Short-term prediction of photovoltaic power based on quadratic decomposition and residual correction

Photovoltaic (PV) power generation is highly nonlinear and stochastic, and accurate prediction of PV power plays a crucial role in PV grid connection and power plant operation and scheduling. A short-term PV power combination prediction model based on quadratic decomposition and residual correction is proposed to improve the prediction accuracy of PV power. The quadratic decomposition method used in this case involves the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), Sample Entropy (SE), and Variational Mode Decomposition (VMD) techniques. These techniques are applied to the PV power data to obtain smoother intrinsic modal function components. Residual correction is a process that involves predicting the sequence of residuals generated from the initial prediction results and using them to correct the initial predictions to obtain the final expected values. The experiments were conducted with the measured data from the DKA (Desert Knowledge Australia) Solar center in Australia, and the results show that the proposed combined model is better than the other models in predicting the results and effectively improves the accuracy of the short-term PV power prediction.