摘要
针对自动机驱动机构产生的4种不同非平稳、非线性信号的特征,从而提出了一种将改进的经验模态分解(empirical mode decomposition,EMD)与SOM神经网络相结合的故障诊断方法。首先,通过改进的EMD方法将原始信号分解为若干个不同的固有模式函数(IMF)分量,由于分解的这些分量中包含的主要有用信息各不相同,因此,采用K-L散度法剔除包含少的IMF分量,从而提取主要的IMF分量;然后,利用相应的数学工具提取出该主要IMF分量相关的特征值,并根据这些特征值组成相应的特征向量;把组成好的特征向量输入到已训练好了的SOM神经网络中进行自动机驱动机构的故障诊断。经过试验结果验证,该方法可有效识别提取自动机驱动机构故障特征,实现不同故障状态识别。
This paper presents an empirical mode decomposition(EMD)and SOM neural network based fault diagnosis method for automaton drive mechanism.First of all,the original signal is decomposed into several IMF components at different frequencies through the improved EMD method,and these components contain different fault information.Therefore,the k-L divergence method is adopted in this paper to eliminate the IMF components that contain less IMF components,so as to extract the main IMF components.Then,the time domain eigenvalues are calculated according to the main IMF components to form the eigenvectors.SOM neural network is used to identify the fault state of SOM.The experimental results show that this method can effectively identify and extract the fault characteristics of the automaton driving mechanism and realize the recognition of different fault states.
作者
范勇
王鹏
张宁超
Fan Yong;Wang Peng;Zhang Ningchao(School of Electronic and Information Engineering,Xi′an Technological University,Xi′an 710021,China)
出处
《国外电子测量技术》
2020年第8期5-10,共6页
Foreign Electronic Measurement Technology
基金
陕西省科技厅重点研发计划项目(2019GY-075)
西安市科技局科技创新引导项目(201805031YD9CG15(1))
西安市科技局高校人才服务企业项目(2019217214GXRC008CG009-GXYD8.1)资助。
