A Physics-Aware and SHAP-Guided Adaptive Method for PV Power Forecasting 物理感知與 SHAP 引導的自適應光伏功率預測方法
- 純資料驅動光伏功率預測於夜間或低輻照段常輸出負功率，違背物理常識，影響電網調度與場站維運決策可信度。於 Autoformer、FEDformer、Informer 等 SOTA 時序骨干中引入非對稱非負損失，顯式抑制非物理負功率輸出並完成多骨干對照實驗，DKASC 上負功率顯著減少、基線 MSE 穩定改善。
- 黑箱調參易使模型過度依賴全球水平輻照度 GHI 等少數特徵，過擬合與貢獻漂移難以提前發現。建置 SHAP 對 GHI 貢獻的監測流程，並接入 TPE 貝葉斯超參搜尋，使超參區間更貼合物理因果，訓練中期可識別貢獻異常與過擬合跡象，提升可診斷性與跨季節穩定性。
- 公開資料存在季節與工況切換帶來的概念漂移，全量重訓練線上成本高。實現 SimpleADWIN 漂移偵測觸發機制，採用凍結骨干並微調末端投影層之輕量線上更新策略，相對全量重訓練線上適應耗時縮短約 92%，SOTA 模型 MSE 平均改善約 17%，形成物理—解釋—演化一體化結論。
摘要
Deep learning models for photovoltaic (PV) power forecasting often face challenges such as black-box opacity, physical law violations, and performance decay caused by concept drift. This research develops an adaptive forecasting framework guided by physics and Shapley additive explanations (SHAP). The proposed framework integrates domain physics with data-driven evolution. We use state-of-the-art (SOTA) forecasting architectures, including Autoformer, FEDformer, and Informer, as the backbone. We embed an asymmetric non-negative penalty into the loss functions to eliminate irrational negative power predictions. For diagnostic transparency, a SHAP-based introspection module monitors the global horizontal irradiance (GHR) contribution ratio to identify feature overfitting. This feedback directs a Bayesian optimization loop via the tree-structured Parzen estimator (TPE) algorithm to align model weights with physical causality. To ensure robustness in non-stationary environments, a lightweight evolution module triggered by SimpleADWIN freezes the SOTA backbone and fine-tunes the terminal projection layer. Validations on the Desert Knowledge Australia Solar Centre (DKASC) dataset demonstrate that the framework rectifies causal biases effectively. The method yields a 17% mean squared error improvement for SOTA models and accelerates online adaptation by 92% compared to global retraining.