Aggregate Distributed Photovoltaic Power Joint Prediction Method Based on LSTM

Accurate prediction of photovoltaic output fluctuations is the key to daily dispatch management and safe and stable operation of the power grid. In this paper, a short-term photovoltaic power prediction model based on Kmeans and LSTM is proposed. Kmeans is used to cluster the photovoltaic power generation of the initial training set and the prediction day, and the LSTM is trained on the initial training set data of each category, and the corresponding LSTM is used to predict the photovoltaic power according to the category of the predicted sample. Finally, the actual historical data of the power grid is used to simulate and analyze the proposed method. The results show that the proposed method can provide predictive data support for the distributed photovoltaic output fluctuation model of the power grid.