东北大学学报:自然科学版 ›› 2018, Vol. 39 ›› Issue (6): 850-855.DOI: 10.12068/j.issn.1005-3026.2018.06.018

• 机械工程 • 上一篇    下一篇

基于人工神经网络的航空轴承疲劳可靠性分析

金燕1,2, 刘少军1   

  1. (1. 中南大学 机电工程学院/高性能复杂制造国家重点实验室, 湖南 长沙410083; 2. 常州工程职业技术学院 机电与汽车工程学院, 江苏 常州213164)
  • 收稿日期:2017-02-16 修回日期:2017-02-16 出版日期:2018-06-15 发布日期:2018-06-22
  • 通讯作者: 金燕
  • 作者简介:金燕(1981-),女,湖北崇阳人,中南大学博士研究生,常州工程职业技术学院副教授; 刘少军(1955-),男,湖南娄底人,中南大学教授.冯明杰(1971-), 男, 河南禹州人, 东北大学副教授; 王恩刚(1962-), 男, 辽宁沈阳人, 东北大学教授,博士生导师.
  • 基金资助:
    国防预研项目(8130208).国家自然科学基金资助项目(51171041).

Fatigue Reliability Analysis of Aviation Bearings Based on ANN

JIN Yan1,2, LIU Shao-jun1   

  1. 1. State Key Laboratory for High Performance Complex Manufacturing/School of Mechanical and Electrical Engineering, Central South University, Changsha 410083, China; 2. Institute of Mechanical and Auto Engineering, Changzhou Vocational Institute of Engineering, Changzhou 213164, China.
  • Received:2017-02-16 Revised:2017-02-16 Online:2018-06-15 Published:2018-06-22
  • Contact: JIN Yan
  • About author:-
  • Supported by:
    -

摘要: 提出一种人工智能方法进行航空轴承疲劳可靠性分析.通过二次多项式近似拟合温度场效应,建立热弹流润滑效应下航空轴承接触应力分析模型,同时考虑热弹流润滑效应、材料属性以及疲劳强度修正系数的随机性,结合应力-强度干涉理论,运用人工神经网络法完成疲劳可靠性分析,基于改进的一次二阶矩法完成可靠性灵敏度分析.数值算例表明,建立的可靠性分析模型能正确反映热弹流润滑效应对航空轴承接触疲劳的影响.与传统蒙特卡罗方法相比,提出的智能方法具有良好的全局搜索能力和高效的计算性能,并通过无交互方差分析滚动轴承疲劳试验对可靠性灵敏度分析结果进行了验证.

关键词: 接触疲劳, 热弹流润滑, 航空轴承, 可靠性, 人工神经网络, 遗传算法

Abstract: An intelligent method is proposed to complete the contact fatigue reliability analysis of aviation bearings. The temperature field is approximated using quadratic polynomial with intercrossing term, and the stress model under thermal elastohydrodynamic lubrication (EHL) is set up. Considering the randomness of the thermal EHL, material properties and fatigue strength correction factors, the probabilistic reliability analysis model is established using artificial neural network (ANN), and the reliability sensitivity analysis is completed based on the advanced first order second moment (AFOSM). The numerical example shows that the established probabilistic reliability analysis model could correctly reflect the effect of the thermal EHL on contact fatigue of aviation bearings, and the proposed intelligent method has an excellent global search capability as well as a highly efficient computing performance compared with the traditional Monte Carlo method, and then the results are verified with the fatigue life test of rolling bearings considering the non-interacting variance analysis.

Key words: contact fatigue, thermal elastohydrodynamic lubrication (EHL), aviation bearing, reliability, artificial neural network (ANN), genetic algorithm (GA)

中图分类号: