摘要
对齿轮故障诊断的特点进行了阐述 ,指出由于环境噪声的干扰 ,在齿轮故障诊断中往往不能获得理想的诊断结果。为此在对齿轮运行状况进行有效特征提取的基础上 ,采用支持向量机的方法对齿轮进行故障诊断。
The paper analyzed the characteristic of gear fault diagnosis. Because the influence of environment noise, the result of reliable diagnosis was always hard to gain. The paper presented the gear fault diagnosis by support vector machine (SVM) on the basis of effective feature extraction. The diagnosis result was compared with other approaches including Fisher discriminant analysis and neural networks. The comparison showed that the SVM could be used successfully in the field of fault diagnosis.
基金
广东省自然科学基金资助项目(0 2 1 34 9)
关键词
齿轮
支持向量机
故障诊断
神经网络
回转振动
Support Vector Machine
Fault Diagnosis
Neural Networks
Rotational Vibration
