Multi-objective optimization and performance analyses of an endoreversible rectangular cycle

Abstract

Based on the endoreversible rectangular cycle model founded in the previous literature, the rectangular cycle performance characteristics with nonlinear variable specific heats are analyzed by applying finite time thermodynamic theory when the objective functions are efficient power and power density, respectively. The different four-objective, three-objective and two-objective combinations are optimized based on the four objectives of the power density, efficient power, power output and efficiency by using the NSGA-II, then the results of the multi-objective optimizations are compared with those of the single-objective optimizations. The results show that, compared with the maximum power output point, although the heat engine efficiency at the maximum power density working point reduces by 10.97%, the heat engine size reduces by 30.84%; compared with the maximum power output point, although the heat engine power output at the maximum efficient power working point reduces by 0.067%, the heat engine efficiency increases by 0.21%. In the MO optimization results, the smallest deviation index is 0.2371; in the single-objective optimization results, the smallest deviation index is 0.2380; so the optimization result of the MO is better than those of the single-objectives.