基于响应面模型与多岛遗传算法的高低齿迷宫密封多参数优化  

Multi-parameter Optimization of a Staggered Labyrinth Seal Based on Response Surface Model and Multi-island Genetic Algorithm

作　　者：钱照生 杨爱玲[1,2] 陈二云 张万福[1] 李国平[3] QIAN Zhaosheng;YANG Ailing;CHEN Eryun;ZHANG Wanfu;LI Guoping(School of Energy and Power Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;Shanghai Key Laboratory of Multiphase Flow and Heat Transfer inPower Engineering,Shanghai 200093,China;Shanghai Marine EquipmentResearch Institute,Shanghai 200031,China)

机构地区：[1]上海理工大学能源与动力工程学院,上海200093 [2]上海市动力工程多相流动与传热重点实验室,上海200093 [3]上海船舶设备研究所,上海200031

出　　处：《动力工程学报》2020年第10期786-792,共7页Journal of Chinese Society of Power Engineering

基　　金：国家自然科学基金资助项目(51106099,50976072,51875361)。

摘　　要：以泄漏量最小为优化目标开展高低齿迷宫密封的多参数优化设计研究,建立高低齿迷宫密封几何参数化模型与二阶响应面模型,并应用多岛遗传算法进行寻优计算。结果表明:高低齿迷宫密封二阶响应面模型的R^2为0.96271,可有效降低优化计算量;由多岛遗传算法获得了一组最优结构参数,与CFD计算结果相符,所得泄漏量误差约为6.8%;密封齿数、凸台高度和凸台的位置对高低齿迷宫密封泄漏特性影响较大,且存在最佳值,是高低齿迷宫密封设计与改造的关键几何参数。Researches were conducted on the multi-parameter design and optimization of a staggered labyrinth seal with minimum leakage flow as the optimization goal,during which,a parameterized geometric model and a second-order response surface model were established for the seal,while the multi-island genetic algorithm was used to carry out corresponding calculation and optimization.Results show that the R^2 of the second-order response surface model for the staggered labyrinth seal is 0.96271,which can effectively reduce the amount of optimization calculation.The optimal set of structural parameters obtained by the multi-island genetic algorithm agrees well with the CFD results,with an error in leakage flow calculation being about 6.8%.Structural parameters,such as the number of sealing teeth,and the height and position of the boss,have a great impact on the leakage characteristics of the staggered labyrinth seal,and there are optimal values,which are the key geometric parameters for the design and modification of staggered labyrinth seals.

关 键 词：高低齿迷宫密封 二阶响应面模型 多岛遗传算法 泄漏特性 优化设计 

分 类 号：TK14[动力工程及工程热物理—热能工程]