基于ITD样本熵与PNN的自动机故障诊断

Automaton Fault Diagnosis Based on the ITD The Sample Entropy Analysis and PNN

针对自动机振动信号短时、非平稳、高冲击的特性,本文提出一种运用固有时间尺度分解(ITD)样本熵和概率神经网络(PNN)进行故障诊断的方法。首先将ITD引入自动机的故障诊断中,通过对ITD分解得到前五层重构信号提取的时频特征来验证ITD方法的有效性,并对信号进行样本熵提取,把其作为特征向量分别用概率神经网络和BP神经网络对自动机进行故障模式识别。实验结果表明:概率神经网络相对于BP神经网络可以提高故障分类的正确率,从而验证了ITD样本熵与PNN的自动机故障诊断方法的优越性。

Aiming at the short lime, non-stationary and high impact characteristic of the automatic vibratitm signals, this paper proposes a method based on the intrinsic time scale decompositiCm （ITD） sample entropy and probabilistic neural network （PNN）. Firstly, ITD is introduced into the fault diagnosis of the automatic machine,The validity of the ITD method is verified by the time-frequency features of the first five layer reconstructed signals obtained by ITD decomposition, the sample entropy of signals are extracted and then are regarded as t^eature ve~.~lOl＇S rc^spectiv~：ly applied to probabilistic neural and BP neural network networks for fault diagnosis of the automaton. The results show that PNN relative to BP neural network can improve the fault diagnosis accuracy. So it proves that the superiority of the method of automaton fault diagnosis method of ITD sample entropy and PNN.