基于证据理论和贝叶斯网络的液压驱动系统可靠性分析

发布时间：2018-01-08 22:04

本文关键词：基于证据理论和贝叶斯网络的液压驱动系统可靠性分析　出处：《液压与气动》2017年04期 　论文类型：期刊论文

【摘要】：为解决因结构复杂、数据缺乏、人的认知水平不足等导致液压系统存在不确定性,以及液压系统存在多性能、多故障状态等多态性问题,提出了液压系统证据理论和贝叶斯网络相结合的可靠性分析方法。证据理论能够很好地处理不确定信息,利用证据理论的似然概率和信任概率描述根节点的失效可能性区间,解决根节点的故障概率存在不确定性及不易精确获取的问题;利用贝叶斯网络描述系统多态性,运用其推理算法给出了叶节点故障概率区间、根节点重要度区间以及根节点的灵敏度区间的计算方法。将该方法运用到工程机械液压驱动系统中,通过分析表明该方法能够有效地描述不确定性及多态性问题。
[Abstract]:In order to solve the problems such as complex structure, lack of data, lack of human cognition and so on, there are uncertainties in hydraulic system, and there are many problems in hydraulic system such as multi-performance, multi-fault state and so on. A reliability analysis method combining evidence theory of hydraulic system with Bayesian network is put forward. Evidence theory can deal with uncertain information well. The probability interval of the root node failure is described by using the likelihood probability and trust probability of the evidence theory to solve the problem that the fault probability of the root node is uncertain and difficult to obtain accurately. The Bayesian network is used to describe the polymorphism of the system and its reasoning algorithm is used to give the fault probability interval of the leaf node. The method of calculating the importance interval of root node and the sensitivity interval of root node is applied to the hydraulic drive system of construction machinery. The analysis shows that this method can effectively describe the uncertainty and polymorphism problems.
【作者单位】：燕山大学河北省重型机械流体动力传输与控制重点实验室;先进锻压成形技术与科学教育部重点实验室(燕山大学);燕山大学河北省工业计算机控制工程重点实验室;
【基金】：国家自然科学基金(51405426,51675460) 河北省自然科学基金(E2016203306)
【分类号】：TH137.7
【正文快照】： Reliability Analysis Based on Evidence Theory and Bayesian Network forHydraulic Drive SystemCHEN Dong-ning1,2,LI Huai-shui1,2,YAO Cheng-yu3,LI Shuo1,2,RAO Le-qing1,2(1.Hebei Provincial Key Laboratory of Heavy Machinery Fluid Power Transmission and Contro

【相似文献】