(489c) Distributed Economic Predictive Control of Integrated Energy Systems for Enhanced Synergy and Grid Response: A Decomposition and Cooperation Strategy

With the growing concerns regarding energy and environmental issues, the popularity of complex and integrated energy systems has increased as they aim to achieve higher energy efficiency and lower environmental costs [1]. Integrated energy systems (IESs) with tightly integrated distributed energy subsystems have emerged as a promising alternative to conventional centralized power plants [2]. Typically functioning as a prosumer, an IES comprises of customers and a diverse range of operating units, such as renewable energy generation systems, generators, chillers, heat supply units, and energy storage. The close integration of these units via material, information, and energy flows results in a complex process network. Due to the variety of operating units, IESs have significant potential as participants in the grid response, enhancing the grid's reliability and boosting the system's profitability. However, designing an appropriate control scheme for IESs is a real challenge due to the strong interactions and tight connections between the various operating units.

Due to the dynamic time-scale multiplicity and complex spatial connectivity of IESs, the direct application of centralized control architecture to IESs’ complex structures can result in computational burden, non-robustness, and ill-conditioned optimization problems [3,4]. An alternative approach is to partition the IES into smaller subsystems and design a non-centralized control framework to operate the IES in a modular fashion. Subsystem decomposition is a crucial step in the development of non-centralized control strategies. Prior research has attempted to manage distributed energy systems through the use of distributed or decentralized control systems based on empirical subsystem configurations [5,6]. However, improper subsystem decomposition can result in reduced control performance [3,7], as distributed/decentralized controllers depend on multiple local decision-making agents to coordinate their actions for system-wide synergies. A systematic subsystem decomposition approach capable of dealing with dynamic time-scale multiplicity and complex spatial connectivity and contributing to the development of cooperative distributed control architectures has been lacking for multi-energy or distributed energy systems.

In order to achieve optimal modular management, a systematic subsystem decomposition method and cooperative distributed economic model predictive control (DEMPC) for IESs are proposed. The decomposition procedure involves two steps to address the complexity of IESs: vertical time-scale separation based on dynamic time-scale multiplicity and horizontal community detection based on spatial connectivity. Qualitative analysis provides a basic guideline for vertical and horizontal decomposition. In complex multi-energy systems, vertical decomposition should be implemented first to facilitate the design of distributed control frameworks. Using the proposed subsystem configuration, a cooperative DEMPC is developed with global objectives of meeting rapid response to the grid's request, customers' cooling demand, and economic gain. In the DEMPC, multiple local agents are sequentially and iteratively performed, sharing their latest information on evaluated actions and local subsystems. These agents can coordinate their decision-making actions to leverage the operating units for the entire system's short-term synergies. Extensive simulations demonstrate the applicability, effectiveness, and superiority of the proposed subsystem decomposition and distributed control framework. The developed DEMPC based on the proposed subsystem and information exchange significantly enhances the system's dynamic performance, including prompt-precise response to the grid's response and boosting operational profit while maintaining indoor temperature.

References

[1] Guo, Min, Mingchao Xia, and Qifang Chen. "A review of regional energy internet in smart city from the perspective of energy community." Energy Reports 8 (2022): 161-182.

[2] Arent, Douglas J., et al. "Multi-input, multi-output hybrid energy systems." Joule 5.1 (2021): 47-58.

[3] Daoutidis, Prodromos, Wentao Tang, and Andrew Allman. "Decomposition of control and optimization problems by network structure: Concepts, methods, and inspirations from biology." AIChE Journal 65.10 (2019): e16708.

[4] Wu, Long, et al. "Economic model predictive control of integrated energy systems: A multi-time-scale framework." Applied Energy 328 (2022): 120187.

[5] Jin, Yuhui, Xiao Wu, and Jiong Shen. "Power-heat coordinated control of multiple energy system for off-grid energy supply using multi-timescale distributed predictive control." Energy 254 (2022): 124336.

[6] De Lorenzi, Andrea, et al. "Predictive control of a combined heat and power plant for grid flexibility under demand uncertainty." Applied Energy 314 (2022): 118934.

[7] Yin, Xunyuan, et al. "Community detection based process decomposition and distributed monitoring for large‐scale processes." AIChE Journal 68.11 (2022): e17826.