Adaptive support segment based short-term wind speed forecasting

Abstract

Accurate wind speed forecasting plays a crucial role in the efficient use of wind energy. This is, however a challenging task due to the volatile and the random nature of the wind speed. To improve the accuracy of wind speed forecasting, we propose the adaptive support segment to characterize the time-varying nature of the wind speed and the inertial property of the airflow. We then propose a hybrid model for wind speed forecasting. In this model, the historical wind speed data is decomposed into narrowband modes using the Variant Mode Decomposition (VMD) method, the adaptive support segment is then estimated and the future measurements for the narrowband modes are forecasted using a modified Reformer model, and the future measurements are finally added together to rebuild the future wind speed. The efficiency of the proposed model is validated through comparative experiments on the wind speed data measured at two wind farms.