基于阶比分析和支持向量机的异步电机故障诊断

Fault Diagnosis Based on Order Tracking and Support Vector Machine for Asynchronous Motor

针对电机故障诊断时转速的变化影响振动信号处理的实际情况,提出了一种阶比分析和支持向量机相结合的故障诊断方法。对故障振动信号进行角域重采样后再进行频谱分析提取故障特征参量,并将故障特征参量作为支持向量机的输入特征向量,以判断电机的故障类型。试验结果表明:该方法能有效提取异步电机故障的特征,在小样本情况下也具有较高的精度和较强的泛化能力,尤其是在变转速工况也能保证足够的诊断速度与精度。

According to the situation of vibration signals processes from a asynchronous motor were affected by the variable speed, a novel fault diagnosis method combined order racking analysis with support vector machine was proposed. Firstly, the vibration signal was sampled at constant time increments, then computer order tracking was used to resample the data in angle-domain. The resampled data was analyzed by frequency spectrum to extracting the fault characteristics. At last, the support vector machines used as fault classifiers were established to identify the fault pattern of the asynchronous motor. Practical examples showed that the proposed approach has high accuracy and good generalization even in the case of small sample and variable speed.