何威,何怡刚,罗旗舞,李志刚,邓芳明,张朝龙.基于连续小波奇异熵的模拟电路故障诊断新方法[J].电子测量与仪器学报,2017,31(12):1967-1973
基于连续小波奇异熵的模拟电路故障诊断新方法
Novel approach for analog circuit fault diagnosis based on continuous wavelet singularity entropy
  
DOI:10.13382/j.jemi.2017.12.013
中文关键词:  模拟电路  故障诊断  连续小波变换  Tsallis奇异熵  超限学习机
英文关键词:analog circuit  fault diagnosis  continuous wavelet transformation  Tsallis singularity entropy  extreme learning machine
基金项目:国家自然科学基金(51577046)、国家自然科学基金重点项目(51637004)、重点研发计划“重大科学仪器设备开发”项目(2016YFF0102200)资助
作者单位
何威 合肥工业大学电气与自动化工程学院合肥230009 
何怡刚 合肥工业大学电气与自动化工程学院合肥230009 
罗旗舞 合肥工业大学电气与自动化工程学院合肥230009 
李志刚 合肥工业大学电气与自动化工程学院合肥230009 
邓芳明 合肥工业大学电气与自动化工程学院合肥230009 
张朝龙 1. 合肥工业大学电气与自动化工程学院合肥230009;2. 安庆师范大学物理与电气工程学院安庆246011 
AuthorInstitution
He Wei School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China 
He Yigang School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China 
Luo Qiwu School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China 
Li Zhigang School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China 
Deng Fangming School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China 
Zhang Chaolong 1. School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China; 2. School of Physics and Electrical Engineering, Anqing Normal University, Anqing 246011, China 
摘要点击次数: 2538
全文下载次数: 7049
中文摘要:
      针对模拟电路的故障诊断和定位问题,为进一步提高故障诊断准确率,提出了一种基于连续小波Tsallis奇异熵和超限学习机的故障诊断方法。首先应用连续小波变换计算被测电路时域响应信号的时频系数矩阵,然后将其分割为8个相同大小的子矩阵,分别计算每个子矩阵的Tsallis奇异熵,组成特征向量,最后将特征应用于超限学习机多类分类器进行区分。仿真结果表明,故障诊断方法能较好地获取故障响应信号的本质特征,并具有较其他现存方法更高的故障诊断正确率。
英文摘要:
      Aiming at the issue of analog circuit fault diagnosis and location, a novel approach for analog circuit fault diagnosis based on continuous wavelet Tsallis singularity entropy (TSE) and extreme learning machine (ELM) is proposed to enhance the accuracy of fault diagnosis. Firstly, the fault response signals are preprocessed by the continuous wavelet transformation to obtain the time frequency coefficient matrix, and the matrix is divided into 8 congruent time frequency blocks. Then, the feature vector is obtained by computing TSE of each block. Finally, the feature vectors are used as the inputs of a kind of multiclass classifier, namely ELM. The simulation results demonstrate that the proposed fault diagnosis approach can not only extract the essential features of fault response signals with better performance, and also achieve higher diagnosis accuracy than other reported approaches.
查看全文  查看/发表评论  下载PDF阅读器