电光与控制, 2010, 17 (5): 66, 网络出版: 2010-06-22   

基于小波支持向量机的模拟电路故障诊断

Fault Diagnosis of Analog Circuit Based on Wavelet Support Vector Machine
作者单位
军械工程学院 光学与电子工程系,石家庄 050003
摘要
在模拟电路故障诊断中,提出了利用小波分析与支持向量机结合的系统方法,利用小波变换对信号进行特征提取得到特征向量并作为支持向量机的训练向量,得到故障分类器。针对激励信号必须能够充分地激励电路的需求,提出一种通用激励信号--连续多抽样函数,利用抽样函数在带通区间内频谱分布均匀且能量相同这一特点作为模拟电路的通用激励信号。仿真结果表明,该激励条件下,利用小波-支持向量机能够较好地对模拟电路进行故障诊断。
Abstract
It is proposed to use the combination of wavelet and Support Vector Machine(SVM) for analog circuit fault diagnosis. The wavelet transform was used to extract the features of the signal, and the obtained feature vector was used as the training vector of SVM for obtaining the fault grader. The stimuli should be able to excite the Circuit Under Test (CUT) so that the faulty-induced effect can be detected. A general excitation signal, a continuous multi-sampling function, was proposed in this paper. The spectrum distribution of sampling function is uniform and energy is the same, so that the signal can be a good stimulus. The results of simulation showed that this signal can excite CUT efficiently, and accuracy of classification of wavelet SVM was good.

张岐龙, 单甘霖, 段修生, 尚裕萌. 基于小波支持向量机的模拟电路故障诊断[J]. 电光与控制, 2010, 17(5): 66. ZHANG Qilong, SHAN Ganlin, DUAN Xiusheng, SHANG Yumeng. Fault Diagnosis of Analog Circuit Based on Wavelet Support Vector Machine[J]. Electronics Optics & Control, 2010, 17(5): 66.

本文已被 2 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!