首页|基于BN-AHP的装甲车辆动力系统故障状态评估

基于BN-AHP的装甲车辆动力系统故障状态评估

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
装甲车辆复杂传统部件的损伤能否及时被发现,这关系到整车战备或作战能力,系统的故障状态评估是至关重要的.将贝叶斯网络模型结合云模型理论,建立云贝叶斯网络模型,针对 4 个不同工况的装甲车辆进行故障状态评估.在获取贝叶斯网络初始节点时更多是依靠专家经验,往往会带来很大的误差,导致条件概率偏差过大,采用证据理论/层次分析法来优化专家经验,确定各个节点的条件概率;将层次分析法转化所得的条件概率值代入到云贝叶斯网络模型中,经过计算可以得到不同损毁等级的概率.将云贝叶斯网络模型计算结果与其他状态评估方法结果进行对比分析,结果表明,所采用的计算方法较其他方法在可靠性和准确性方面有所提高.
Fault state evaluation of armored vehicle power system based on BN-AHP
Whether the damage of complex traditional components of armored vehicles can be detected in time is related to the combat readiness or combat capability of the whole vehicle,and the fault state assessment of the system is crucial.In this paper,based on the Bayesian network,the cloud model theory is fused to establish the cloud Bayesian network model,which is used to evaluate the fault status of armored vehicles under four different working conditions.When obtaining the initial nodes of Bayesian network,we rely more on the experience of experts,which often leads to large error,resulting in too large deviation of conditional probability.In this paper,we use evidence theory/analytic hierarchy process to optimize the expert experience and determine the conditional probability of each node.The conditional probability value converted by the analytic hierarchy process is substituted into the Yun-Baier network model,and the probability of different damage levels can be obtained through calculation.Comparing the results of the cloud Bayesian network model with the results of other state assessment methods,the results show that the calculation method used in this paper has improved reliability and accuracy compared with other methods.

Bayesian networkanalytic hierarchy processcloud model transformationfault state assessmentevidence theoryexpert experience

王文顺、崔俊杰、刘勇、张江、夏添、武一博

展开 >

中北大学 机电工程学院,太原 030051

中北大学 能源与动力工程学院,太原 030051

贝叶斯网络 层次分析法 云模型转换 故障状态评估 证据理论 专家经验

机电系统测控北京市重点实验室开放基金国防科技重点实验室项目山西省研究生创新项目

KF2022222320361422132004072021Y577

2024

10.11809/bqzbgcxb2024.03.011

兵器装备工程学报
重庆市（四川省）兵工学会 重庆理工大学

兵器装备工程学报

CSTPCD北大核心
影响因子：0.478
ISSN：2096-2304
年,卷(期)：2024.45(3)
  • 16