基于遗传算法优化的人工神经网络下高速滚动轴承的疲劳可靠性

考虑到高速滚动轴承中热弹流润滑效应的影响，提出一种人工智能方法进行航空轴承疲劳可靠性分析。通过带交叉项的二次多项式近似拟合温度场效应，并将热应力映射到滚动轴承赫兹接触区内，完成热弹流润滑效应下航空轴承接触应力分析模型的建立，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，运用人工神经网络法完成热弹流润滑效应下航空滚动轴承疲劳可靠性分析。采用遗传算法完成最小可靠性指标寻优和惩罚函数最佳设计点。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，所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。与传统的Monte Carlo法相比，两种计算结果的失效概率之差为2.0×10-4，相对误差为23.8%，而所提方法耗时只有蒙特卡洛方法的0.15%，具有良好的全局搜索能力和高效的计算性能。 

Fatigue reliability of high speed bearing based on genetic algorithm optimized artificial neural network

To complete the fatigue reliability analysis of aviation bearing under thermal elastohydrodynamic lubrication (EHL) efficiently and accurately, an artificial intelligent method was proposed. The heat stress from temperature was approximated using quadratic polynomial with intercrossing term and then mapped into the Hertz contact zone. The contact stress model, which included the thermal EHL, was established. Considering the randomness of the thermal EHL, the material properties and fatigue strength correction factors, the probabilistic reliability analysis model was established using artificial neural network (ANN). Genetic algorithm(GA) was employed to search the minimum reliability index and the design point by introducing an adjusting factor in penalty function. Reliability sensitivity analysis was completed based on the advanced first order second moment (AFOSM). Numerical example shows that the established probabilistic reliability analysis model could correctly reflect the effect of thermal EHL on contact fatigue of aviation bearing. Compared with the traditional Monte Carlo method, the proposed method presents a difference in failure probability of 2.0×10-4, a relative error of 23.8% and a relative time consumption of only 0.15%, which means that the proposed method has an excellent global search capability and a highly efficiency.