Research on A Forecasting Model of Wind Power based on Recurrent Neural Network with Long Short-term Memory

The forecasting of the wind power, which is very important for the stability and reliability of the grid, is challenge for the uncertainty and randomness of the wind power. This paper proposed a forecasting model of wind power based on recurrent neural network (RNN) with long short-term memory (LSTM). With the help of the sequence-dependence characteristic of RNN, the proposed model can be trained by smaller data sets but achieve more accurate forecast of wind power. LSTM is involved into the model to improve the oblivion and avoid gradient disappearance. The experimental results verify the effectiveness of the proposed forecasting model.