A distributed AGC power sharing strategy considering BESS participation factors

• Published in: Electric power systems research Vol. 217; p. 109117
• Main Authors: Zhang, Shengqi, Yu, Qicheng, Liu, Heyu, Mishra, Yateendra, Wang, Fei
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.04.2023
• Subjects: Automatic generation control; Battery energy storage system; Distributed algorithm; Fuzzy-logic theory; Battery energy storage system; Fuzzy-logic theory; Automatic generation control; Distributed algorithm
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2023.109117

•To keep the state of charge balance of battery energy storage systems, a participation factor is proposed in the cost function.•The distributed algorithm is proposed to shorten the computation time.•Massive battery energy storage systems are managed by the distributed algorithm.•The strategy considers the diverse characteristics of frequency regulation units, and can be collaborate with the current AGC system. Battery energy storage systems (BESSs) in power system automatic generation control (AGC) are regarded as an effective way to improve the frequency stability when the system has a high penetration level of renewable energy. Compared with conventional generators, BESSs have different characteristics and operation constraints, which bring challenges to AGC power sharing in an accurate and effective way. Therefore, a participation factor (PF)-based AGC power sharing strategy (APSS) is proposed in this paper, aiming at managing a large quantity of BESSs with various frequency regulation (FR) characteristics. Specifically, the concept of the BESS PF is introduced in its AGC cost function, and the PF value of each BESS is dynamically determined on the real-time AGC power and state of charge (SoC). The numerical simulation results demonstrate that the PF-based APSS is able to make the exact match between the characteristic components of AGC power and the corresponding BESSs under the condition of SoC balancing. Meanwhile, the computational burden of the control center is effectively relieved by the distributed algorithm (DA) regardless of the BESS quantity, and BESS plug-and-play is also realized.