摘要
为实现小样本数据下链式刀库系统可靠性薄弱环节评估,提出一种基于贝叶斯网络的链式刀库系统重要度分析方法。采用结构化分析和设计技术确定部件间的功能依赖关系,应用故障模式影响分析方法(FMEA)确定部件间故障影响关系;结合现场故障信息计算贝叶斯网络节点条件概率并构建其贝叶斯网络结构模型,应用证据理论融合多元信息计算根节点故障率,经贝叶斯推理实现节点重要度分析以确定关键节点,为提高系统可靠性提供参考。最后,以国产某系列卧式加工中心故障信息为例验证了方法的有效性。
In order to evaluate the reliability of chain-type tool magazine system with small sample data,a chain-type tool magazine system importance analysis method based on Bayesian network is proposed.Structural analysis and design technology are used to determine the functional dependence among components,and Fault Mode Effects Analysis(FMEA)is applied to determine the fault impact relationship among components.The conditional probability of Bayesian network mode with the field fault information is calculated and the Bayesian network structure model is constructed.Evidence theory is used to fuse multi-element information to calculate the failure rate of root nodes,and Bayesian reasoning is used to realize node importance analysis to determine the key nodes,which can provide reference for improving the reliability of the system.A series of domestic horizontal machining centre fault information are taken as examples to verify the validity of the proposed method.
作者
郑玉彬
申桂香
张英芝
牟黎明
秦猛猛
ZHENG Yu-bin;SHEN Gui-xiang;ZHANG Ying-zhi;MU Li-ming;QIN Meng-meng(School of Mechanical and Aerospace Engineering,Jilin University,Changchun 130022,China)
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2019年第2期466-471,共6页
Journal of Jilin University:Engineering and Technology Edition
基金
吉林省科技发展计划项目(20190302104GX)
吉林省自然科学基金项目(20170101212JC)
关键词
机床
链式刀库系统
贝叶斯网络
条件概率
证据理论
重要度分析
machine tool
chain-type tool magazine system
Bayesian network
conditional probability
evidence theory
importance analysis
