Open access
Jul 2026
Photovoltaic Power Generation Forecasting Based on CNN-LSTM-PINNs Hybrid Model
Results indicate that embedding physical constraints into data-driven forecasting models can improve PV power prediction accuracy, and shows stronger robustness and generalization performance under heterogeneous operating conditions, although its effectiveness is contingent on relatively stable data distributions.
Jiabo Gou, Xiaoqiao Liao, Sheng Li et al.
· Energies · 0 citations