摘要
车身结构的复杂性及知识表达的不精确性,使得车身故障症状与故障原因之间的映射表现为随机和不确定。针对这些特点,在大量车身测量数据和历史诊断案例的基础上,将贝叶斯网络引入到车身偏差故障诊断中去。对贝叶斯网络的参数学习进行了探讨,结合实例统计和相关性分析建立了车身偏差诊断的贝叶斯网络模型。最后用以某车型的偏差诊断案例对该方法进行了验证,结果表明该方法在工程实际中有一定的指导性。
The mapping of body＇s fault symptom and source register as randomization and uncertainty because of the complexity and uncertainty of body deviation. Based on massive measurement data of body and historical cases, Bayesian network is combined with deviation diagnosis of body. Parameter study of Bayesian network is investigated. According to the methods of example statistics and correlation analysis, Bayesian diagnosis model of body deviation is established. The fault diagnosis case of one model has proved the feasibility of this method finally and it can guide our diagnosis in practice.
出处
《机械》
2009年第3期67-70,共4页
Machinery
关键词
车身偏差
贝叶斯网络
故障诊断
body deviation
Bayesian network
fault diagnosis
