[1]张 成,韩宏宇,李 元.去主元相关性 DKPCA 故障检测与诊断方法[J].计算机技术与发展,2023,33(04):161-167.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 024]
　ZHANG Cheng,HAN Hong-yu,LI Yuan.Fault Detection and Diagnosis Method Based on Removing Principal Component Correlation DKPCA[J].,2023,33(04):161-167.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 024]

点击复制

去主元相关性 DKPCA 故障检测与诊断方法()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年04期

页码:

161-167

栏目:

人工智能

出版日期:

2023-04-10

文章信息/Info

Title:

Fault Detection and Diagnosis Method Based on Removing Principal Component Correlation DKPCA

文章编号:

1673-629X(2023)04-0161-07

作者:

张 成¹ ; 韩宏宇² ; 李 元³

1. 沈阳化工大学 理学院,辽宁 沈阳 110142;

2. 沈阳化工大学 计算机科学与技术学院,辽宁 沈阳 110142;

3. 沈阳化工大学 信息工程学院,辽宁 沈阳 110142

Author(s):

ZHANG Cheng1 ; HAN Hong-yu2 ; LI Yuan3

1. School of Science,Shenyang University of Chemical Technology,Shenyang 110142,China;

2. School of Computer Science and Technology,Shenyang University of Chemical Technology,Shenyang 110142,China;

3. School of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China

关键词:

动态核主元分析; 过程监控; 自相关性; 去主元相关性; 故障诊断

Keywords:

dynamic kernel principal component analysis; process monitoring; autocorrelation; remove principal component correlation;  fault diagnosis

分类号:

TP277

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 04. 024

摘要:

针对动态核主元分析(Dynamic Kernel Principal Component Analysis,DKPCA) 在动态非线性过程监控中没有降

低数据动态性的影响,导致统计量 T2 具有显著自相关性的问题, 提出一种基于去主元相关性的 DKPCA ( Dynamic KernelPrincipal Component Analysis based on Removing Principal Component Correlation,DKPCA-RPCC) 故障检测与诊断方法。 首先,对原始数据 X 进行时滞扩展生成增广矩阵 Y 并使用 KPCA 计算主成分 M;其次,利用已知数据重构增广矩阵 Y^,再使用KPCA 计算主成分 M^ ;然后,通过主成分之间的差异来构建统计量进行故障检测;最后,利用基于变量贡献图的方法进行故障诊断。 通过数值例子和田纳西-伊斯曼( Tennessee-Eastman, TE) 过程进行仿真验证,并将仿真结果与 KPCA、DPCA 和DKPCA 的结果进行对比。 仿真结果说明,该方法在动态非线性过程监控中构建的统计量故障检测性能更高且具有较低的自相关性。

Abstract:

Aiming at the problem that Dynamic Kernel Principal Component Analysis ( DKPCA) does not reduce the influence of datadynamics in dynamic nonlinear process monitoring,resulting in significant autocorrelation?

of statistic T2 ,a fault detection and diagnosismethod for DKPCA based on removing principal component correlation ( DKPCA-RPCC) was proposed. Firstly,the original data X wastime-delayed to generate the augmented matrix Y ,and the principal component M was calculated by KPCA. Secondly,the known data

was used to reconstruct the augmented matrix Y^ and then calculate the main component M^ by KPCA.?

Then, the difference betweenprincipal components was used to construct statistics for fault detection. Finally,

the method based on variable contribution graph wasused for fault diagnosis. Numerical examples and Tennessee - Eastman ( TE ) process were used for simulation verification, and thesimulation results were compared with those of KPCA,DPCA and DKPCA. The simulation results show that the proposed method hashigher fault detection performance and lower autocorrelation in dynamic nonlinear process monitoring.

常用功能