| 模拟电路故障诊断的改进分析 |
| |
| 引用本文： | 李涛柱,李红波,曾繁景,李铁峰.模拟电路故障诊断的改进分析[J].信息工程大学学报,2012,13(5):568-572. |
| |
| 作者姓名： | 李涛柱 李红波 曾繁景 李铁峰 |
| |
| 作者单位： | 信息工程大学信息系统工程学院,河南郑州450002 |
| |
| 摘 要： | 针对模拟电路故障诊断识别率较低的问题,文章结合模拟电路智能故障诊断流程的重要环节对特征选择、特征提取和诊断识别进行了改进分析。首先将支持向量机(support vectormachine,SVM)和传统特征选择算法相结合,改进了现有特征选择算法,接着将主成分分析(principle component analysis,PCA)和独立成分分析(independent component analysis,ICA)相结合提出双空间特征提取算法,并将双空间提取算法和融合特权信息支持向量机(SVM of learn-ing using privileged information,LUPI-SVM)算法相结合,提出基于双空间提取算法的融合特权信息支持向量机模拟电路故障诊断方法。最后对改进后方法进行了应用分析,通过对两个典型电路的仿真测试,验证了改进后方法的可行性和有效性,改进后方法提高了模拟电路故障诊断的性能。
|
| 关 键 词： | 特征选择 特征提取 模拟电路 |
Improvement and Analysis of Analog Circuits Fault Diagnosis |
| |
| LI Tao zhu,LI Hong bo,ZENG Fan jing,LI Tie feng.Improvement and Analysis of Analog Circuits Fault Diagnosis[J].Journal of Information Engineering University,2012,13(5):568-572. |
| |
| Authors: | LI Tao zhu LI Hong bo ZENG Fan jing LI Tie feng |
| |
| Affiliation: | (Institute of Information systems Engineering,Information Engineering University, Zhengzhou 450002, China) |
| |
| Abstract: | To improve accuracy of analog circuit fault diagnosis, this paper suggests to optimize criti- cal steps including feature selection, feature extraction and fault diagnosis. Feature selection is en- hanced by combining the traditional algorithm with the support vector machine （SVM）. Based on principal component analysis （PCA） and independent component analysis （ICA） , a double-space feature extraction algorithm is introduced. And then the double-space feature extraction algorithm is combined with the SVM of Learning Using Privileged Information （LUPI-SVM）. This improved method is simulated and analyzed, with its feasibility and effectiveness validated. |
| |
| Keywords: | feature selection feature extraction analog circuits |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《信息工程大学学报》浏览原始摘要信息 |
|
点击此处可从《信息工程大学学报》下载全文 |
|
