Multi-mode process monitoring method based on PCA mixture model

许仙珍

您当前的位置：

首页 >

文章列表页 >

Multi-mode process monitoring method based on PCA mixture model

更新时间：2025-12-10

- Multi-mode process monitoring method based on PCA mixture model
- CIESC Journal Issue 3, (2011)
- 作者机构：
  
  浙江大学智能系统与控制研究所，工业控制技术国家重点实验室
- 作者简介：
- 基金信息：
- DOI：
  CLC：
- Online First：05 March 2011，
  
  Published：05 March 2011
- 稿件说明：
移动端阅览
许仙珍. Multi-mode process monitoring method based on PCA mixture model[J]. CIESC Journal, 2011, (3).
DOI：

许仙珍. Multi-mode process monitoring method based on PCA mixture model[J]. CIESC Journal, 2011, (3). DOI：

摘要

针对传统多元统计故障检测方法大多假设测量数据服从单一高斯分布的不足，提出了一种基于PCA(principal component analysis)混合模型的多工况过程监测方法。首先通过直接对混合模型的各高斯成分的协方差进行PCA降维变换，使得协方差阵对角化，既减少了运算量又避免了变量相关而导致的奇异性问题；同时采用BYY增量EM算法自动获取混合模型的最佳混合分量数目，避免了常规EM算法的不足。所得的混合模型，除包括均值、协方差和先验概率等参数外，还包括了PCA载荷阵，即对每个混合元建立了PCA模型。然后给出了统计量定义，实现对多工况过程的故障检测。数值例子和TE过程的应用表明，本文提出的方法无需过程先验知识，能自动获取工况数目、精确估计各个工况的统计特性，并更准确及时地检测出多工况过程的各种故障。

Abstract

Traditional multivariate statistical fault detection methods are designed for single operating conditions and may produce erroneous conclusions if used for multi-mode process monitoring.A novel multi-mode process monitoring approach based on PCA mixture model is proposed in this paper.First

the PCA technique is used to reduce the dimension of original variables and to guarantee the nonsingular covariance matrix.In order to overcome the limitations of EM (expectation-maximization)

the BYY(Bayes Ying-Yang)scale-incremental EM algorithm is then adopted to automatically optimize the number of mixture components.With the obtained PCA mixture model

a novel process monitoring scheme is derived for fault detection of multi-mode processes.The validity and effectiveness of the proposed monitoring approach are illustrated by numerical example and the TE process.The results show that the proposed algorithm can achieve good parameter estimation of the mixture model with correct model selection.Therefore

it can achieve accurate and early detection of various types of faults in multi-mode processes.

关键词

Keywords

references

Views

1087

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Mixed matrix membranes based on nitrogen-rich Schiff base network polymer and their CO2 separation performance

Study on performance enhancement and cold retention characteristics of solid packed bed cold energy storage

Prediction model for COD in petrochemical wastewater influent based on temporal-spatial dual-stream convolutional neural network

Thermodynamic study of germanium-impurity competitive deposition in GeCl4-MxCly-H2 system

Corrosion health status assessment method for refining unit pipeline based on GLAE-MLP

Related Author

No data

Related Institution

No data

⁰