AUTHOR=Fadnes Fredrik Skaug , Olsen Ernst , Assadi Mohsen 

TITLE=Holistic management of a smart city thermal energy plant with sewage heat pumps, solar heating, and grey water recycling

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2023.1078603

DOI=10.3389/fenrg.2023.1078603

ISSN=2296-598X

ABSTRACT=This article presents and evaluates a modern, thermal energy plant consisting of several renewable production sources, including sewage heat pumps, thermal solar collectors, and greywater recycling. The project is a collaboration between the industrial plant designer, the municipal plant owner, and operator, and the local academic institution, and the overall target is to investigate how the experience and competence of the three partners together can lead to improved operation in terms of reduced cost, equipment wear, and reductions in energy consumption, peak loads, and greenhouse gas emissions. The partners investigate the potential for using existing measuring equipment and sensors to develop data-driven models for improved operational management and control. The industrial partner, with vast experience in the design of thermal energy plants and heat pump systems, is allowed to closely follow up on its design and increase its knowledge within the field of artificial intelligence and data-driven methods. The municipal partner is given a “free-of-charge” system review by the academic and industrial partners, with new knowledge and reduced life cycle costs and emissions as possible results. The academic partner gets access to a “living green laboratory," a unique dataset, and the opportunity to validate developed models and optimization strategies, including changing system setpoints in cooperation with the municipal operator. This article contains a literature review of methods from the AI domain applied to thermal energy plants and heat pump systems and introduces the case study energy plant. The plant represents the state-of-the-art for a medium scaled, local thermal energy production and distribution system in an existing building cluster. The design energy and emission targets are presented and compared to the operational results. Though the municipal partner can report good agreement between targets and results, a more detailed evaluation of the plant’s operation, based on access to the control system, identifies several potential improvements which AI can help accomplish. The paper concludes with a description of plans for future work and, returning to the three-part cooperation, a broader discussion of the impacts of introducing data-driven methods in real-life systems.