Wind speed forecasting based on Time series - Adaptive Kalman filtering algorithm

Short-term wind speed forecasting is important for the accuracy of wind power prediction. Based on convenience to establish the Time series prediction model and high accuracy of Kalman filtering algorithm, this paper proposes a hybrid algorithm to forecast wind speed combining Time series analysis and Kalman filter algorithm. First using Time series analysis theory, set up the regression forecasting model of wind speed sequence and then establish the state equation and measurement equation of Kalman filter. Because the input noise covariance and measurement noise covariance of the Kalman filtering method take the fixed value, thus we use the Adaptive Kalman filtering method, to realize hybrid forecast of the wind speed sequence. Simulation results showed that the proposed hybrid algorithm can effectively improve the predictive accuracy of wind speed, and can also solve the time delay of Time series method to predict the wind speed and lack of adaptability of Kalman filtering method exists.