Two-Step Wind Power Prediction Approach With Improved Complementary Ensemble Empirical Mode Decomposition and Reinforcement Learning

The strong stochastic nature of wind power generation makes it extremely challenging to accurately predict and support the planning operation modern systems with significant penetration renewable energy. This article proposes a two-step prediction method, which consists two phases: long time-scale coarse short fine correction. In phase, complementary ensemble empirical mode decomposition-based sigma point Kalman filter approach is proposed coarsely merely historical data. deep deterministic policy gradient learns from real-time weather information correct result, results in an improved accuracy. A real-life case study confirms that method can properly have better accuracy than existing techniques, thus offering viable promising alternative for predicting generation.