Optimization of interconnected absorption cycle heat pumps with micro-genetic algorithms. (May 2017)

• Record Type: Journal Article
• Authors: Vinther, K.; Nielsen, Rene J.; Andersen, Palle; Bendtsen, Jan D.
• Is Part Of: Journal of process control. Volume 53(2017)
• Page Start: 26
• Page End: 36
• Publication Date: 2017-05
• Subjects: Global optimization -- Micro-genetic algorithm -- Heat pumps -- Absorption -- District heating; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2017.02.011
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract : Highlights: A dynamic district heating plant model, with absorption heat pumps, is presented. The model can predict solution crystallization and mass maldistribution issues. Constrained global plant optimization is solved with a micro-genetic algorithm. Real operational data from a plant in Denmark is used as baseline for comparison. Several percent reduction in expensive wood-chip energy is achieved. Abstract: Production of hot water in district heating plants needs to be adjusted on a day-to-day basis to match the expected demand and availability and prices of energy resources. However, such plants are often highly nonlinear and complex. It therefore makes sense to attempt to automate optimization of the operating conditions within the physical boundaries set by the plant equipment. In this work, we investigate the use of micro-genetic algorithms to achieve constrained global set-point optimization based on a dedicated simulation model. The model is based on a real district heating plant consisting of four interconnected LiBr–water based absorption cycle heat pumps primarily driven by a geothermal reservoir and wood chip boilers. Various scenarios are considered, and it is found that the proposed genetic algorithm is able to find combinations of valid set-points that provide savings of several percent of current operating costs compared to a baseline scenario in reasonable computation time.