摘要
组建完成了一个用于机械故障诊断的具有两层节点结构的Bayes网络模型,利用Bayes网络的强大学习功能,探讨了当学习样本集完整时,采用MLE(最大似然估计)法则来合理地调整网络模型的CPT表值,使之更加符合特定机组的运行情况,从而为网络模型的判断推理做好了准备。
On the base of Bayes network, the model of fault diagnosis for machine contains two-layer nodes is constructed in the paper. With the help of the strong learning capability of the model, when the learning data samples are complete, MLE algorithm is used to adjust the value of the conditional probability table of the model to the special machine. Consequently the model is ready to reason and calculate.
出处
《河南科学》
2007年第1期98-100,共3页
Henan Science
关键词
故障诊断
BAYES网络
网络学习
faults diagnosis
Bayes network
learning of the network
