滚动轴承的运行状态直接影响机械设备的正常运行。为及时准确识别滚动轴承的运行状态，通过对滚动轴承运行过程中的振动信号分析，采用小波包变换提取各频带内的能量熵，以此作为反映轴承运行状态的特征向量，并利用支持向量机对提取的特征向量进行模式识别。研究结果表明，支持向量机的轴承故障识别准确率均达到99%以上，而采用多项式核函数的支持向量机识别准确率最高，可达99.6%。%The running status of rolling bearings directly affect the normal operation of machinery and equipment, in order to precisely recognize the running status of rolling bearings, through analyzing the vibration signals during the running process of rolling bearings, by adopting wavelet packet transform method to extract the energy entropy in each frequency band, which reflects the feature vector of the running status, and by using support vector machine, the pattern recognition for feature vector extracted is conducted. The result of research indicates that the accurate rate of this method is above 99%, while the highest recognition accuracy by using polynomial kernel function support vector machine is 99. 6%.
展开▼
