A Method for Identifying Line Losses in Low Voltage Stations in Mountainous Areas under Distributed Photovoltaic Access

Due to the integration of distributed photovoltaics, the instantaneous changes in load in low-voltage substations increase, making it more difficult to analyze load variables and increasing the error in line loss identification. Therefore, a method for identifying line losses in low-voltage substations in mountainous areas under distributed photovoltaics integration is proposed. After standardizing the line loss of low-voltage substations in mountainous areas, the bilateral total variation filtering algorithm is used to denoise the line loss data, in order to improve the quality of the line loss data. By combining wavelet analysis with artificial neural networks, a wavelet neural network recognition model is constructed, and the denoised line loss data is input into the recognition model to complete the identification of line loss in low voltage substations in mountainous areas. The experimental results show that compared with other existing methods, the RMS error of line loss identification of the proposed method is the smallest under different number of samples. Therefore, it shows that the line loss identification performance of this method is good.