Mitigating large-scale offshore wind farm shutdowns during a severe storm using an adaptive robust optimization approach
Damanik, O.; Van Hertem, D.; Ergun, H. (2025). Mitigating large-scale offshore wind farm shutdowns during a severe storm using an adaptive robust optimization approach. Sustainable Energy Grids & Networks 43: 101816. https://dx.doi.org/10.1016/j.segan.2025.101816
In: Sustainable Energy Grids & Networks. Elsevier: Amsterdam. ISSN 2352-4677; e-ISSN 2352-4677, more
|  |
| Author keywords |
Unit commitment;Robust optimization;AC/DC grids;Power system resilience;Wind farm shutdown;Storm;power |
| Authors | | Top |
- Damanik, O.
- Van Hertem, D., more
- Ergun, H., more
|
|
|
| Abstract |
This paper presents an adaptive robust optimization (ARO) approach to address a day-ahead unit commitment and dispatch planning to anticipate a large-scale offshore wind farm (OWF) uncertainty due to possible shutdown operations caused by severe storms. In this paper, we construct the uncertainty set of the OWF shutdown operation based on the wind speed and OWF power forecasts. The ARO model is formulated as a two-stage robust optimization model. In the first stage, the day-ahead unit commitment and dispatch decisions and constraints are implemented. In the second stage, we optimize the redispatch decisions based on the obtained first-stage decisions and the uncertainty realization. In other words, the aim of the optimization model is to ensure the robustness and cost-effectiveness of the first-stage unit commitment decisions against all possible realizations of the uncertainty set. We implement the proposed ARO model on a hybrid AC/DC grid test system with a meshed HVDC offshore grid connected to two onshore AC grids. The OWF operation represents actual North Sea-based geographic locations and wind speed data. The results show that the proposed ARO model can effectively protect the system against the uncertainty of the wind farm shutdown operations while maintaining the system cost efficiency by still allowing more OWF generation in the day-ahead planning. |
|
