EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Compute-Constrained Adaptive Electricity Load Forecasting via Offline Policy Learning

Jiapei Liu, Zhigang Zhao, Li Xu, Zhaoyuan Zhu

Energies · 29.10.2026 · 19 (21) · 5155

DOI: 10.3390/en19215155 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Energies, online erschienen am 29.10.2026. Verfasst von Jiapei Liu und 3 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Electricity load forecasting services must allocate inference across models with different accuracy and compute costs under a shared daily budget.

Methods

We evaluate offline policies using only pre-call information, complete-day rolling folds, outcome-time embargoes, and a reservation rule that preserves minimum service for later origins. Four public datasets are tested at four horizons and three budgets, comparing contextual allocation with fitted Q iteration (FQI), double FQI, and neural FQI against a feasible static forecaster.

Results

All 960 formal policy evaluations satisfy the budget. The mean gains in mean absolute error (MAE) are 9.22%, 7.37%, 7.77%, and 3.83% for the four methods, respectively. The contextual gains are positive on Steel, Tetouan, and Open Power System Data (OPSD) but negative on UCI ElectricityLoadDiagrams (UCI–ECL). A disjoint deployment gate yields a 10.63% aggregate gain at 51.7% coverage with six harmful activations and 61 missed beneficial cases; this empirical risk-control filter is not a safety guarantee. The lightweight multilayer perceptron (MLP) extension and sequential audits find no uniform benefit from a larger action pool or deeper Bellman backups; FQI improves over contextual allocation only at the smallest tested switching penalty in the cache-state experiment.

Conclusions

Offline policy learning can support budget-feasible adaptive forecasting, but its benefit depends on dataset and runtime conditions. Chronological validation, explicit feasibility checks, and static fallback are needed; the results do not establish the universal superiority of sequential reinforcement learning.

Zitieren

APA

Liu, J., Zhao, Z., Xu, L., & Zhu, Z. (2026). Compute-Constrained Adaptive Electricity Load Forecasting via Offline Policy Learning. Energies, 19(21), 5155. https://doi.org/10.3390/en19215155

Vancouver

Liu J, Zhao Z, Xu L, Zhu Z. Compute-Constrained Adaptive Electricity Load Forecasting via Offline Policy Learning. Energies. 2026;19(21):5155. doi:10.3390/en19215155

BibTeX
@article{Liu2026,
  author = {Liu, Jiapei and Zhao, Zhigang and Xu, Li and Zhu, Zhaoyuan},
  title = {Compute-Constrained Adaptive Electricity Load Forecasting via Offline Policy Learning},
  journal = {Energies},
  year = {2026},
  volume = {19},
  number = {21},
  pages = {5155},
  doi = {10.3390/en19215155}
}
Kostenlos starten →