Regular Papers

International Journal of Control, Automation and Systems 2017; 15(3): 1466-1477

Published online May 22, 2017

https://doi.org/10.1007/s12555-015-0463-7

© The International Journal of Control, Automation, and Systems

Elongation Prediction of Steel-Strips in Annealing Furnace with Deep Learning via Improved Incremental Extreme Learning Machine

Chao Wang*, Jian-HuiWang, Shu-Sheng Gu, XiaoWang, and Yu-Xian Zhang

Northeastern University

Abstract

The elongation of steel-strips in annealing furnace is an important factor that affects the position of welding line and safety of air-knife since there is no extra space to install welding line detector in field conditions. Therefore, predicting the elongation of steel-strips in the annealing process is important to fulfill the requirements of eliminating security risks and improving economic performance. In this paper, we propose a deep architectures called I-ELM/MLCSA autoencoders with the concept of stacked generalization philosophy to solve large and complex data mining problems. The comparison results of the case studies indicate that D-ELMs-AE/MLCSA is a promising prediction algorithm and can be employed for steel-strips elongation predictions with excellent performance."

Keywords Baldwinian learning, Clone selection algorithm, deep learning, elongation prediction, incremental extreme learning machine, Lamarckian learning.

Article

Regular Papers

International Journal of Control, Automation and Systems 2017; 15(3): 1466-1477

Published online June 1, 2017 https://doi.org/10.1007/s12555-015-0463-7

Copyright © The International Journal of Control, Automation, and Systems.

Elongation Prediction of Steel-Strips in Annealing Furnace with Deep Learning via Improved Incremental Extreme Learning Machine

Chao Wang*, Jian-HuiWang, Shu-Sheng Gu, XiaoWang, and Yu-Xian Zhang

Northeastern University

Abstract

The elongation of steel-strips in annealing furnace is an important factor that affects the position of welding line and safety of air-knife since there is no extra space to install welding line detector in field conditions. Therefore, predicting the elongation of steel-strips in the annealing process is important to fulfill the requirements of eliminating security risks and improving economic performance. In this paper, we propose a deep architectures called I-ELM/MLCSA autoencoders with the concept of stacked generalization philosophy to solve large and complex data mining problems. The comparison results of the case studies indicate that D-ELMs-AE/MLCSA is a promising prediction algorithm and can be employed for steel-strips elongation predictions with excellent performance."

Keywords: Baldwinian learning, Clone selection algorithm, deep learning, elongation prediction, incremental extreme learning machine, Lamarckian learning.

IJCAS
February 2025

Vol. 23, No. 2, pp. 359~682

Stats or Metrics

Share this article on

  • line

Related articles in IJCAS

IJCAS

eISSN 2005-4092
pISSN 1598-6446