Long short-term memory neural networks for clogging detection in the submerged entry nozzle

  • Ana P. M. Diniz Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil
  • Patrick M. Ciarelli Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil
  • Evandro O. T. Salles Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil
  • Klaus F. Coco Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil
Keywords: Continuous Casting, Submerged Entry Nozzle, Clogging, LSTM, Deep Learning.


The clogging in the Submerged Entry Nozzle (SEN), responsible for controlling the steel flow in continuous casting, is one of the main problems faced by steelmaking process, since it can increase the frequency of interruptions in the operation for the maintenance and/or exchange of its equipment. Although it is a problem inherent to the process, not identifying the clogging can result in losses associated with the process yield, as well as compromising the product quality. In order to detect the occurrences of clogging in a real steel industry from historical data of process variables, in this paper, different models of Long Short-Term Memory (LSTM) neural networks were tested and discussed. The overall performance of the classifiers developed here showed very promising results in real data with class imbalance.


Download data is not yet available.


Barati, H.,Wu, M., Kharicha, A. & Ludwig, A. (2018). A transient model for nozzle clogging. Powder Technology, 329, 181-198. DOI: https://doi.org/10.1016/j.powtec.2018.01.053

Bessho, N., Yoda, R., Yamasaki, H., Fujii, T., Nozaki, T. & Takatori, S. (1991). Numerical analysis of fuid flow in continuous casting mold by bubble dispersion model. ISIJ international, 31(1), 40-45. DOI: https://doi.org/10.2355/isijinternational.31.40

Boughorbel, S., Jarray, F. & El-Anbari, M. (2017) Optimal classifier for imbalanced data using Matthews Correlation Coeficient metric. PloS one, 12(6), e0177678. DOI: https://doi.org/10.1371/journal.pone.0177678

Buduma, N. & Locascio, N. (2017). Fundamentals of deep learning: Designing next generation machine intelligence algorithms. Sebastopol: O'Reilly Media, Inc.

Essien, A. & Giannetti, C. (2020). A deep learning model for smart manufacturing using convolutional LSTM neural network autoencoders. IEEE Transactions on Industrial Informatics, 16(9), 6069-6078.

Fawcett, T. (2016). An introduction to ROC analysis, Pattern recognition letters, 27(8), 861-874. DOI: https://doi.org/10.1016/j.patrec.2005.10.010

Goodfellow, I., Bengio, Y. & Courville, A. (2016). Deep Learning. Cambridge: MIT Press.

Haykin, S. O. (2011). Neural Networks and Learning Machines. New Jersey: Pearson Education.

Ikaheimonen, J., Leiviska, K., Ruuska, J. & Matkala, J. (2002). Nozzle clogging prediction in continuous casting of steel. IFAC Proceedings Volumes, 35(1), 143-147. DOI: https://doi.org/10.3182/20020721-6-ES-1901.01175

Kingma, D. P. & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.

Mourão, M. B., Yokoji, A., Malynowskyj, A., Leandro, C. A. S., Takano, C., Quites, E. E. C., Gentile, E. F., Silva, G. F. B. L., Bolota, J. R. & Gonçalves, M. (2011). Introdução à Siderurgia. São Paulo: Associação Brasileira de Metalurgia, Materiais e Mineração.

Pellegrini, G., Sandri, M., Villagrossi, E., Challapalli, S., Cestari, L., Polo, A. & Ometto, M. (2019). Successful use case applications of artificial intelligence in the steel industry. In: Iron & Steel Technology (AIST), Digital Transformations, 44-53.

Rackers, K. G. (1995). Mechanism and mitigation of clogging in continuous casting nozzles. Master's thesis, University of Illinois at Urbana-Champaign.

Rout, B. K., Singh, R. K., Choudhary, S. K. & Das, C. L. (2013). Development and application of nozzle clogging index to improve the castabilty in continuous slab casting. In: International Conference on Advances in refractories and Clean Steel Makinge, At Ranchi, India.

Schmidt, M., Russo, T. J. & Bederka, D. J. (1991). Steel shrouding and tundish flow control to improve cleanliness and reduce plugging. Iron and Steel Society, Tundish Metallurgy, 2, 3-12

Skiena, S. S. (2017). The data science design manual. New York: Springer. DOI: https://doi.org/10.1007/978-3-319-55444-0

Thomas, B. G. & Bai, H. (2011). Tundish nozzle clogging-application of computational models. In Steelmaking Conference Proceedings, 84, 895-912

Vannucci, M. & Colla, V. (2011). Novel classification method for sensitive problems and uneven datasets based on neural networks and fuzzy logic. Applied Soft Computing, 11(2), 2383-2390. DOI: https://doi.org/10.1016/j.asoc.2010.09.001

Wang, B., Mao, Z. & Huang, K. (2018). A prediction and outlier detection scheme of molten steel temperature in ladle furnace. Chemical Engineering Research and Design, 138, 229-247.

Wang, R., Li, H., Guerra, F., Cathcart, C. & Chattopadhyay, K. (2021). Development of quantitative indices and machine learning-based predictive models for SEN clogging. In: AISTech 2021, proceedings of the iron & steel technology conference, 1892-190.

Wu, X., Jin, H., Ye, X., Wang, J., Lei, Z., Liu, Y. & Guo, Y. (2021). Multiscale Convolutional and Recurrent Neural Network for Quality Prediction of Continuous Casting Slabs. Processes, 9(1), 33.

Yildirim, O., Baloglu, U. B., Tan, R. S., Ciaccio, E. J., Acharya, U. R. (2019) A new approach for arrhythmia classification using deep coded features and LSTM networks. Computer methods and programs in biomedicine, 176, 121-133.

How to Cite
Diniz, A. P. M., Ciarelli, P. M., Salles, E. O. T., & Coco, K. F. (2022). Long short-term memory neural networks for clogging detection in the submerged entry nozzle. Decision Making: Applications in Management and Engineering, 5(1), 154-168. https://doi.org/10.31181/dmame0313052022d