摘要
建立了一种基于贝叶斯网络的机械设备故障诊断模型,引入评分函数和蚁群算法对模型进行了优化,在模型建立过程中引入知识进行自我学习,减少了因检测对象造成的不确定信息,提高了机械设备故障检测的可信度,最后通过实例进行了验证。
It establishes a model of mechanical equipment fault diagnosis based on Bayesian network. Based on the score function and ant colony algorithm it optimizes the model. Appling adopting self-learning in the process of building model,this model reduces the uncertainty caused by the detection object,and improves the reliability of mechanical equipment fault detection. An example shows that the model is effective.
作者
郭日红
董忠文
谢国锋
GUO Rihong DONG Zhongwen XIE Guofeng(66440 Unit of PLA, Hebei Shijiazhuang, 050001, China)
出处
《机械设计与制造工程》
2016年第10期87-91,共5页
Machine Design and Manufacturing Engineering
关键词
机械故障
贝叶斯网络
蚁群算法
故障诊断
mechanism fault
Bayesian networks
ant colony algorithm
fault diagnosis