面向一次性机械的轴承钢疲劳弹性寿命分散带多尺度预测方法

Shape instability life scatters are the fundamental basis to design disposable machinery without loss of reliability. The purpose of this project is to provide theoretical basis and methodology in predicting the shape instability life scatter of bearing steel GCr15. The research will lay the foundation for the design of screws in the disposable steering engine without loss of reliability. Firstly, by extending the computational framework of discrete dislocation dynamics and introducing the time-varying temperature field, the evolution law of dislocation density under sustained rising temperature during the shape instability failure process is theoretically and experimentally investigated. Using dislocation density as an internal variable, a multiscale simulation model is proposed by coupling the discrete dislocation dynamics and crystal plastic finite element method. Then, the micro mechanism of shape instability failure behavior is revealed through comparing the results of multiscale simulation and in-situ fatigue test. After investigating the intrinsic uncertainty factors that related to the shape instability life of metallic materials, the corresponding control parameters in the multiscale simulation model can be determined. Hereafter, according to the scales of the control parameters, a layered probabilistic quantitative analysis method is proposed to examine the uncertainty in the grain scale that transmitted from the uncertainty of control parameters in the dislocation scale. The uncertainty is introduced into the multiscale simulation by integrating the probabilistic distribution features of control parameters in grain scales. Finally, a multiscale uncertainty simulation model is established and used to calculate the shape instability life scatter by Monte Carlo method. The proposed theory and method will be verified by the ground performance test of the prototype.

疲劳弹性寿命分散带是一次性机械高可靠度设计的重要依据。本课题以GCr15轴承钢为研究对象，旨在提供理论基础和方法，预测疲劳弹性寿命分散带，指导一次性电动舵机行星滚柱丝杠的高可靠度设计。首先，拓展离散位错动力学理论计算框架，引入时变温度场，通过理论和试验研究位错密度在疲劳弹性失效过程中连续升温条件下的演化规律，建立离散位错动力学-晶体塑性有限元多尺度仿真模型。其次，对比研究多尺度仿真和原位疲劳试验结果，揭示疲劳弹性失效行为微观机理，探求影响疲劳弹性寿命的材料内禀不确定性因素，定位其映射在多尺度仿真模型中的控制参数。然后，根据控制参数所处尺度提出分层概率量化分析方法，研究位错尺度控制参数的不确定性传递至晶粒尺度引起的不确定性，结合晶粒尺度控制参数的概率分布，将不确定性引入多尺度仿真模型。最后，基于该模型，采用蒙特卡洛方法计算疲劳弹性寿命分散带，研制样机进行地面性能测试，验证理论和方法的有效性。

一次性机械主要应用于一些军事及航空领域，是近些年提出的一个新的概念。一次性机械和常规机械不同，其主要指一类不重复使用，以自毁为主要失效形式的机械。疲劳弹性寿命是一次性机械的重要性能指标。为了实现一次性机械的高可靠度设计，需要更准确的的疲劳弹性寿命预测结果。.本项目以 GCr15 轴承钢为研究对象，以一次性电动舵机行星滚柱丝杠为应用载体，基于SEM-EBSD图像的拓扑信息，提取真实微观组织结构晶粒形状和晶向特征，再将位错密度演化规律和塑性应变率方程相结合，以位错密度为模型控制参数，建立基于位错密度的概率晶体塑性有限元模型，用以仿真预测疲劳弹性寿命分散带。通过初步仿真发现，当位错偶极子间距分布符合高斯分布时，可以实现循环载荷下的位错偶极子累积，进而影响材料塑性应变率，揭示了材料出现在疲劳弹性失效下的局部塑性累积应变的微观机理。对比试验和仿真结果，利用所建立的基于位错密度的概率晶体塑性有限元模型预测的疲劳弹性寿命分散带与试验结果基本一致，验证了疲劳弹性寿命分散带预测模型的可靠性。.本项目研究成果为一次性机械的高可靠度设计提供了理论基础，极具推广和应用价值。同时，研究过程中原创的SEM-EBSD晶粒图像处理和模型重构方法在微结构仿真研究上具有广泛的应用前景。.