摘要
基于叠加的对数线性过程(S-LLP),提出了多台数控机床在最小维修条件下的全寿命周期可靠性评估方法,建立了包含早期故障期和耗损故障期的浴盆曲线模型,定量分析了早期故障期的持续时间,给出了模型参数和可靠性指标的极大似然点估计和区间估计;同时,以多台数控机床现场故障截尾数据为例对模型进行验证.AIC值和拟合优度计算结果表明,S-LLP模型优于叠加幂律过程模型,适用于数控机床的全寿命周期可靠性评估.
Based on superposed log-linear process （S-LLP）, a whole lifecycle reliability assessment method for multiple numerically-controlled （NC） machine tools with minimal repair was proposed. A bathtub curve model, which includes early failure period and wear failure period, was presented. The lasting time of early failure period was quantitatively analyzed, and the point maximum likelihood and interval estimators of model parameters and reliability indices were given. An example of real data with failure truncation from multi-NC machine tools was taken to prove the proposed method. The results of AIC and goodness- of-fit show that the S-LLP model is better than that of the superposed power-law process model, which is suitable for whole lifecycle reliability assessment of NC machine tools.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2015年第1期19-23,共5页
Journal of Shanghai Jiaotong University
基金
国家科技重大专项项目资助(2010ZX04001-032)
甘肃省省青年科技基金计划资助项目(145RJYA307)
关键词
数控机床
最小维修
极大似然估计
全寿命周期
可靠性评估
numerically-controlled machine tools
minimal repair
maximum likelihood estimation
whole lifecycle
reliability assessment
