Pre-Trained Large Model as Distributed PV Power Forecaster: Integrating Scenario-Consistent Retrieval-Augmented Generation and Cross-Modal Semantic Alignment
IEEE Transactions on Sustainable Energy · 2026-07 · 17 (3) · 2394-2407
DOI: 10.1109/tste.2025.3650479 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Sustainable Energy, 2026-07. Verfasst von Mingyang Gao und 6 weiteren Autor:innen. Die Arbeit wurde bisher 18-mal zitiert (Crossref). Gefördert von: Science and Technology Project of State Grid.
Abstract (Originalsprache)
The rapid proliferation of distributed photovoltaic (PV) systems presents significant challenges for accurate power generation forecasting due to their inherent intermittency and spatial-temporal variability. This paper proposes a novel framework that leverages pre-trained large models for distributed PV power forecasting by integrating retrieval-augmented generation with cross-modal semantic alignment. To unlock the reasoning capabilities of pre-trained large models, we design a scenario-consistent RAG module that retrieves relevant historical PV sequences guided by long-term dependencies and temporal correlations, and integrates them through a Chain-of-Thought prompting paradigm. Furthermore, we develop a cross-modal semantic alignment mechanism to bridge the representational gap between numerical PV series data and the language semantic space of PLMs, enabling effective pattern recognition across modalities. To ensure physically meaningful forecasting within the training and inference processes, we also establish a constrained activation function that explicitly considers the inherent characteristics of PV power generation. Extensive experiments on nine real-world distributed PV datasets demonstrate that our framework achieves state-of-the-art performance, significantly outperforming six benchmark models.
Förderung
- Science and Technology Project of State Grid
Zitieren
APA
Gao, M., Zhou, S., Gu, W., Fan, J., Liu, H., Guan, A., & Xu, H. (2026). Pre-Trained Large Model as Distributed PV Power Forecaster: Integrating Scenario-Consistent Retrieval-Augmented Generation and Cross-Modal Semantic Alignment. IEEE Transactions on Sustainable Energy, 17(3), 2394-2407. https://doi.org/10.1109/tste.2025.3650479
Vancouver
Gao M, Zhou S, Gu W, Fan J, Liu H, Guan A, et al. Pre-Trained Large Model as Distributed PV Power Forecaster: Integrating Scenario-Consistent Retrieval-Augmented Generation and Cross-Modal Semantic Alignment. IEEE Transactions on Sustainable Energy. 2026;17(3):2394-2407. doi:10.1109/tste.2025.3650479
BibTeX
@article{Gao2026,
author = {Gao, Mingyang and Zhou, Suyang and Gu, Wei and Fan, Jili and Liu, Haiquan and Guan, Aobo and Xu, Honghua},
title = {Pre-Trained Large Model as Distributed PV Power Forecaster: Integrating Scenario-Consistent Retrieval-Augmented Generation and Cross-Modal Semantic Alignment},
journal = {IEEE Transactions on Sustainable Energy},
year = {2026},
volume = {17},
number = {3},
pages = {2394--2407},
doi = {10.1109/tste.2025.3650479}
}
EUVIMED