基于奇异值分解和支持向量机的齿轮故障诊断

Gear Fault Diagnosis based on Singular Value Decomposition and Support Vector Machine

针对大型设备的齿轮运行环境噪声干扰大和缺少故障样本的特点,提出了基于奇异值分解和支持向量机相结合的齿轮故障诊断方法。分析了奇异值分解法在信号特征提取中的应用与优势、支持向量机的原理与算法,并通过试验验证了基于奇异值分解和支持向量机的齿轮故障诊断方法可以实现对齿轮进行快速、准确的故障诊断。

For large equipment gear running environmental noise and lack of fault sample characteristics, the method of gears fault diagnosis based on singular value decomposition and support vector machine is proposed. Applica- tions and advantages of the singular value decomposition of the signal feature extraction is analyzed. Verified by experi- ments based on singular value decomposition and support vector machine gear fault diagnosis method can achieve quickly, accurately gears fault diagnosis.