electric Heating

In order to analyze the impact of large-scale photovoltaic (PV) and electric heating on line loss of distribution network station area, a prediction method for main influencing factor line loss based on a deep neural network (DNN) prediction model is proposed in this paper. Firstly, influence factors of line loss in the station area are analyzed, and main influence factors of line loss are obtained by principal component analysis (PCA). Secondly, PV power generation model and thermal storage electric heating power consumption characteristic model are established, and then power flow of large-scale PV and electric heating connected to low-voltage distribution station area is calculated by using OpenDSS, and sample data is obtained; finally, main line loss factors are used as input, taking line loss of low-voltage station area as output; prediction model of line loss of low-voltage station area is established, and simulation shows that the method proposed in this paper has good prediction accuracy. The line loss prediction method in this paper can effectively solve the problem of complex calculation and large amount of calculation in traditional methods.