Please wait a minute...
工程设计学报
Chinese Journal of Engineering Design  2014, Vol. 21 Issue (5): 405-411    DOI: 10.3785/j.issn. 1006-754X.2014.05.001
    
Failure probability analysis of multi-state system based on Bayesian network with fuzzy support radius variable
ZHANG Lu-lu1， ZHANG Rui-jun1，2， WANG Xiao-wei1，2， WANG Nan-nan1
1. School of Mechanical and Electronic Engineering， Shandong Jianzhu University， Jinan 250101, China；
2. Key Laboratory of Mechanical Engineering & Innovation Technology in Universities of Shandong（Shandong Jianzhu University）， Jinan 250101, China
Download: HTML     PDF(1924KB)
Export: BibTeX | EndNote (RIS)      

Abstract  In order to overcome the shortage of describing the fault states of system and components with existing membership functions, a membership function with fuzzy support radius variable is constructed to describe the fault states and a failure probability calculation method of multistate system based on Bayesian network with variable fuzzy support radius is proposed. The variable was introduced to describe fuzzy support radius of membership function and used to construct the membership function. Introducing the established membership function with variable into Bayesian network, the leaf node failure probability curve was obtained by analyzing the failure probability of leaf nodes on the multistate system using bucket elimination method. The proposed method was compared with Bayesian network method based on membership function with constant fuzzy support radius to verify the validity and generality of the method. Finally, the method was applied to analyze the CNC machine tool spindle system, and calculated the leaf node failure probability. The results show that the method can solve the subjectivity of fault states membership function selection in multistate system effectively.

Key words： fuzzy support radius variable      membership function      Bayesian network      multi-state system      failure probability      spindle system     
Received: 05 June 2014      Published: 28 October 2014
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

ZHANG Lu-lu， ZHANG Rui-jun， WANG Xiao-wei， WANG Nan-nan. Failure probability analysis of multi-state system based on Bayesian network with fuzzy support radius variable. Chinese Journal of Engineering Design, 2014, 21(5): 405-411.

URL:

https://www.zjujournals.com/gcsjxb/10.3785/j.issn. 1006-754X.2014.05.001     OR     https://www.zjujournals.com/gcsjxb/Y2014/V21/I5/405


模糊支撑半径变量贝叶斯网络多态系统故障概率分析

针对现有隶属函数描述系统及部件故障状态的不足，构建了一种含模糊支撑半径变量的隶属函数来描述部件故障状态，提出了一种基于模糊支撑半径变量的贝叶斯网络多态系统故障概率计算方法.在隶属函数的构造中，用变量代替精确值描述隶属函数模糊支撑半径，建立了含模糊支撑半径变量的隶属函数；并将其引入贝叶斯网络，利用桶消元法对多态系统叶节点故障概率进行分析，得到叶节点故障概率变化曲线.将所提出的方法与文献中基于模糊支撑半径为定值的隶属函数贝叶斯网络方法相比较，验证了方法的有效性.最后，将该方法应用到数控机床主轴系统故障分析实例中，对叶节点故障概率进行计算.结果表明，该方法能够有效解决多态系统故障状态隶属函数选择中的主观性问题.

关键词： 模糊支撑半径变量,  隶属函数,  贝叶斯网络,  多态系统,  故障概率,  主轴系统 
[1] DENG Xiao-lei, PANG Shi-jie, LI Rui-qi, ZHOU Yi-bo, WANG Jian-chen, FU Jian-zhong. Thermal design of cooling structure for CNC machine tool spindle system based on insect wing vein bionic channel[J]. Chinese Journal of Engineering Design, 2018, 25(5): 583-589.
[2] HUANG Qiang, DING Zhi-Hua, WU Chong-Jun, LIU Xin. Fuzzy control with heterogeneous distribution membership function for
fluctuation in fuel pressure[J]. Chinese Journal of Engineering Design, 2007, 14(6): 478-481.
[3] XIE Li-Yang, YANG Xiao-Fang, ZHOU Jin-Yu, WANG Xue-Min. Framework of segment partition method for reliability design of long pipeline and rope[J]. Chinese Journal of Engineering Design, 2004, 11(6): 312-315.