Operation of coupled multi-owner district heating networks via distributed optimization. (October 2021)

• Record Type: Journal Article
• Authors: Kaisermayer, Valentin; Muschick, Daniel; Horn, Martin; Gölles, Markus
• Is Part Of: Energy reports. Volume 7(2021)Supplement 4
• Page Start: 273
• Page End: 281
• Publication Date: 2021-10
• Subjects: District heating networks -- Energy management system -- Multi-owner -- Cooperative -- Non-cooperative; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2021.08.145
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: The growth of district heating and cooling (DHC) networks introduces the possibility of connecting them with neighbouring networks. Coupling networks can save costs by reducing operating hours of peak load or backup boilers, or free up production capacity for network expansion. Optimization-based energy management systems (EMS) already provide operators of individual DHC networks with solutions to the unit commitment and economic dispatch problem. They are especially useful for complex networks with multiple producers and integrated renewable energy sources, where incorporating forecasts is important. Time-dependent constraints and network capacity limitations can easily be considered. For coupled networks, a centralized optimization would provide a minimum with respect to an objective function which can incorporate fuel costs, operational costs and costs for emissions. However, the individual coupled networks are generally owned by different organizations with competing objectives. The centralized solution might not be accepted, as each company aims to optimize its own objective. Additionally, all data has to be shared with a centralized EMS, and it represents a single point of failure. A decentralized EMS may therefore be a better choice in a multi-owner setting. In this article, a novel decentralized EMS is presented that can handle multi-owner structures with cooperative and non-cooperative coupling. Each local EMS solves its own optimization problem, and an iterative Jacobi-style algorithm ensures consensus among the networks. The distributed EMS is compared to a centralized EMS based on a representative real-world example consisting of three coupled district heating networks operated by two companies.