基于两阶段Gamma过程的机械产品剩余寿命预测研究  

Remaining useful lifetime prediction of the mechanical products based on the two-phase Gamma process

作　　者：姜洋[1] JIANG Yang(School of Modern Information Technology,Zhejiang Institute of Mechanical&Electrical Engineering,Hangzhou 310053,China)

机构地区：[1]浙江机电职业技术学院现代信息技术学院,浙江杭州310053

出　　处：《机电工程》2021年第6期802-806,共5页Journal of Mechanical & Electrical Engineering

基　　金：浙江省科技厅重点研发计划资助项目(2021C01149);高校国内访问工程师校企合作资助项目(FG2019072)。

摘　　要：在进行机械产品寿命预测时,传统的两状态可靠性评估存在较大的估计误差,针对这一问题,在计及冲击载荷条件下,提出了一种基于两阶段Gamma过程模型的三态机械产品的寿命预测方法。首先,在对产品连续退化轨迹研究的基础上,采用两阶段Gamma过程描述了产品的连续退化过程;同时,基于复合泊松过程描述了外部随机冲击造成的离散退化;然后,基于累积损伤模型,建立了冲击失效与退化失效竞争条件下机械产品的可靠性估计模型,利用该模型进行了机械产品剩余寿命预测;最后,通过对金属材料的裂纹扩展性能退化过程的仿真分析,验证了所提方法的有效性。研究结果表明:相对于传统两状态的可靠性评估模型,基于两阶段Gamma过程模型的三态寿命预测方法在机械产品的可靠性评估方面具有更高的精度。In order to solve the problem of large estimation error in traditional two-state reliability assessment,a life prediction method of three-state mechanical products was proposed based on the two-phase Gamma process model under the condition of impact load.Firstly,based on the study of the continuous degradation trajectory of the product,a two-phase Gamma process model was used to describe the continuous degradation of the product,and the discrete degradation caused by external random impact was described based on compound Poisson process.Then,based on the cumulative damage model,the reliability estimation model of mechanical products under the competitive condition of impact failure and degradation failure was established,and the residual life of mechanical products was predicted by using the model.Finally,the correctness and effectiveness of the proposed method were verified by the simulation analysis of the degradation process of the crack growth performance of metal materials.The research results indicate that comparing with the traditional two-state reliability evaluation model,the proposed three-state life prediction method has higher accuracy for the reliability evaluation results.

关 键 词：两阶段伽马过程模型 三态机械产品 复合泊松过程 寿命预测 可靠性评估 

分 类 号：TH17[机械工程—机械制造及自动化] TB114.3[理学—概率论与数理统计]