Modeling flexible generator operating regions via chance-constrained stochastic unit commitment

Language
en
Document Type
Article
Issue Date
2023-05-23
First published
2020-06-01
Issue Year
2020
Authors
Singh, Bismark
Knueven, Bernard
Watson, Jean-Paul
Editor
Publisher
Springer Berlin Heidelberg
Abstract

Abstract We introduce a novel chance-constrained stochastic unit commitment model to address uncertainty in renewables’ production in operations of power systems. For most thermal generators, underlying technical constraints that are universally treated as “hard” by deterministic unit commitment models are in fact based on engineering judgments, such that system operators can periodically request operation outside these limits in non-nominal situations, e.g., to ensure reliability. We incorporate this practical consideration into a chance-constrained stochastic unit commitment model, specifically by infrequently allowing minor deviations from the minimum and maximum thermal generator power output levels. We demonstrate that an extensive form of our model is computationally tractable for medium-sized power systems given modest numbers of scenarios for renewables’ production. We show that the model is able to potentially save significant annual production costs by allowing infrequent and controlled violation of the traditionally hard bounds imposed on thermal generator production limits. Finally, we conduct a sensitivity analysis of optimal solutions to our model under two restricted regimes and observe similar qualitative results.

Journal Title
Computational Management Science
Volume
17
Issue
2
Citation

Computational Management Science 17.2 (2020): 309-326. https://link.springer.com/article/10.1007/s10287-020-00368-3

Zugehörige ORCIDs