基于PNN和第二代小波熵的智能故障诊断研究

本文结合转管自动机运动循环过程和振动信号的非线性短时冲击特性,提出了截取振动信号中有用信息进行第二代小波分析,在此基础上提取第二代小波熵,并把其作为特征向量应用概率神经网络对转管自动机进行故障模式识别,实现了对实际射击信号的转管自动机故障诊断,从而验证了该方法的有效性。

Intelligent Diagnosis Based on PNN and The second Generation Wavelet Entropy

Combining gatling automaton circle motion process analysis with the nonlinear, short-term impact characteristics of the vibration signal, this paper proposed the way of using the second generation of wavelet to analysis the intercepted useful information of the vibration signal. Based on this, the second generation of wavelet entropy was extracted and then was regarded as a feature vector applied to Probabilistic Neural Networks for fault diagnosis of the actual firing signal of gatling automaton so as to verify the effectiveness of this new method.