Performance Improvement of Air-Source Heat Pump Using Optimum Control Based on Artificial Neural Network

Air source heat pump (ASHP) systems are an economically viable solution for building decarbonization. However, ASHP’s energy performance deteriorates under severe conditions around a building. Several researchers have conducted studies for optimal compressor control to overcome these limitations. Despite these efforts, it is rare for a user to optimally regulate the heat pump system because manufacturers are reluctant to relinquish control to users. Therefore, in this study, we developed an artificial neural network (ANN)-based optimum control logic (OCL) to enable users to control and optimize the heat pump system without a compressor. The developed ANN-based OCL controls the secondary side working fluid of the heat pump considering general building conditions. We verified the real applicability of the ANN-based OCL using dynamic simulations. As a result of comparing the energy performance of the conventional and optimum models, the cooling/heating system COP improved by 1.52% and 3.58%, respectively, while the heat pump COP improved by 0.76% and 0.81%, respectively. These results demonstrate the potential to reduce a building's carbon footprint while maintaining comfort by enabling energy-efficient operation and stable load response of the heating and cooling system.