基于Kriging模型的调心滚子轴承优化设计  

Kriging Model Based Optimization of Spherical Roller Bearing

作　　者：杨霞[1] 王先正 赵春江[1] 寇保福[1] 张涨 YANG Xia;WANG Xianzheng;ZHAO Chunjiang;KOU Baofu;ZHANG Zhang(School of Mechanical Engineering,Taiyuan university of Science and Technology,Taiyuan 030024,China)

机构地区：[1]太原科技大学机械工程学院,太原030024

出　　处：《机械设计与研究》2021年第6期101-106,111,共7页Machine Design And Research

基　　金：山西省应用基础研究计划基金项目(201901D111240);山西省科技重大专项(20181102015)。

摘　　要：为了提高调心滚子轴承的综合性能,对其结构参数进行分析和优化设计。首先,建立满足精度要求的轴承单个滚子静力学模型。然后,选取接触角、滚子半径、滚子长度和滚子轮廓半径作为设计变量,以接触应力最小、切应力最小和质量最小为目标函数,考虑几何结构的约束,采用最优空间填充设计(OSF)进行试验设计。基于试验设计的结果,采用Kriging模型建立目标函数的响应面优化模型,并通过自适应细化和手动细化两种方法来提高模型的精度。最后利用多目标遺传算法(MOGA)得到最佳优化结果,并通过设计变量对目标函数的影响分析和目标函数对设计变量的局部敏感度分析来验证优化结果的可靠性。结果表明:优化结果可靠且优化后模型的接触应力、切应力和质量分别降低了15.17%、32.13%和0.47%。To improve the comprehensive performance of spherical roller bearings.the structural parameters are analyzed and optimized.First,a static model of a single roller of a bearing is established that meets the accuracy requirements.Then,the contact angle,radius of the rller,the roller length and the roller radius contour are selected as the design variables.In order to minimize contact stress,shear stress and the minimum quality as the objective function,considering constraints geometry,experimental design is conducted using the optimal spac-illig design(OSF).Based on the results of the experimental design,the Kriging model is used to establish the response surface optimizaion model of the objective function,and the accuracy of the model is improved by two methods of adaptive refinement and manual refinement.Finally,the multi-objective genetic algorithm(M0GA)is used to obtain the best optimization results,and the reliability of the optimization results is verifed through the analysis of the influence of design variables on the objective function and the analysis of the local sensitivity of the objcive function on the design variables.The results show that the optimization results are reliable and the contact stress,shear stress and quality of the optimized model are reduced by 15.17%,32.13%and 0.47%,respectively.

关 键 词：调心滚子轴承 响应面优化 最优空间填充设计(OSF) KRIGING模型 多目标遗传算法(MOGA) 

分 类 号：TH133.3[机械工程—机械制造及自动化]