双相干谱和RBF网络在旋转机械故障诊断中的应用

BICOHERENCE SPECTRUM AND RADIAL BASIS FUNCTION NETWORK APPLIED TO FAULT DIAGNOSIS OF ROTATING MACHINERY

双相干谱保留了信号的相位信息 ,可以用来描述非线性相位的耦合 ,径向基函数网络具有良好的推广能力和分类能力。文中将双相干谱和径向基函数网络结合 ,提出一种基于双相干谱与径向基函数网络相结合的旋转机械故障诊断方法 ,即以双相干谱为故障特征向量 ,以径向基函数网络作为分类器 ,对旋转机械的故障进行分类 ,并以转子不平衡、转轴碰摩、油膜涡动为例进行实验研究。实验结果表明 。

Bicoherence spectrum retained the system phase information, could describe nonlinear phase coupling. Radial basis function (RBF) network had good extensible and classified ability. Combined bicoherence spectrum with RBF network, a fault diagnosis approach using bicoherence spectrum and RBFN, in rotating machinery, was presented. They were that bicoherence spectrum was used as fault feature, and an RBFN as the classifier, fault classification in rotating machinery were successfully completed. At the same time, as an example of three typical fault i.e. rotor imbalance, rotor-to-stator contact and oil whirl. The experiment research had been made. The experimental results show that this fault diagnosis method is very effective in rotating machinery.