基于谐波小波包和IAGA-SVM的滚动轴承故障诊断

传统方法在诊断滚动轴承故障时受人为因素影响,故障成因复杂,因此在已有理论上提出一种基于谐波小波包和自适应支持向量机相结合的捣固车故障诊断方法。谐波小波包对不同故障下的振动信号展开分解及重构后所提取的频带能量即为特征向量,再把特征值输入支持向量机(SVM)模型中训练并对核函数和惩罚系数进行优化。用自适应支持向量机构建从特征向量到故障类型间的对应,从而完成滚动轴承故障的诊断。该方法能高效准确地诊断出故障类型且有实用价值。通过与GA-SVM及AGA-SVM对比,证明此方法在故障诊断领域中的卓越性。

FAULT DIAGNOSIS OF ROLLING BEARING BASED ON HARMONIC WAVELET PACKET AND IAGA-SVM

The traditional method is affected by human factors in the diagnosis of rolling bearing faults,and the causes of faults are complex.Therefore,a fault diagnosis method based on wavelet packet analysis and adaptive support vector machine was proposed in theory.The harmonic wavelet packet was used to decompose and reconstruct the vibration signal under different faults,and the extracted band energy was the feature vector.Then the eigenvalue was input into the support vector machine (SVM) model and the kernel function and penalty coefficient were optimized.Finally,the adaptive support vector machine was used to construct the correspondence from the feature vector to the fault type,thus completing the diagnosis of the rolling bearing fault.Therefore,the method can diagnose the fault type efficiently and accurately and has practical value.In addition,compare with GA-SVM and AGA-SVM,this method is proved to be excellent in the field of fault diagnosis.