摘要
为提高数控机床可靠性评估的准确性,基于边界强度过程理论,提出一种初始时刻故障强度不为零的三参数边界强度过程模型。通过对故障强度函数进行求导运算,证实了模型的特性,并推导了三参数边界强度过程和齐次泊松过程之间的关系。利用极大似然估计法和优化方法,给出了模型参数及可靠性指标的计算公式。分别以单台数控机床故障截尾数据和多台数控机床时间截尾数据为例对模型进行验证,对数似然函数及拟合优度计算结果均表明:三参数边界强度过程模型优于边界强度过程模型,而且更符合工程实际,可为制定合理的维修策略提供一定的理论依据。
To improve the reliability assessment accuracy of NC machine tools, a three-parameter bounded intensity process（3-BIP）model by intensity function with non-zero initial condition is proposed. The characteristic of the model is confirmed by differentiating with respect to the failure intensity function, and the relationship between the three-parameter bounded intensity process and the homogenous Poisson process is derived. The calculating formula for model parameters and reliability indexes are deduced by the maximum likelihood estimation and optimization. Two examples of the failure truncated data from single NC machine tool and the time truncated data from multiple NC machine tools are taken respectively to verify the proposed method. The results of log-likelihood function and goodness-of-fit show that the 3-BIP model is better than the BIP model, so the 3-BIP model accords with practical engineering well and provides a theoretical basis for making a reasonable maintenance strategy.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2014年第5期107-112,共6页
Journal of Xi'an Jiaotong University
基金
国家重大数控专项资助项目(2010ZX04001-032)
国家自然科学基金资助项目(51165018)
关键词
三参数边界强度过程
极大似然估计
数控机床
可靠性评估
three-parameter bounded intensity process
maximum likelihood estimation
NC machine tool
reliability assessment
