Automatic Multi-steps Prediction Modelling for Wind Power Forecasting

Wind power is an important source of renewable energy. Owing to the randomness of wind speed, wind power forecasting has always been a challenging issue and is of paramount significance to the operation safety of power systems. In this paper, we proposed a hybrid method for multi-steps wind power forecasting, which combines the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Long Short-Term Memory (LSTM) neural network with modified Genetic Algorithm optimization. The unknown parameters of LSTM and component aggregation weights in result reconstruction are optimized to improve the forecasting performance. A case study concerning the real wind power datasets from ELIA is carried out to validate the effectiveness of the proposed method.