Evaluation of CO₂ heat pump water heater model for operational planning

Energy management is systematic activity to improve energy performances of target system, and it has already been introduced in the industrial and commercial field. Energy management system is expected to solve operational planning problem and report or suggest opportunities for performance improvement. In recent years, CO₂ heat pump water heater (HPWH) is attracting attention as one of the high-efficiency devices to meet domestic hot water demand. Equipment model is required to reflect the characteristics of the actual equipment’s performance and to have a simple structure to apply to operational planning problem and should diagnose the changes with the performance degradation overtime. However, few studies have been reported which proposes the thermodynamically-sound model of HPWH for energy management. In this paper, we proposed a simple model suitable for solving operational planning problem and diagnosing aging deterioration or degradation of equipment in energy management system. Firstly, we constructed function model consisting of a heat pump unit (HP) and a hot water storage tank (ST). HP model is simply composed of start-up, steady state, and shut-down status-transition model. The steady state model is constructed based on the Lorentz efficiencyand theoretical maximum coefficient of performance (COP) for trans-critical heat pump cycle. The Lorentz efficiency is identified by experiment on typical domestic hot water demand with variation of ambient temperature, inlet/outlet water temperature at the gas-cooler. While the heat pump system is operating, unit COP of the HP is influenced on the thermal stratification of ST. The ST model is a simplified model that can describe temperature distribution in ST. Secondly, we validated the model. Proposed model reproduced experimental values with sufficient accuracy for day-ahead planning problem. Finally, we confirmed that the proposed model is preferable in its simplicity and robust performance for wide temperature range by comparing with a conventional statistical regression model.