Department of Control Science, University of Science, Pyongyang 355, North Korea
BibTex Citation Data :
@article{JCERP20856, author = {CholJin O and Hyok Jo and GumHyok Pak and Jusong Rim and Boknam Huang and Cholryong Huang}, title = {Sensitivity Improvement of an In-Line Electromagnetic Metal Detector for Chemical Process Streams Using a Radial Basis Function Neural Network}, journal = {Journal of Chemical Engineering Research Progress}, volume = {4}, number = {1}, year = {2027}, keywords = {metal detector; electromagnetic induction; chemical process; product effect; sensitivity; radial basis function neural network; orthogonal least squares; in-line instrumentation}, abstract = { Tramp-metal contamination in chemical process streams damages crushers, mills, and extruders, degrades product purity, and can provide ignition sources in combustible-dust atmospheres. In-line electromagnetic induction (EMI) metal detectors are the standard countermeasure, but their practical sensitivity is limited by the product effect—the channel-correlated background signal produced by conductive or moist process materials—which forces conservative detection thresholds and causes false alarms or missed small fragments. This paper presents an in-line EMI metal detector whose sensitivity and false-alarm stability are improved by a radial basis function (RBF) neural network that estimates the size and cross-sectional position of a metallic fragment from the induced voltages of a six-coil array mounted on a process pipe. A prototype was dynamically validated over 200 passes with 125 metallic objects, achieving 99.2% detection accuracy and a 3 mm minimum detectable radius. Because the prototype's component-level documentation is no longer available, a fully parameterized simulation of the in-line configuration was conducted with representative design values and a product-effect model: on 2400 training and 1500 test samples, the RBF network matched a conventional amplitude-threshold detector at the nominal product level (AUC 0.949 versus 0.962) while a two-hidden-layer multilayer perceptron baseline performed comparably (AUC 0.945); the decisive advantage appeared under product-effect drift, where the false-alarm rate of the RBF detector rose to only 20.3% at 2.5× the nominal product level against 71.1% for the amplitude threshold, without recalibration. The RBF estimator also localized fragments with a mean absolute error of 2.0 mm (pipe radius 45 mm), an estimation capability the threshold detector does not provide. These results indicate that a lightweight trained estimator can substantially improve the operability of in-line metal detection in chemical plants. Copyright © 2027 by Authors, Published by Universitas Diponegoro and BCREC Publishing Group. This is an open access article under the CC BY-SA License ( https://creativecommons.org/licenses/by-sa/4.0 ). }, issn = {3032-7059}, pages = {45--52} doi = {10.9767/jcerp.20856}, url = {https://journal.bcrec.id/index.php/jcerp/article/view/20856} }
Refworks Citation Data :
Tramp-metal contamination in chemical process streams damages crushers, mills, and extruders, degrades product purity, and can provide ignition sources in combustible-dust atmospheres. In-line electromagnetic induction (EMI) metal detectors are the standard countermeasure, but their practical sensitivity is limited by the product effect—the channel-correlated background signal produced by conductive or moist process materials—which forces conservative detection thresholds and causes false alarms or missed small fragments. This paper presents an in-line EMI metal detector whose sensitivity and false-alarm stability are improved by a radial basis function (RBF) neural network that estimates the size and cross-sectional position of a metallic fragment from the induced voltages of a six-coil array mounted on a process pipe. A prototype was dynamically validated over 200 passes with 125 metallic objects, achieving 99.2% detection accuracy and a 3 mm minimum detectable radius. Because the prototype's component-level documentation is no longer available, a fully parameterized simulation of the in-line configuration was conducted with representative design values and a product-effect model: on 2400 training and 1500 test samples, the RBF network matched a conventional amplitude-threshold detector at the nominal product level (AUC 0.949 versus 0.962) while a two-hidden-layer multilayer perceptron baseline performed comparably (AUC 0.945); the decisive advantage appeared under product-effect drift, where the false-alarm rate of the RBF detector rose to only 20.3% at 2.5× the nominal product level against 71.1% for the amplitude threshold, without recalibration. The RBF estimator also localized fragments with a mean absolute error of 2.0 mm (pipe radius 45 mm), an estimation capability the threshold detector does not provide. These results indicate that a lightweight trained estimator can substantially improve the operability of in-line metal detection in chemical plants. Copyright © 2027 by Authors, Published by Universitas Diponegoro and BCREC Publishing Group. This is an open access article under the CC BY-SA License (https://creativecommons.org/licenses/by-sa/4.0).
Article Metrics:
Last update:
Authors who publish in the Journal of Chemical Engineering Research Progress (JCERP) retain full copyright ownership of their work. In keeping with the journal’s commitment to open access, all articles are published under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0).
This license permits anyone to access, use, share, adapt, remix, transform, and build upon the work for any purpose, including commercial use, provided that appropriate credit is given to the original author or authors, a link to the license is provided, changes to the work (if any) are clearly indicated, and any derivative works are distributed under the same license.
Authors are encouraged to disseminate their work as widely as possible. They retain the right to reuse their published article in future scholarly works, such as books, conference presentations, or teaching materials. They may also deposit the final published version in institutional or subject-based repositories, and share it freely on personal websites, academic platforms, or professional networks. These rights are fully preserved under the CC BY-SA license, and all such uses must comply with its terms.
Readers and third parties may also use the content in accordance with the CC BY-SA license. This includes the ability to reproduce, modify, and build upon the article, even for commercial purposes, as long as proper attribution is given, and any resulting work is distributed under the same license.
License to Publish Agreement (Non-Exclusive License for Publishing Rights)
To enable publication and global dissemination of accepted manuscripts, the Journal of Chemical Engineering Research Progress is published by UPT Laboratorium Terpadu Universitas Diponegoro jointly with Masyarakat Katalis Indonesia - Indonesian Catalyst Society (MKICS) Publisher, and the technical management of the JCERP journal is supported by with BCREC Publishing Group, requires that authors grant the publisher a non-exclusive license to publish the work. This license authorizes the publisher to reproduce, distribute, and communicate the article to the public in all forms and media. The license to publish does not transfer copyright; authors remain the sole copyright holders.
This arrangement is formalized through a License to Publish Agreement, which the corresponding author must complete after the manuscript is accepted for publication. The agreement confirms that the author or authors grant the publisher the non-exclusive right to publish the article and to distribute it under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0).
Authors retain all other rights to their work. They remain free to use and share their article in any manner consistent with the terms of the CC BY-SA license, including in future publications, educational settings, and commercial applications, provided proper attribution is given and the license is preserved.
After acceptance, the corresponding author will receive an email containing instructions for completing and electronically signing the License to Publish Agreement. The signed agreement must be returned to the Editorial Office in order to proceed with publication. The License to Publish Agreement form is available for download on the journal’s official website.
The non-exclusive Transfer Agreement for Publishing Right (RTAP) Form can be downloaded here: [Transfer Agreement for Publishing Right (RTAP) Form JCERP 2024]
The (non-exclusive) publishing right form should be signed electronically and send to the Editorial Office in the form of original e-mail below: Prof. Dr. I. Istadi (Editor-in-Chief)Editorial Office of Journal of Chemical Engineering Researc ProgressLaboratory of Plasma-Catalysis (R3.5), UPT Laboratorium Terpadu, Universitas DiponegoroJl. Prof. Soedarto, Semarang, Central Java, Indonesia 50275Telp/Whatsapp: +62-81-316426342E-mail: jcerp[at]live.undip.ac.id
(This policy statements has been updated at 25th August 2026)
View statistics
The Journal of Chemical Engineering Research Progress is published by UPT Laboratorium Terpadu Universitas Diponegoro jointly with Masyarakat Katalis Indonesia - Indonesian Catalyst Society (MKICS) Publisher. The technical management of the JCERP journal is supported by with BCREC Publishing Group.