Probabilistic ultra-short-term solar photovoltaic power forecasting using natural gradient boosting with attention-enhanced neural networks

作者：Zhe Song, Fu Xiao, Zhe Chen, Henrik Madsen · 发表于：Energy and AI · 年份：2025 · DOI：10.1016/j.egyai.2025.100496 · 被引用次数：34 · 研究领域：Solar Radiation and Photovoltaics、Energy Load and Power Forecasting、Photovoltaic System Optimization Techniques

• NGBoost and neural networks are integrated for probabilistic PV power forecasting. • A hybrid deep neural network is employed for automatic feature extraction. • The proposed framework enhances the reliability and sharpness of probabilistic forecasts. • PV power forecasting uncertainty is effectively quantified with high accuracy. Probabilistic forecasting provides insights in estimating the uncertainty of photovoltaic (PV) power forecasts. In this study, an innovative probabilistic ultra-short-term PV power forecasting framework that integrates natural gradient boosting (NGBoost) and deep neural networks is developed. Specifically, an attention-enhanced neural network combining convolutional neural networks (CNN) and bidirectional long short-term memory (BiLSTM) networks is employed for feature engineering to extract abstract features from time-series data. The extracted features are then fed into an optimized NGBoost model to yield probabilistic forecasts. In comparison to the benchmark models, i.e., the recently reported quantile regression (QR)-based deep learning methods and NGBoost, the proposed model demonstrates an enhanced ability to capture variation patterns in PV power output, further improving the forecast skill score by approximately 15-60% in deterministic forecasting. In terms of probabilistic forecasting, the proposed model shows superior forecast reliability and sharpness compared to all benchmark methods. Its continuous ranked probability score (CRPS) ran...