基于遗传算法的板翅式换热器优化策略  

Optimization Strategy of Plate-fin Heat Exchanger Based on Genetic Algorithm

作　　者：张帅[1,2] 欧阳峥嵘 ZHANG Shuai;OUYANG Zheng-Rong(High Magnetic Field Laboratory,Chinese Academy of Sciences,Hefei 230031,China;Science Island Branch,Graduate School of USTC,Hefei 230026,China)

机构地区：[1]中国科学院强磁场科学中心,合肥230031 [2]中国科学技术大学研究生院科学岛分院,合肥230026

出　　处：《工程热物理学报》2021年第11期2919-2925,共7页Journal of Engineering Thermophysics

摘　　要：换热器的传热温差与阻力压降是一对相互矛盾的参数。在低温板翅式换热器设计中需要对换热器结构及翅片进行优化,使传热温差与阻力压降达到最优。在约束条件范围内换热器与锯齿形翅片的几何参数可能是连续的或离散的。本文利用基于基因遗传算法的优化方法对其进行计算,分别利用SEGA算法进行单目标优化分析,利用NSGA Ⅱ算法进行多目标优化分析。为了避免"熵产悖论",在单目标优化中分别采用修正熵产数与耗散数最小作为适值函数,而在多目标优化中分别以换热器的传热温差及阻力压降造成的修正熵产数与耗散数最小作为适值函数,并利用TOPSIS(优劣解距离法)策略在Pareto解中得到最优解。结果表明,单目标优化中以修正熵产数或耗散数为目标函数具有相同的优化效果,而多目标优化的结果更具针对性,以修正熵产数与耗散数为目标函数分别对压降及传热温差有更好的优化效果。The heat transfer temperature difference and pressure drop of the heat exchanger are a pair of contradictory parameters. In the design of cryogenic plate-fin heat exchangers, the heat exchanger structure and the fins need to be optimized to reach the best performance. The geometric parameters of the heat exchanger and offset fins can be continuous or discrete within the constraints. The optimization methods based on the genetic algorithm have been adopted in this paper. Single-objective optimization analysis is performed using SEGA, and multi-objective optimization analysis is performed using NSGA Ⅱ. In order to avoid the ‘entropy generation paradox’,in the single-objective optimization, the minimum corrected entropy generation number and the entransy dissipation number are used as the fitness functions. In the multi-objective optimization, the minimum corrected entropy generation number and the entransy dissipation number caused by the heat transfer temperature differences and pressure drops are used as the fitness functions. And the TOPSIS strategy is used to obtain the best solution of the multi-objective optimization in the Pareto solutions. The results show that the two objective functions have the same optimization effect in the single-objective optimization. The results of multi-objective optimization are more targeted. Taking the modified entropy generation number and the entransy dissipation number as the objective functions have better optimization effects on pressure drops and heat transfer temperature differences,respectively.

关 键 词：板翅式换热器 遗传算法 优化 熵产 耗散 

分 类 号：TK172[动力工程及工程热物理—热能工程]