Short-Term Wind Power Prediction Based on Dynamic STARMA Model

Wind power prediction plays an important role in the safe integration of wind power. In this paper, we propose a dynamic spatial-temporal autoregressive and moving average (STARMA) model for short-term wind power prediction. The adjacency matrix is determined based on the variations of wind speed and direction, and predicted by using the autoregressive integrated moving average (ARIMA) model. The proposed wind power prediction method can reduce the influence of wind direction on prediction accuracy. Experimental results demonstrate the effectiveness of the proposed prediction method.