Fault detection based on IJB-PCA-ICA

LIU Shurui; PENG Hui; LI Shuai; ZHOU Xiaofeng

doi:10.11949/j.issn.0438-1157.20180643

您当前的位置：

首页 >

文章列表页 >

Fault detection based on IJB-PCA-ICA

更新时间：2025-12-10

- Fault detection based on IJB-PCA-ICA
- CIESC Journal Issue 12, Pages: 5146-5154(2018)
- 作者机构：
  
  1. 中国科学院沈阳自动化研究所,辽宁,沈阳,110016
  2. 中国科学院大学,北京,100049
  3. 中国科学院网络化控制系统重点实验室,辽宁,沈阳,110016
  4. 中国科学院机器人与智能制造创新研究院,辽宁,沈阳,110016
- 作者简介：
- 基金信息：
- DOI：10.11949/j.issn.0438-1157.20180643
  CLC： P277
- Online First：05 December 2018，
  
  Published：05 December 2018
- 稿件说明：
移动端阅览
LIU Shurui, PENG Hui, LI Shuai, et al. Fault detection based on IJB-PCA-ICA[J]. CIESC Journal, 2018, (12): 5146-5154.
DOI：

LIU Shurui, PENG Hui, LI Shuai, et al. Fault detection based on IJB-PCA-ICA[J]. CIESC Journal, 2018, (12): 5146-5154. DOI： 10.11949/j.issn.0438-1157.20180643.

摘要

针对现代工业过程数据的高维性和分布复杂性等问题，提出了一种基于IJB-PCA-ICA（improved Jarque-Bera-principal component analysis-independent component analysis）的故障检测方法。首先采用改进的Jarque-Bera检测方法（J-B test）对原始数据划分高斯与非高斯核心部分，并对其中的高斯性与非高斯性均不明显的变量划分半高斯部分。将半高斯部分通过高斯分布置信概率加权与高斯核心部分和非高斯核心部分分别建立高斯子空间和分高斯子空间，然后对高斯子空间进行相关性划分后采用PCA方法得到高斯子空间的统计量；对非高斯子空间进行主元投影划分后采用ICA方法得到非高斯子空间的统计量，接着通过贝叶斯推断得到的联合统计量进行故障检测。最后通过Tenessee Eastman（TE）仿真实验，有效验证了所提出方法的有效性。

Abstract

In the view of the high dimensionality and the distribution complexity of modern industrial data

a fault detection method based on IJB-PCA-ICA (improved Jarque-Bera-principal component analysis-independent component analysis) is proposed. Through the method of Jarque-Bera test (J-B test)

the original data are divided into Gaussian part

non-Gaussian part and semi-Gaussian part. The semi-Gaussian part divided from those variables with not obvious Gaussianity or non-Gaussianity is weighted to participate into Gaussian subspace and non-Gaussian subspace by the Gaussian confidence probability. After the partition by the correlation and principal component projection

the statistics of the Gaussian and non-Gaussian subspaces are obtained by PCA and ICA

respectively. Then the Bayesian inference is applied to obtain the comprehensive statistics for fault detection.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Principal component selection algorithm based on ReliefF and its application in process monitoring

Research and prospect of early warning and diagnosis technology for ORC power generation system process

Industrial data driven transition state detection with multi-mode switching of a hydrocracking unit

Fault detection and diagnosis method based on weighted statistical feature KICA

Fault detection using hierarchical variational Gaussian mixture model and principal polynomial analysis

Related Author

TAO Yang

WANG Fan

SHI Hongbo

SONG Bing

WANG Jinjiang

LU Zhenjie

AN Weizheng

YANG Fengyun

Related Institution

Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education, East China University of Science and Technology

College of Artificial Intelligence, China University of Petroleum

College of Safety and Ocean Engineering, China University of Petroleum

Key Laboratory of Oil and Gas Production Equipment Quality Inspection and Health Diagnosis, State Administration for Market Regulation

CNOOC Research Institute Co., Ltd.

⁰