基于可靠性的裂纹齿轮点蚀寿命参数灵敏度评估  

Parameter sensitivity analysis of gear teeth life with crack in regard to surface pitting based on reliability model

作　　者：王倩倩[1] 张义民[1] 吕昊[1] 

机构地区：[1]东北大学机械工程与自动化学院,辽宁沈阳110819

出　　处：《工程设计学报》2013年第4期282-286,297,共6页Chinese Journal of Engineering Design

基　　金：国家自然科学基金资助项目(51135003;U1234208);长江学者和创新团队发展计划资助项目(IRT0816);"高档数控机床与基础制造装备"科技重大专项课题(2013ZX04011011);中央高校基本科研业务费资助项目(N110603001)

摘　　要：考虑含微裂纹齿轮在接触载荷作用下发生点蚀破坏的问题,将齿轮裂纹扩展模型与可靠性分析、可靠性灵敏度分析方法相接合,分析计算了齿轮点蚀破坏的随机参数可靠性灵敏度.首先建立了齿轮接触等效二维模型,在良好润滑条件下,初始裂纹出现在接触应力最大点处,基于赫兹接触理论利用有限元方法模拟了齿轮接触时裂纹的增长情况,并应用最小二乘法拟合裂纹长度与应力强度因子之间的函数关系,应用BCS短裂纹扩展速率模型建立了齿轮寿命模型.之后考虑齿轮参数及载荷参数的随机性,结合可靠性评估的降维-点估计方法及可靠性灵敏度分析理论,计算了可靠度对各随机参数的均值及方差的灵敏度,得到了随机参数对可靠度影响程度的排序,为防止含有微裂纹的齿轮点蚀破坏提供了理论依据.As to the problem of surface pitting of a gear pair with initial crack subjected to rolling and sliding contact, the crack growth model was combined with the reliability and reliability based sensitivity analysis method to calculate the effect of the random parameters on the crack growth. The equivalent FEM contact model based on the Hertz contact theory was first presented for evaluating the crack growth of the gear pair with fine surface finishing and good lubrication, where the cracks leading to pitting were initiated in the area of largest contact stress at certain depth under the contacting surface. And then the relationship between the crack length and the stress intensity factor was approximated by the least squares method, which was applied to the BCS model of short crack growth to evaluate the left life of the micro cracked gear teeth. Taking the uncertainty of the gear structure parameters and load into consideration, the dimension reduction- point estimate method was used for the reliability and reliability based sensitivity analysis. Sensitivities of parameters＇ mean value and standard variance with respect to life reliability were obtained by derivation method. It provides significant theoretical base for resistance of pitting failure.

关 键 词：裂纹 点蚀 可靠性 灵敏度 

分 类 号：TB114.3[理学—概率论与数理统计]