ACM e-Energy · 2025
Learning-Augmented Online Control for Decarbonizing Water Infrastructures
ACM International Conference on Future and Sustainable Energy Systems
Can learned pump control reduce carbon costs while meeting safety constraints?
LAOC restricts online decisions to safe action sets while exploiting learned predictions for pump scheduling. Analysis characterizes the relationship between safety requirements and operating costs, with a building water supply case study.
Research themes
Connections
Guarantees meet physical infrastructure
Aggregate flexibility communicates feasible actions for coordinated charging. Robust dispatch and safe water control connect performance guarantees to operational costs and physical constraints.
Open questions
Coordinated low carbon infrastructure
Can distributed energy and water resources adapt together under changing conditions?
Explore open directionsCite this paper
Jianyi Yang, Pengfei Li, Tongxin Li, Adam Wierman, Shaolei Ren. Learning-Augmented Online Control for Decarbonizing Water Infrastructures. ACM International Conference on Future and Sustainable Energy Systems, 2025. https://doi.org/10.1145/3679240.3734595
BibTeX
@inproceedings{tongxin-water-control,
title = {{Learning-Augmented Online Control for Decarbonizing Water Infrastructures}},
author = {Jianyi Yang and Pengfei Li and Tongxin Li and Adam Wierman and Shaolei Ren},
year = {2025},
booktitle = {ACM International Conference on Future and Sustainable Energy Systems},
url = {https://doi.org/10.1145/3679240.3734595},
doi = {10.1145/3679240.3734595}
}
Related papers
Robust machine-learned algorithms for efficient grid operation
RobustML combines a learned dispatch algorithm with a bound on worst case cost. A combined cycle cogeneration case study examines dispatch and ramping costs as renewable penetration increases, including performance under distribution shift.
Information Aggregation for Constrained Online Control
Entropic aggregation summarizes a local controller's feasible actions for a remote decision maker. Penalized Predictive Control can approach the regret of full information control under causal invariance and a sufficiently long prediction window.
Learning-Based Predictive Control via Real-Time Aggregate Flexibility
Maximum entropy feedback summarizes the feasible actions of controllable loads. Reinforcement learning approximates this feedback for Penalized Predictive Control, reducing information and computation requirements. Charging data demonstrates the coordination approach.