Model-based analysis of a heat pump cascade system using seawater and ammonia as working fluids

Large-scale heat pumps are a key-technology for decarbonizing industrial and residential heating demands. A promising novel technology for a large-scale heat pump which uses seawater as heat source to supply district heating has been developed and installed in Aarhus, Denmark. The system is constructed as a cascade system with an open steam compression cycle as the bottom cycle and an ammonia cycle as top cycle. To ensure optimal operation of heat pumps and increase their reliability, digital twin-based solutions are a promising approach. This article presents a developed dynamic model made in Dymola, which is used as the basis for a digital twin of the cascade heat pump system. The simulation model predicted a COP of 3.35 in a part load case for the heat pump. The study further presents suitable modelling approaches and parametrization considering the trade-off between modelling detail and accuracy. The dynamic model was used in a novel optimization framework where the aim was to perform continuously set-point optimization of the steam compressor for maximizing the COP in a large-scale heat pump. The presented framework was found to be a promising approach for developing further digital twin-based services, and for using it as a basis for optimizing the operation of large-scale heat pump systems. This was exemplified with a suggested set point optimization for the compressor speed in the bottom steam cycle, which yielded an increase of 1.5 % for the COP compared to a representative operating point for the system without using the developed optimization framework.