Thermal performance, parametric analysis, and multi-objective optimization of a direct-expansion solar-assisted heat pump water heater using NSGA-II and decision makings

Direct-expansion solar-assisted heat pump water heaters (DX-SAHPWHs) are conducive to the environment due to the use of ambient energy and solar radiation. This paper aims to design and develop a thermodynamic model and a multi-criteria optimization of thermal performance for a DX-SAHPWH using R-134a as the working refrigerant that supplies domestic hot water for a typical building throughout the whole year. The thermodynamic model of the DX-SAHPWH is developed so that the performance of the system is solved through the try and error technique with the least initial data of the thermodynamic cycle of DX-SAHPWH. After the validation of the thermodynamic model, the performance of the DX-SAHPWH is analyzed for a typical building located at the temperate climate of Iran. Then, a parametric study is conducted to identify how various design parameters may affect the performance of the DX-SAHPWH system. In the current investigation, the effects of design parameters including solar radiation intensity, ambient air temperature, outlet water temperature of the condenser, solar collector area, compressor speed, length of tube in the condenser, external diameter of the tube in collector plate, fin thickness, and thermal conductivity of collector plate on the performance of the DX-SAHPWH are investigated. Besides, single and bi-criteria optimizations are carried out using genetic algorithm (GA) to obtain the optimal solutions of design parameters, where the coefficient of performance (COP) and the solar collector efficiency (SCE) are selected as two fitness functions. The optimal solutions achieved from the bi-criteria optimization process will be given as Pareto frontier. The final optimum solution from the available solutions on the Pareto optimal frontier is selected using decision-making methods, such as LINMAP, TOPSIS, and Shannon's Entropy. The comparison of single and bi-criteria optimization results illustrate that the bi-criteria optimization method yields more proper results than single ones, mainly because of the lower deviation index from the ideal solution. The results of bi-criteria optimization show that SCE decreases 1.6% compared to the initial model while the COP increases close to 20% that makes the optimum solution more desirable compared to the single ones. Moreover, the performance of the optimized DX-SAHPWH system is compared with the initial model in each month of the year. The results indicate that the performance of the optimized DX-SAHPWH is highly improved so that the working hours of the system decrease close to 109 h compared to the initial model during the whole year.