S-CO_(2)箔片气体动压轴承静动态特性分析  

Static and dynamic analysis of gas foil bearings lubricated by supercritical carbon dioxide

作　　者：曲智旭 郝开元[2] 杨靖贵 朱鹏程 刘水华 李文俊[1] 冯凯[1] QU Zhixu;HAO Kaiyuan;YANG Jinggui;ZHU Pengcheng;LIU Shuihua;LI Wenjun;FENG Kai(State Key Laboratory of Advanced Design and Manufacturing Technology for Vehicle,Hunan University,Changsha 410082,China;Beijing Aerospace Propulsion Institute,Beijing 100076,China)

机构地区：[1]湖南大学整车先进设计制造技术全国重点实验室,长沙410082 [2]北京航天动力研究所,北京100076

出　　处：《振动与冲击》2024年第14期131-141,共11页Journal of Vibration and Shock

基　　金：国家重点研发计划课题(2021YFF0600208)。

摘　　要：以超临界二氧化碳(supercritical carbon dioxide, S-CO_(2))为介质的动力循环系统因其结构紧凑、效率高、能源清洁等优势受到广泛的关注。针对S-CO_(2)介质径向箔片气体动压轴承建立力学计算模型,通过物性数据库获取S-CO_(2)物性数据并基于反向传播(back propagation, BP)神经网络训练S-CO_(2)物性模型,与变黏度变密度的湍流气体润滑雷诺方程进行耦合。利用热平衡计算轴承气膜的平均温升,计算S-CO_(2)径向箔片气体动压轴承的静态特性并对程序进行验证。基于小扰动法计算动态刚度与阻尼系数,分析雷诺数、静态承载和轴承名义间隙对S-CO_(2)介质径向箔片轴承动态特性的影响规律。结果表明,利用神经网络训练获得的S-CO_(2)物性模型,获取到的不同温度与压力状态下的S-CO_(2)物性值可靠性相当高。适当增大局部雷诺数,减少名义间隙,提高静态载荷,可以获得更好的轴承动态性能。The power circulation system that utilizes supercritical carbon dioxide(S-CO_(2))as the medium has garnered widespread attention due to its compact structure,high efficiency,and energy cleaning.A mechanical model was built for gas foil bearings lubricated by S-CO_(2).The property data of S-CO_(2) were obtained from the database,and the property model of S-CO_(2) was trained based on the back propagation(BP)neural network.The property model was coupled with the Reynolds equation of turbulent gas lubrication with variable viscosity and density.The average temperature rise of the bearing gas film was calculated by using the thermal balance equation.The static mechanical model was verified,and the static characteristics of gas foil bearings lubricated by S-CO_(2) were calculated.Based on the small perturbation method,the dynamic stiffness and damping coefficient were calculated,and the influences of Reynolds number,static load,and nominal clearance on the dynamic characteristics of gas foil bearings lubricated by S-CO_(2) were analyzed.The results show that the S-CO_(2) property model obtained through neural network training embraces high reliability.By appropriately increasing the Reynolds number,reducing the nominal clearance,and increasing the static load,a better dynamic performance of the gas foil bearings can be obtained.

关 键 词：超临界二氧化碳 径向箔片轴承 神经网络 静态特性 动态特性 

分 类 号：TH133.37[机械工程—机械制造及自动化]