不可逆往复式Brayton循环性能分析与优化:(2)多目标优化

Performance Analysis and Optimization for Irreversible Reciprocating Brayton Cycle:(2)Multi-Objective Optimizations

在第(1)部分导出循环功率密度表达式的基础上,应用NSGA-Ⅱ算法,以循环压缩比为优化变量,引入功率、热效率、功率密度和生态学函数为目标函数,对不可逆往复式Brayton循环进行不同目标组合的单、双、三和四目标优化,并比较了LINMAP、TOPSIS和Shannon Entropy 3种决策方式下多目标优化的结果。结果显示,多目标优化得到的偏差指数小于单目标优化得到的偏差指数,当以P^(-)、E^(-)和P_(d)^(-)或P^(-)和E^(-)为优化目标时,采用TOPSIS决策方式得到的偏差指数为0.1309;当以P^(-)和η为优化目标时,采用LINMAP决策方式得到的偏差指数为0.1309。这3种优化方案得到的结果偏差远小于单目标优化得到的结果偏差,其设计方案更接近于理想方案。研究结果可为实际热机优化提供参考。

Based on the cycle power density expression derived in Part(1),the cycle compression ratio is taken as the optimization variable,and the power output,thermal efficiency,power density and ecological function are introduced as the objective functions so as to perform the single-,bi-,tri-,and quadru-objective optimizations for the irreversible reciprocating Brayton cycle by using the NSGA-II algorithm with different objective combinations.The results of the multi-objective optimization with the three decision-making methods of LINMAP,TOPSIS and Shannon Entropy are compared.The results show that the deviation index obtained by multi-objective optimization is smaller than that obtained by single-objective optimization.When the multi-objective optimization is performed with P^(-),E^(-) and P_(d) or P^(-) and E^(-) as objective functions,the deviation index obtained by the TOPSIS decision method is 0.1309.When the multi-objective optimization is performed with P^(-) andηas objective functions,the deviation index obtained by the LINMAP decision method is 0.1309.The deviation results obtained by these three optimization schemes are much smaller than those obtained by single-objective optimization,and the design scheme is closer to the ideal scheme.The research results can provide guidance for the actual heat engine optimization.