Short-Term Wind Power Forecasting Model Based on Low-Rank Tensor Function Representation and Informer

With the increasing global demand for energy and heightened environmental awareness, wind energy has emerged as a pivotal renewable resource. However, the inherent intermittency of wind poses challenges to grid stability and efficient power dispatch. Accurate wind power forecasting is essential for optimizing the integration of wind energy into power systems, reducing curtailment, and ensuring reliable power supply. This paper proposes a novel hybrid deep learning model leveraging Low-Rank Tensor Function Representation (LRTFR) and the Informer model to address the challenges of missing data and long-time series processing. The LRTFR method can effectively compensate for incomplete data due to wind turbine sensor failures, etc., while the Informer model solves the efficiency problem caused by long time series through its advanced self-concerned mechanism and can significantly improve forecasting accuracy. The experimental results show that the proposed LRTFR_Informer model outperforms the existing methods and achieves excellent accuracy in the wind power forecasting task.