Chance-constrained unit commitment with energy storage systems in electric power systems

Ying Yi Hong*, Gerard Francesco DG Apolinario, Tai Ken Lu, Chia Chi Chu

*Corresponding author for this work

Research output: Contribution to journalJournal Article peer-review

16 Scopus citations

Abstract

The high penetration of renewables has amplified errors in forecasts of wind and solar power generation that are made to support the operation of power systems. To guarantee the security and reliability of the power system, sufficient dispatchable power generation capacity must be reserved. Various studies of the use of risk assessment to improve operational performance have been carried out. However, most methods of so doing are designed to cover a given risk level of uncertainty, which is determined from forecasts rather than actual results. As renewable generation and load capacity increase, defining the risk level without considering unit commitment (UC) may limit scheduling efficiency. In this study, chance-constrained programming is combined with goal programming to optimize risk-based UC. Solar power generation, wind power generation, and demand forecast errors are handled by the optimization of reserve schedules and adjustments of risk by imposing a penalty cost. The proposed risk-based UC model is transformed into a mixed-integer linear programming (MILP) problem using the equivalent deterministic form and piecewise linear functions to find an efficient solution. Case studies of a realistic Taiwan power system in 2025 are considered; these include peak (Summer 2025) and off-peak (Winter 2025) scenarios. Simulation results reveal that the flexible operation of a battery energy storage system (BESS) (upward/downward reserve) reduces risk costs by a significant amount (19.12%∼100%). Sensitivity analysis reveals how the risk preferences of the System Operator (SO) can be represented in terms of predefined risk levels and confidence intervals to capture the uncertainties in load, wind power generation, and solar power generation forecasts. The spinning reserve level is optimized to reduce the risk cost and the addition of BESS in flexible operations further reduces the risk cost. Lastly, pumped storage units as a downward reserve can reduce wind spillage and excess solar power generation costs.

Original languageEnglish
Pages (from-to)1067-1090
Number of pages24
JournalEnergy Reports
Volume8
DOIs
StatePublished - 11 2022
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2021 The Author(s)

Keywords

  • Battery energy storage systems
  • Chance-constrained goal programming
  • Pumped storage units
  • Risk level
  • Unit commitment
  • Upward/downward reserves

Fingerprint

Dive into the research topics of 'Chance-constrained unit commitment with energy storage systems in electric power systems'. Together they form a unique fingerprint.

Cite this