摘要
采用多状态贝叶斯网络方法对复杂系统进行可靠性分析时,由于复杂系统中的设备对系统状态的影响程度难以确定,系统中存在较大的认知不确定性。多态贝叶斯网络中的认知不确定性问题可结合专家经验通过构造条件概率表来解决,但随着系统规模的增加,专家打分工作量呈指数增长且具有一定的模糊性和主观性。针对以上问题,基于层次分析法与三角模糊数的模糊群决策方法,并将其与多态多父节点条件概率计算模型相结合构造条件概率表,提出计及认知不确定性的多态贝叶斯网络方法。将其应用于某船舶态势感知系统的多状态可靠性分析中,诊断态势感知系统处于不同故障状态下的薄弱环节,为船舶态势感知系统不同故障状态下的运行和维护提供参考。
When using the multi-state Bayesian network method for reliability analysis of complex systems,there is significant cognitive uncertainty in the system due to the difficulty in determining the degree of influence of equipment on the system's state.The cognitive uncertainty problem in multi-state Bayesian networks can be solved by constructing conditional probability tables based on experts'experience.However,as the system size increases,the workload of expert scoring increases exponentially and has a certain degree of ambiguity and subjectivity.In response to the above issues,a fuzzy group decision-making method based on Analytic Hierarchy Process and triangular fuzzy numbers is combined with a multi-state and multi-parent node conditional probability calculation model to construct a conditional probability table,and a multi-state Bayesian network method considering cognitive uncertainty is proposed.This method is applied to the multi-state reliability analysis of a ship situational awareness system to diagnose the weak links of the system under different fault statesproviding reference for the operation and maintenance of the ship's situational awareness system under different fault states.
作者
罗小芳
李雨珊
白旭
杨立
张海华
李永正
LUO Xiaofang;LI Yushan;BAI Xu;YANG Li;ZHANG Haihua;LI Yongzheng(Jiangsu University of Science and Technology,School of Economics and Management,Zhenjiang 212003,Jiangsu,China;Jiangsu University of Science and Technology,Ship and Ocean Engineering Intelligent Technology Research Center,Zhenjiang 212003,Jiangsu,China;Shanghai Branch,China Ship Scientific Research Center,Shanghai 200011,China)
出处
《船舶工程》
CSCD
北大核心
2023年第8期101-106,169,共7页
Ship Engineering
关键词
认知不确定
条件概率
多状态可靠性分析
多态贝叶斯网络
态势感知系统
epistemic uncertainty
conditional probability
multi-state reliability analysis
multi-state Bayesian network
situational awareness system
