Optimal Scheduling Model of a Battery Energy Storage System in the Unit Commitment Problem Using Special Ordered Set

• Published in: Energies (Basel) Vol. 15; no. 9; p. 3079
• Main Authors: Do, Insu, Lee, Siyoung
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01-05-2022
• Subjects: battery energy storage system (BESS); Charging; Distributed generation; Efficiency; Energy storage; Error correction; Error reduction; Internal energy; Linear programming; microgrid; mixed-integer linear programming (MILP); Operational problems; Optimization; piecewise linearization; Power conditioning; Power efficiency; Power management; Renewable resources; Scheduling; special ordered set of type 2 (SOS2); Unit commitment
• ISSN: 1996-1073; 1996-1073
• DOI: 10.3390/en15093079

Nonlinear characteristics of a battery energy storage system (BESS) may cause errors in the stored energy between the operation plan and the actual operation. These errors may hinder the reliability of the power system especially in environments such as microgrids with limited power generation resources and high uncertainty. This study proposes a method to alleviate the occurrence of such errors in the charging/discharging scheduling process of the BESS by piecewise linearizing its nonlinear characteristics. Specifically, the stored energy in a BESS that changes nonlinearly according to the size of the charging/discharging power was modeled using the special ordered set of the type 2 (SOS2) method. The proposed model and the typical BESS-operation models with constant power conditioning system (PCS) input/output power efficiency were applied to the unit commitment (UC) problem in a microgrid environment, and the results were compared with the actual operation results. The proposed model operated similarly to the actual operation compared to the typical model, reducing the error in charging/discharging energy. Consequently, the proposed model was made cost-effective by reducing the cost of error correction and reduced the risk of deviating from operating range of the BESS. This study demonstrates that the proposed method can efficiently solve the operational problems caused by the nonlinear characteristics of BESS.