Energies, Vol. 19, Pages 455: Multi-Scale Photovoltaic Power Forecasting with WDT–CRMABIL–Fusion: A Two-Stage Hybrid Deep Learning Framework
Energies doi: 10.3390/en19020455
Authors:
Reza Khodabakhshi Palandi
Loredana Cristaldi
Luca Martiri
Ultra-short-term photovoltaic (PV) power forecasts are vital for secure grid operation as solar penetration rises. We propose a two-stage hybrid framework, WDT–CRMABIL–Fusion. In Stage 1, we apply a three-level discrete wavelet transform to PV power and key meteorological series (shortwave radiation and panel irradiance). We then forecast the approximation and detail sub-series using specialized component predictors: a 1D-CNN with dual residual multi-head attention (feature-wise and time-wise) together with a BiLSTM. In Stage 2, a compact dense fusion network recombines the component forecasts into the final PV power trajectory. We use 5-minute data from a PV plant in Milan and evaluate 5-, 10-, and 15-minute horizons. The proposed approach outperforms strong baselines (DCC+LSTM, CNN+LSTM, CNN+BiLSTM, CRMABIL direct, and WDT+CRMABIL direct). For the 5-minute horizon, it achieves MAE = 1.60 W and RMSE = 4.21 W with R2 = 0.943 and CORR = 0.973, compared with the best benchmark (MAE = 3.87 W; RMSE = 7.89 W). The gains persist across K-means++ weather clusters (rainy/sunny/cloudy) and across seasons. By combining explicit multi-scale decomposition, attention-based sequence learning, and learned fusion, WDT–CRMABIL–Fusion provides accurate and robust ultra-short-term PV forecasts suitable for storage dispatch and reserve scheduling.