Toward Cost-Efficient and Reliable Unit Commitment under Uncertainty

Hrvoje Pandzic, Yury Dvorkin, Ting Qiu, Yishen Wang, Daniel S. Kirschen

Research output: Contribution to journalArticlepeer-review


Large-scale integration of wind farms causes volatile bus net injections. Although these fluctuations are anticipated, their timing, magnitude and duration cannot be predicted accurately. In order to maintain the operational reliability of the system, this uncertainty must be adequately addressed at the day-ahead generation scheduling stage. The ad-hoc reserve rules incorporated in deterministic unit commitment formulations do not adequately account for this uncertainty. Scenario-based stochastic unit commitment formulations model this uncertainty more precisely, but require computationally demanding simulations. Interval and robust optimization techniques require less computing resources, but produce overly conservative and thus expensive generation schedules. This paper proposes a transmission-constrained unit commitment formulation that improves the performance of the interval unit commitment. The uncertainty is modeled using upper and lower bounds, as in the interval formulation, but inter-hour ramp requirements are based on net load scenarios. This improved interval formulation has been tested using the IEEE RTS-96 and compared with existing stochastic, interval and robust unit commitment techniques in terms of solution robustness and cost. These results show that the proposed method outperforms the existing interval technique both in terms of cost and computing time.

Original languageEnglish (US)
Article number7115977
Pages (from-to)970-982
Number of pages13
JournalIEEE Transactions on Power Systems
Issue number2
StatePublished - Mar 2016


  • Interval optimization
  • stochastic optimization
  • uncertainty
  • unit commitment

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering


Dive into the research topics of 'Toward Cost-Efficient and Reliable Unit Commitment under Uncertainty'. Together they form a unique fingerprint.

Cite this