A Bayesian-optimized CNN–GRU framework for multimodal short-term solar photovoltaic power forecasting

The integration of solar photovoltaic (PV) systems into the energy grid presents several challenges for controlling energy systems, primarily due to their intermittent nature. To manage this intermittency, accurate short-term forecasts are required. This study proposes a Bayesian-optimized sequence-aligned hybrid Convolutional Neural Network and Gated Recurrent Unit (CNN–GRU) framework to predict PV power 15 min ahead using multimodal sky images ( 6 4 × 6 4 × 4 8 ) and historical PV time-series data (16 × 1). Tree-structured Parzen Estimator (TPE) optimization with Hyperband pruning is employed to systematically search the hyperparameter space. On a standardized 1-year benchmark dataset, the proposed model achieves an overall Root Mean Square Error (RMSE) of 2 . 4 1 9 k W , a Mean Absolute Error (MAE) of 1 . 1 5 3 k W , and a Forecast Skill (FS) of 18.96%. These results demonstrate a 3.36% improvement in RMSE, a 7.83% improvement in MAE, and a 17.40% relative improvement in Forecast Skill compared to the state-of-the-art (SOTA) model, while achieving an 80.0% reduction in model parameters.