摘要
提出了一种基于时间序列AR模型和支持向量机的故障诊断方法。首先利用AR模型对振动信号进行建模,然后将AR模型自回归系数组成的特征向量输入到支持向量机,最后支持向量机完成对不同工况的分类识别。测试系统采用LabVIEW虚拟仪器构建,并通过计算机自动完成测试和分析。实验结果表明:这种基于时间序列AR模型和支持向量机的故障诊断方法是可行的。
In this paper,a new fault diagnosis method based on time series autoregressive（AR） model and support vector machine has been put forward.First,the AR model of the relief valve body vibration signals is established.The autoregressive coefficients are regarded as the feature vectors,and are used as an input of the support vector machine.Then,the normal state and artificial faults are classified by support vector machine.The experiments and diagnosis process based on virtual instruments LabVIEW are automatically controlled by computer.The results and analysis of the experiments indicate that this new fault diagnosis method based on time series autoregressive model and support vector machine is feasible and effective.
出处
《机械科学与技术》
CSCD
北大核心
2010年第7期972-975,980,共5页
Mechanical Science and Technology for Aerospace Engineering
基金
福建省重大专项专题项目(2008HZ0201-3)
国家自然科学基金项目(50975098)资助
关键词
溢流阀
故障诊断
AR模型
支持向量机
relief valve
fault diagnosis
autoregressive（AR） model
support vector machine
