Short-Term Wind Power Output Forecasting Based on ISSA-Optimized CEEMDAN-BILSTM

Given the significance of wind power forecasting in enhancing the regulation capacity of power systems and improving wind energy utilization, this study proposes a shortterm wind power output prediction model based on ISSAoptimized CEEMDAN-BILSTM. The model first performs correlation analysis to screen meteorological feature sets. Subsequently, an improved sparrow search algorithm (ISSA) incorporating refraction reverse learning, sine-cosine strategy, and Levy flight is employed to optimize both Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Bidirectional Long Short-Term Memory (BILSTM). The subsequences obtained through CEEMDAN decomposition are then fed into the BILSTM model for regression prediction. Finally, the predicted values of the subsequences are superimposed to derive the final forecasting results. The research demonstrates that this model significantly improves the accuracy of wind power output prediction.