Energy system device sizing under closed-loop operation with imperfect forecasts

The value of renewable energy devices depends on their operation, which must adapt to weather and demand using limited, imperfect forecasts that update over time. Conventional energy systems planning practices assume unlimited foresight in a single-level formulation, potentially leading to long-term investment in suboptimal device sizes. Accounting for realistic operation corresponds to a bi-level problem: the upper level selects device sizes, while the lower level contains closed-loop operation under evolving forecasts. We solve this problem using backpropagation MPC (BP-MPC), a gradient-based method, with a deterministic MPC as the lower-level controller. Using forecast errors generated by a parameterized martingale model of forecast evolution, we investigate how forecast accuracy affects the benefits of operation-aware energy systems planning and the resulting device sizes. In a case study of an energy system comprising photovoltaics, a battery, a heat pump, and thermal storage with uncertain weather and energy demands, accounting for closed-loop operation yields higher average cost savings as forecast error increases. The resulting designs achieve these savings by progressively shifting investment from battery storage to solar generation and thermal storage, which buffers forecast mismatches.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-12
DOI
https://doi.org/10.48550/arxiv.2611.16699
Primary Topic
Integrated Energy Systems Optimization
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Energy system device sizing under closed-loop operation with imperfect forecasts

John Lygeros, Kevin Wallington, Georgios Mavromatidis, Cara Koepele
arXiv (Cornell University)
Integrated Energy Systems Optimization
preprint

Energy system device sizing under closed-loop operation with imperfect forecasts

John Lygeros, Kevin Wallington, Georgios Mavromatidis, Cara Koepele
preprint en

Abstract

The value of renewable energy devices depends on their operation, which must adapt to weather and demand using limited, imperfect forecasts that update over time. Conventional energy systems planning practices assume unlimited foresight in a single-level formulation, potentially leading to long-term investment in suboptimal device sizes. Accounting for realistic operation corresponds to a bi-level problem: the upper level selects device sizes, while the lower level contains closed-loop operation under evolving forecasts. We solve this problem using backpropagation MPC (BP-MPC), a gradient-based method, with a deterministic MPC as the lower-level controller. Using forecast errors generated by a parameterized martingale model of forecast evolution, we investigate how forecast accuracy affects the benefits of operation-aware energy systems planning and the resulting device sizes. In a case study of an energy system comprising photovoltaics, a battery, a heat pump, and thermal storage with uncertain weather and energy demands, accounting for closed-loop operation yields higher average cost savings as forecast error increases. The resulting designs achieve these savings by progressively shifting investment from battery storage to solar generation and thermal storage, which buffers forecast mismatches.

arXiv (Cornell University)
Integrated Energy Systems Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon