Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms

Jose Gomez Fuentes; Sirkka-Liisa Jämsä-Jounela; David Moseley; Tom Skuse

doi:10.1109/SYSTOL.2019.8864797

Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms

Jose Gomez Fuentes, Sirkka-Liisa Jämsä-Jounela, David Moseley, Tom Skuse

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

27 Downloads (Pure)

Abstract

An enhanced control strategy for a multiple hearth furnace for the purpose of kaolin production is developed and presented in this paper. Mineralogy-driven machine learning algorithms play a key role in the optimization strategy of the furnace. First, the capacity and temperature setpoints for furnace control are determined based on the feed ore mineralogy. Next, the capacity is optimized by combining the prediction of soluble alumina content and mullite content, while maintaining the quality of the product. The stabilizing control level compensates the disturbances with a feedforward control, which uses a spinel phase reaction rate soft sensor, aimed at minimizing the energy use of the furnace. The control concept is successfully tested by simulation using industrial data. Finally, a sampling campaign and software testing of the soft sensors and machine learning algorithms are performed at the industrial site. The results are presented and discussed in the paper.

Original language	English
Title of host publication	4th Conference on Control and Fault Tolerant Systems (SysTol)
Publisher	IEEE
Pages	250-256
ISBN (Electronic)	978-1-7281-0380-8
ISBN (Print)	978-1-7281-0381-5
DOIs	https://doi.org/10.1109/SYSTOL.2019.8864797
Publication status	Published - 14 Oct 2019
MoE publication type	A4 Article in a conference publication
Event	International Conference on Control and Fault-Tolerant Systems - Casabalanca, Morocco Duration: 18 Sep 2019 → 20 Sep 2019 Conference number: 4 http://www.systol.org/systol19/

Publication series

Name	Conference on Control and Fault Tolerant Systems (SysTol)
Publisher	IEEE
ISSN (Print)	2162-1195
ISSN (Electronic)	2162-1209

Conference

Conference	International Conference on Control and Fault-Tolerant Systems
Abbreviated title	SysTol
Country	Morocco
City	Casabalanca
Period	18/09/2019 → 20/09/2019
Internet address	http://www.systol.org/systol19/

Keywords

Mineral processing
Quality control
Multiple hearth furnace (MHF)
Soft sensor
Advanced Control

Access to Document

10.1109/SYSTOL.2019.8864797

CHEM-Fuentes et al-Control strategy-IEEEAccepted author manuscript, 810 KBLicence: Unspecified

Fingerprint Dive into the research topics of 'Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms'. Together they form a unique fingerprint.

Cite this

Gomez Fuentes, J., Jämsä-Jounela, S-L., Moseley, D., & Skuse, T. (2019). Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms. In 4th Conference on Control and Fault Tolerant Systems (SysTol) (pp. 250-256). (Conference on Control and Fault Tolerant Systems (SysTol)). IEEE. https://doi.org/10.1109/SYSTOL.2019.8864797

@inproceedings{e6f58ffd3d47456393523f34288bff06,

title = "Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms",

abstract = "An enhanced control strategy for a multiple hearth furnace for the purpose of kaolin production is developed and presented in this paper. Mineralogy-driven machine learning algorithms play a key role in the optimization strategy of the furnace. First, the capacity and temperature setpoints for furnace control are determined based on the feed ore mineralogy. Next, the capacity is optimized by combining the prediction of soluble alumina content and mullite content, while maintaining the quality of the product. The stabilizing control level compensates the disturbances with a feedforward control, which uses a spinel phase reaction rate soft sensor, aimed at minimizing the energy use of the furnace. The control concept is successfully tested by simulation using industrial data. Finally, a sampling campaign and software testing of the soft sensors and machine learning algorithms are performed at the industrial site. The results are presented and discussed in the paper.",

keywords = "Mineral processing, Quality control, Multiple hearth furnace (MHF), Soft sensor, Advanced Control",

author = "{Gomez Fuentes}, Jose and Sirkka-Liisa J{\"a}ms{\"a}-Jounela and David Moseley and Tom Skuse",

year = "2019",

month = "10",

day = "14",

doi = "10.1109/SYSTOL.2019.8864797",

language = "English",

isbn = "978-1-7281-0381-5",

series = "Conference on Control and Fault Tolerant Systems (SysTol)",

publisher = "IEEE",

pages = "250--256",

booktitle = "4th Conference on Control and Fault Tolerant Systems (SysTol)",

address = "United States",

}

Gomez Fuentes, J , Jämsä-Jounela, S-L, Moseley, D & Skuse, T 2019, Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms. in 4th Conference on Control and Fault Tolerant Systems (SysTol). Conference on Control and Fault Tolerant Systems (SysTol), IEEE, pp. 250-256, International Conference on Control and Fault-Tolerant Systems, Casabalanca, Morocco, 18/09/2019. https://doi.org/10.1109/SYSTOL.2019.8864797

Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms. / Gomez Fuentes, Jose ; Jämsä-Jounela, Sirkka-Liisa; Moseley, David; Skuse, Tom.

4th Conference on Control and Fault Tolerant Systems (SysTol). IEEE, 2019. p. 250-256 (Conference on Control and Fault Tolerant Systems (SysTol)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Control Strategy of a Multiple Hearth Furnace Enhanced by Machine Learning Algorithms

AU - Gomez Fuentes, Jose

AU - Jämsä-Jounela, Sirkka-Liisa

AU - Moseley, David

AU - Skuse, Tom

PY - 2019/10/14

Y1 - 2019/10/14

N2 - An enhanced control strategy for a multiple hearth furnace for the purpose of kaolin production is developed and presented in this paper. Mineralogy-driven machine learning algorithms play a key role in the optimization strategy of the furnace. First, the capacity and temperature setpoints for furnace control are determined based on the feed ore mineralogy. Next, the capacity is optimized by combining the prediction of soluble alumina content and mullite content, while maintaining the quality of the product. The stabilizing control level compensates the disturbances with a feedforward control, which uses a spinel phase reaction rate soft sensor, aimed at minimizing the energy use of the furnace. The control concept is successfully tested by simulation using industrial data. Finally, a sampling campaign and software testing of the soft sensors and machine learning algorithms are performed at the industrial site. The results are presented and discussed in the paper.

AB - An enhanced control strategy for a multiple hearth furnace for the purpose of kaolin production is developed and presented in this paper. Mineralogy-driven machine learning algorithms play a key role in the optimization strategy of the furnace. First, the capacity and temperature setpoints for furnace control are determined based on the feed ore mineralogy. Next, the capacity is optimized by combining the prediction of soluble alumina content and mullite content, while maintaining the quality of the product. The stabilizing control level compensates the disturbances with a feedforward control, which uses a spinel phase reaction rate soft sensor, aimed at minimizing the energy use of the furnace. The control concept is successfully tested by simulation using industrial data. Finally, a sampling campaign and software testing of the soft sensors and machine learning algorithms are performed at the industrial site. The results are presented and discussed in the paper.

KW - Mineral processing

KW - Quality control

KW - Multiple hearth furnace (MHF)

KW - Soft sensor

KW - Advanced Control

U2 - 10.1109/SYSTOL.2019.8864797

DO - 10.1109/SYSTOL.2019.8864797

M3 - Conference contribution

SN - 978-1-7281-0381-5

T3 - Conference on Control and Fault Tolerant Systems (SysTol)

SP - 250

EP - 256

BT - 4th Conference on Control and Fault Tolerant Systems (SysTol)

PB - IEEE

ER -