at Google ScholarTitle data
Burkel, Chris ; Griesbach, Marco ; Wirtz, Marco ; Heberle, Florian ; Jess, Andreas ; Brüggemann, Dieter:
Numerical preprocessing of ground-coupled thermal impacts on seasonal ice energy storage systems for mixed-integer linear programming.
In: Journal of Energy Storage.
Vol. 176
(2026)
.
- 123382.
ISSN 2352-1538
DOI: https://doi.org/10.1016/j.est.2026.123382
Abstract in another language
Mixed-integer linear programming (MILP) is widely applied for the optimization of complex energy systems due to its high computational efficiency. However, the requirement for linear formulations often leads to simplified thermal energy storage (TES) models, in which ambient heat losses are neglected or assumed. While acceptable for short-term storage, these simplifications can cause significant inaccuracies in seasonal applications. Ice energy storage (IES) poses an additional challenge due to its nonlinear phase-change behavior and its direct thermal interaction with the surrounding ground. This paper presents a computationally efficient approach to represent nonlinear, ground-coupled IES behavior within hourly resolved MILP-based operation optimization. The method is based on a numerical preprocessing strategy, in which the thermal behavior of the storage system is evaluated externally and subsequently incorporated into the MILP model in a linearized form. A key element of the methodology is the derivation of system-specific annual loss characteristics based on the physical properties of the storage and its geothermal coupling. The methodology is validated using a real-world trigeneration system, including a 500 m3 IES. Simulation results show good agreement with measurement data, capturing both temporal dynamics and energy quantities within a computation time of approximately 30 min. The simulated and measured annual IES temperature profiles exhibit a strong correlation, with a coefficient of determination of 0.77 for the reference year. Furthermore, the proposed preprocessing approach enables the integration of system-specific loss characteristics into design optimization and is transferable to other seasonal TES technologies, such as aquifer, cavern or pit thermal energy storage systems.