University of Oulu

Tero Vuolio, Ville-Valtteri Visuri, Aki Sorsa, Seppo Ollila, Timo Fabritius, Application of a genetic algorithm based model selection algorithm for identification of carbide-based hot metal desulfurization, Applied Soft Computing, Volume 92, 2020, 106330, ISSN 1568-4946, https://doi.org/10.1016/j.asoc.2020.106330

Application of a genetic algorithm based model selection algorithm for identification of carbide-based hot metal desulfurization

Saved in:
Author: Vuolio, Tero1; Visuri, Ville-Valtteri1; Sorsa, Aki2;
Organizations: 1Process Metallurgy Research Unit, University of Oulu, P.O. Box 4300, FI–90014, University of Oulu, Finland
2Control Engineering, University of Oulu, P.O. Box 4300, FI–90014, University of Oulu, Finland
3SSAB Europe Oy, Rautaruukintie 155, P.O. Box 93, FI–92101, Raahe, Finland
Format: article
Version: accepted version
Access: embargoed
Persistent link: http://urn.fi/urn:nbn:fi-fe20201215100721
Language: English
Published: Elsevier, 2020
Publish Date: 2022-04-20
Description:

Abstract

Sulfur is considered as one of the main impurities in hot metal. Hot metal desulfurization is often carried out with pneumatic injection of a fine-grade desulfurization reagent using a submerged lance. The aim of this study was to develop a data-driven model for the process. The model selection algorithm carries out a simultaneous variable selection and optimization of number of hidden neurons with a combination of binary and integer coded Genetic Algorithm. The objective function applied in the search is repeated Leave-Multiple-Out cross-validation. The model considered is a feedforward neural network with a single hidden layer. In the inner loop of the algorithm, the computational load is reduced by making use of Extreme Learning Machine (ELM) architecture. The final model is trained using the Bayesian regularization. The results show that a well-generalizing data-driven model with good prediction performance can be repeatedly selected based on noisy industrial data with the help of a Genetic Algorithm, provided that the model is validated comprehensively with internal and external data sets.

see all

Series: Applied soft computing
ISSN: 1568-4946
ISSN-E: 1872-9681
ISSN-L: 1568-4946
Volume: 92
Article number: 106330
DOI: 10.1016/j.asoc.2020.106330
OADOI: https://oadoi.org/10.1016/j.asoc.2020.106330
Type of Publication: A1 Journal article – refereed
Field of Science: 215 Chemical engineering
213 Electronic, automation and communications engineering, electronics
Subjects:
Funding: This work was conducted within the Symbiosis of Metal Production and Nature (SYMMET) research program funded by Business Finland. The financial support of Technology Industries of the Finland Centennial Foundation, the Tauno Tönning foundation, the Finnish Foundation of Technology Promotion and Walter Ahlström foundation are also acknowledged.
Copyright information: © 2020 Elsevier B.V. All rights reserved. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/.
  https://creativecommons.org/licenses/by-nc-nd/4.0/