论文部分内容阅读
针对自动机运作时的瞬态冲击、非线性、非平稳信号特征,提出一种基于排列熵和支持向量机对小口径高速自动机进行故障诊断的方法。首先,引入排列熵对信号进行分析,发现排列熵能很好地反映自动机工作状态;其次,将排列熵特征量分别作为概率神经网络PNN和SVM的输入参数以识别自动机故障,结果表明:SVM相比于PNN可以提高分类正确率。同时证明基于排列熵和SVM在自动机故障诊断中的有效性。
Aiming at the characteristics of transient impact, nonlinear and non-stationary signals when automaton operates, a method of fault diagnosis of small aperture high-speed automata based on permutation entropy and support vector machine is proposed. Firstly, the permutation entropy is introduced to analyze the signal, and it is found that permutation entropy can reflect the working status of the automaton well. Secondly, the permutation entropy is regarded as the input parameter of PNN and SVM respectively to identify the automata fault. The results show that: SVM compared to PNN can improve the classification accuracy. At the same time, we prove the validity of the proposed method in automata fault diagnosis based on permutation entropy and SVM.