Day-Ahead PV and EVs Power Forecasting of Distribution Network Based on LSTM

The massive access of photovoltaic system (PV) and electrical vehicles (EVs) in urban cities significantly increases the proportion of distributed generation in distribution network, which has a negative effect on the forecasting accuracy of PV and EVs due to the fluctuation and intermittence. To solve the above problems and ensure the power supply reliability and power quality of the distribution network, this paper proposes a day-ahead PV and EVs power forecasting method for distribution network based on Long-Short Term Memory (LSTM) neural network. Firstly, the structure of RNN and LSTM is analyzed. Then the model of LSTM is established, and the forecasting method based on LSTM is proposed. Finally, the real field data of PV and EVs in applied to verify the proposed method. At the end of the paper, the results show that the method proposed in this paper has a good effect on forecasting of PV and EVs.