基于最优分数阶傅里叶变换的模拟电路故障特征提取新方法

被引：26

作者：

罗慧

王友仁

崔江

机构：

[1] 南京航空航天大学自动化学院

来源：

仪器仪表学报 | 2009年 / 30卷 / 05期

关键词：

分数阶傅里叶变换; 模拟电路; 故障诊断; 特征提取; 类内类间距离;

D O I：

10.19650/j.cnki.cjsi.2009.05.019

中图分类号：

TN710 [电子电路];

学科分类号：

摘要：

模拟电路故障诊断中故障特征提取是关键技术之一,为获得有效的故障特征信息,研究提出一种基于最优分数阶傅里叶变换(FRFT)的模拟电路故障特征提取方法。该方法把在原始数据空间中可分性差的特征数据映射到合适的分数阶空间,用类内类间距离作为判据,通过遗传算法确定分数阶p值,找出在分数阶Fourier域内类内类间距离最大的故障特征,然后用PCA降维和神经网络分类诊断。实验表明该方法能增强不同故障模式的可分性,提高了诊断结果的准确率。

引用

页码：997 / 1001

页数：5

共 7 条

[1] Adaptive Fault Diagnosis of Analog Circuits by Operation-Region Model and X–Y Zoning Method [J].

Yukiya Miura ;

Jiro Kato .

Journal of Electronic Testing, 2008, 24 :223-233

[2]

Soft Fault Diagnosis for Analog Circuits Based on Slope Fault Feature and BP Neural Networks[J] . Mei Hu,Hong Wang,Geng Hu,Shiyuan Yang.Tsinghua Science ＆ Technology . 2007

[3] Tuning of FIR filter transition bandwidth using fractional Fourier transform [J].

Sharma, S. N. ;

Saxena, Rajiv ;

Saxena, S. C. .

SIGNAL PROCESSING, 2007, 87 (12) :3147-3154

[4]

Research progress of the fractional Fourier transform in signal processing[J] . Ran Tao,Bing Deng,Yue Wang.Science in China Series F . 2006 (1)

[5]

Detection and parameter estimation of multicomponent LFM signal based on the fractional fourier transform[J] . Lin Qi,Ran Tao,Siyong Zhou,Yue Wang.Science in China Series F: Information Sciences . 2004 (2)

[6] Fractional Fourier transform pre-processing for neural networks and its application to object recognition [J].

Barshan, B ;

Ayrulu, B .

NEURAL NETWORKS, 2002, 15 (01) :131-140

[7]

Fault Diagnosis of Analog Circuits Using Bayesian Neural Networks with Wavelet Transform as Preprocessor[J] . Farzan Aminian,Mehran Aminian.Journal of Electronic Testing . 2001 (1)

← 1 →