Performance analysis and multi-objective optimization of irreversible Diesel cycle with non-ideal gas working fluid

Abstract

In the early research process, the ideal gas was taken as the research object, but in practice, the working fluid was all non-ideal gas, so it is of great significance to study performance of actual internal combustion engine with non-ideal gas. This study utilizes an irreversible Diesel cycle model, which has been established in the previous literature, and considers various irreversible loss terms and specific heat model of non-ideal gas working fluid, to perform cycle performance analysis and multi-objective optimization. Compression ratio (\(\gamma\)) is taken as optimization variable to optimize efficiency (\(\eta\)), dimensionless power (\(\overline{P}\)), dimensionless power density (\(\overline{{P_{{\text{d}}} }}\)) and dimensionless ecological function (\(\overline{E}\)). The results show that there are optimal \(\gamma\) s to maximize the four-objective functions (\(\eta_{\max }\), \(\overline{P}_{\max }\), \(\overline{{P_{{\text{d}}} }}_{\max }\) and \(\overline{E}_{\max }\)); with the rises of irreversible loss terms, the \(\eta_{\max }\), \(\overline{P}_{\max }\), \(\overline{{P_{{\text{d}}} }}_{\max }\) and \(\overline{E}_{\max }\) all drop. As freedom degree of monatomic gas changes from 1 to 3, only \(\eta_{\max }\) drops and the other three-objective functions rise. When \(\overline{P} - \eta - \overline{E} - \overline{P}_{{\text{d}}}\) is optimized and \(\gamma_{{{\text{opt}}}}\) is mainly concentrated between 3.6 and 5.3, the calculation results of \(\overline{P}_{{}}\) are distributed between 0.85 and 1. The calculation results of \(\eta\) are distributed between 0.46 and 0.52. The calculation results of \(\overline{E}\) are distributed between 0.6 and 1. The calculation results of \(\overline{{P_{{\text{d}}} }}\) are distributed between 0.9 and 1. When \(\overline{P} - \eta - \overline{E} - \overline{P}_{{\text{d}}}\) and \(\overline{P} - \overline{E} - \overline{P}_{{\text{d}}}\) are optimized, deviation indexes obtained by using LINMAP decision-making are the smallest and the best among all optimization results. Multi-objective optimization algorithm is an optimization method to solve multiple conflicting objectives by simulating the competition mechanism in nature. It can find a balance point among multiple objective extremes and thus improve comprehensive performance of Diesel cycle.