LAPSE:2022.0120
Published Article
LAPSE:2022.0120
A Modified Expectation Maximization Approach for Process Data Rectification
Weiwei Jiang, Rongqiang Li, Deshun Cao, Chuankun Li, Shaohui Tao
October 31, 2022
Process measurements are contaminated by random and/or gross measuring errors, which degenerates performances of data-based strategies for enhancing process performances, such as online optimization and advanced control. Many approaches have been proposed to reduce the influence of measuring errors, among which expectation maximization (EM) is a novel and parameter-free one proposed recently. In this study, we studied the EM approach in detail and argued that the original EM approach is not feasible to rectify measurements contaminated by persistent biases, which is a pitfall of the original EM approach. So, we propose a modified EM approach here to circumvent this pitfall by fixing the standard deviation of random error mode. The modified EM approach was evaluated by several benchmark cases of process data rectification from literatures. The results show advantages of the proposed approach to the original EM in solving efficiency and performance of data rectification.
Keywords
bias detection, data rectification, expectation maximization
Suggested Citation
Jiang W, Li R, Cao D, Li C, Tao S. A Modified Expectation Maximization Approach for Process Data Rectification. (2022). LAPSE:2022.0120
Author Affiliations
Jiang W: State Key Laboratory of Safety and Control for Chemicals, SINOPEC Qingdao Research Institute of Safety Engineering, Qingdao 266071, China
Li R: State Key Laboratory of Safety and Control for Chemicals, SINOPEC Qingdao Research Institute of Safety Engineering, Qingdao 266071, China
Cao D: State Key Laboratory of Safety and Control for Chemicals, SINOPEC Qingdao Research Institute of Safety Engineering, Qingdao 266071, China
Li C: State Key Laboratory of Safety and Control for Chemicals, SINOPEC Qingdao Research Institute of Safety Engineering, Qingdao 266071, China
Tao S: College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao 266042, China [ORCID]
Journal Name
Processes
Volume
9
Issue
2
First Page
270
Year
2021
Publication Date
2021-01-30
Published Version
ISSN
2227-9717
Version Comments
Original Submission
Other Meta
PII: pr9020270, Publication Type: Journal Article
Record Map
Published Article

LAPSE:2022.0120
This Record
External Link

doi:10.3390/pr9020270
Publisher Version
Download
Files
[Download 1v1.pdf] (11.2 MB)
Oct 31, 2022
Main Article
License
CC BY 4.0
Meta
Record Statistics
Record Views
55
Version History
[v1] (Original Submission)
Oct 31, 2022
 
Verified by curator on
Oct 31, 2022
This Version Number
v1
Citations
Most Recent
This Version
URL Here
https://psecommunity.org/LAPSE:2022.0120
 
Original Submitter
Mina Naeini
Links to Related Works
Directly Related to This Work
Publisher Version