Robust PPLS model and its applications in process monitoring

CHEN Jiayi; ZHAO Zhonggai; LIU Fei

doi:10.11949/j.issn.0438-1157.20151439

您当前的位置：

首页 >

文章列表页 >

Robust PPLS model and its applications in process monitoring

更新时间：2025-11-26

- Robust PPLS model and its applications in process monitoring
- CIESC Journal Issue 7, Pages: 2907-2915(2016)
- 作者机构：
  
  江南大学自动化研究所, 轻工过程先进控制教育部重点实验室,江苏,无锡,214122
- 作者简介：
- 基金信息：
- DOI：10.11949/j.issn.0438-1157.20151439
  CLC： P277
- Online First：05 July 2016，
  
  Published：05 July 2016
- 稿件说明：
移动端阅览
CHEN Jiayi, ZHAO Zhonggai, LIU Fei. Robust PPLS model and its applications in process monitoring[J]. CIESC Journal, 2016, (7): 2907-2915.
DOI：

CHEN Jiayi, ZHAO Zhonggai, LIU Fei. Robust PPLS model and its applications in process monitoring[J]. CIESC Journal, 2016, (7): 2907-2915. DOI： 10.11949/j.issn.0438-1157.20151439.

摘要

概率偏最小二乘（PPLS）模型建立的条件是主元和误差都服从高斯分布，但是高斯分布的期望和方差容易受到离群点的影响，导致模型的鲁棒性较差。针对PPLS模型的不足，提出一种鲁棒概率偏最小二乘（RPPLS）方法，用拖尾更宽的T分布代替高斯分布，通过调整自由度参数，使模型对含离群点数据的拟合效果更好。更进一步，将RPPLS引入过程监控中，提出GT2和GSPE两个监控指标，分别监控过程的受控状态以及模型关系的变化。PPLS和RPPLS在TE过程监控的应用结果表明RPPLS不仅能更准确检测故障的产生，而且能更有效降低故障的漏报率。

Abstract

A probability model can be developed by probabilistic partial least squares (PPLS) under the conditions that both principal components and errors satisfy Gaussian distribution. However

the expectation and variance of the Gaussian distribution is susceptible to outliers. As a result

the model is not robust for the real industrial process. This paper improves the robustness of the PPLS model based on the assumption that the raw data satisfy T distribution rather than Gaussian distribution. By adjusting the freedom degree of T distribution

the proposed robust probabilistic partial least squares (RPPLS) model can overcome the shortcomings of PPLS model. Furthermore

on the basis of RPPLS model

two monitoring indicators GT2 and GSPE are proposed to monitor the process state and the model changes

respectively. Comparing the monitoring performance in the TE process based on PPLS and RPPLS shows that RPPLS is more effective than PPLS in terms of the fault accuracy and the missing alarm rate.

关键词

Keywords

references

Views

187

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Recognizing historical operating conditions by determining the density peaks at kernel density estimation of heat diffusion

Interval model for predicting effluent quality variables of wastewater treatment plants

Prediction of lipid yield and substrate distribution based on consumed C/N ratio

Triboelectrification characteristics of associated minerals in coal

Water bloom prediction and factor analysis based on multidimensional time series analysis

Related Author

Rongshan BI

Zhihui HAN

Shaohui TAO

Xiaoyan SUN

Shuguang XIANG

Wei CHAI

Longhang GUO

Binbin CHI

Related Institution

College of Chemical Engineering, Qingdao University of Science and Technology

Faculty of Information Technology, School of Automation, Beijing University of Technology

Beijing Key Laboratory of Computational Intelligence and Intelligent Systems

Dalian Institute of Chemical Physics, Chinese Academy of Sciences

State Key Laboratory of Chemical Engineering, Department of Chemical and Biological Engineering, Zhejiang University

⁰