免疫支持向量机复合故障诊断方法及试验研究

研究了传统分类算法在故障诊断中的不足，融合人工免疫系统中的实值否定选择(RNS)算法和支持向量机(SVM)算法提出了一种复合的故障诊断方法。在新方法中使用RNS算法产生检测器(非己集合)当作故障样本，这些样本再作为SVM算法的输入进行训练，这样就能解决分类算法所面临的训练样本不足的难题。轴向柱塞泵发生故障时，由于滑靴对斜盘冲击产生的振动信号被高频谐振信号调制，通过小波簇包络解调方法将调制信号解调出来，然后对包络信号用小波包分解子带特征能量法进行特征提取。最后用轴向柱塞泵多松靴和配流盘磨损多故障模式样本进行诊断测试，正确率可达90%以上，验证了复合诊断方法的有效性。

Composite fault diagnosis method and its verification experiments

A composite fault diagnosis approach was proposed,combining the real-valued negative selection(RNS) algorithm with the support vector machine(SVM),and considering the shortcoming of the conventional classification algorithm applied in fault diagnosis.In the new method,the RNS algorithm was used to generate the detector(non-self) as unknown fault samples,which were then used as input to SVM algorithm for training purpose.The difficult problem of lacking training samples was solved by using the new method in ...