Calculation of Line Loss with Machine Learning in Distribution Network Considering Power Supply Curve and Load Distribution Characteristics

In order to make full use of power supply curve and load distribution characteristics and improve the efficiency of line loss calculation, this paper proposes a method using machine learning to calculate distribution network line loss. First of all, the 24-hour power supply curve of representative day, the load monthly active electricity energy and the distribution line length are chosen as the original data. Calculate the sum and standard deviation of the monthly active power energy of each distribution line, and principal component analysis (PCA) is used to reduce the dimension of the 24-hour active power supply data. Then, a neural network model is established to learn the input features, and a complex nonlinear relationship between the input features and the line loss is fitted. Finally, to verify the validity and practicability of the proposed method, simulation is applied to a certain area with 217 distribution lines, and the result is satisfied.