基于LMD多尺度熵和极限学习机的模拟电路故障诊断

发布时间：2018-12-20 09:30

【摘要】：为了高速、高效的测试和诊断模拟电路,提出一种将局部均值分解(LMD)多尺度熵和极限学习机相结合的模拟电路故障诊断的新方法。该方法中,首先采用LMD将故障信号分解为若干个乘积函数(production function,PF);然后,求出各PF分量的多尺度熵并构造故障特征向量;最后,将特征向量输入到极限学习机中进行训练和测试。仿真实验结果显示采用该方法诊断时间只需0.028 74 s,诊断精度达到了98.89%。相较于其他3种方法有效减少诊断时间,提高故障诊断精度。
[Abstract]:In order to test and diagnose analog circuits with high speed and efficiency, a new method for fault diagnosis of analog circuits is proposed, which combines the local mean decomposition of (LMD) multi-scale entropy with the ultimate learning machine. In this method, the fault signal is first decomposed into several product functions (production function,PF) by LMD, then the multi-scale entropy of each PF component is obtained and the fault eigenvector is constructed. Finally, the eigenvector is input into the extreme learning machine for training and testing. The simulation results show that the diagnostic time is only 0.028 74 s and the diagnostic accuracy is 98.89 s. Compared with the other three methods, the diagnosis time is reduced and the fault diagnosis accuracy is improved.
【作者单位】：湖南师范大学物理与信息科学学院;合肥工业大学电气工程博士后流动站;合肥工业大学电气与自动化工程学院;国网湖南省邵阳供电公司;
【基金】：国家自然科学基金(51577046);国家自然科学基金重点项目(51637004) 国家重点研发计划“重大科学仪器设备开发”项目(2016YFF0102200) 湖南省教育厅项目(17C0956)资助
【分类号】：TN710

【相似文献】