1. Proceedings
  2. I3M
  3. 2021
  4. MAS
  5. 10.46354/i3m.2021.mas.014

Prediction of steel coils mechanical properties and microstructure by using deep learning and advanced data preprocessing techniques

  • a Marco Vannucci , 
  • b Valentina Colla,
  • c Claudio Mocci
  • d Frenk Van Den Berg,
  • e  Danique Fintelman,
  • f  Haibing Yang
  • a,b,c TeCIP Institute, Scuola Superiore Sant’Anna, via Moruzzi 1, 50127, Pisa, Italy
  • d,e,f Tata Steel, PO.Box 10.000, IJmuiden, 1970CA,The Netherlands
Cite as
Vannucci M., Colla V., Mocci C., Van Den Berg F., Fintelman D., Yang H. (2021). Prediction of steel coils mechanical properties and microstructure by using deep learning and advanced data preprocessing techniques. Proceedings of the 20th International Conference on Modeling & Applied Simulation (MAS 2021), pp. 111-119. DOI: https://doi.org/10.46354/i3m.2021.mas.014

Abstract

In the production of steel strips, the fulfilment of required product properties is a key factor to improve the company’s productivity and competitiveness. Product characteristics can be evaluated online throughout the length of the strip by means of non–destructive tests such as the IMPOC whose output signal is related to mechanical properties and their uniformity. In this work, a novel approach based on the use of deep–neural–networks and advanced analytics is used to develop a model for the prediction of IMPOC signal from process parameters. The model provides plant managers with an insight into the relationships among process conditions, product characteristics and mechanical properties in order to suitably set up process parameters to meet product requirements. In this work, different model architectures and data processing techniques are evaluated leading an overall prediction error lower than 5% that puts the basis for their integration into the plant.

References

  1. Ahmad, A. U. and Starkey, A. J. (2018). Application of feature selection methods for automated cluster ing analysis: a review on synthetic datasets. Neural Computing and Applications, 29(7):317–328. 
  2. Bärwald, M., Sommers, U., Biglari, M., Montagna, E., Beugeling, W., and Lhoest, A. (2015). Online quality monitoring in galvanizing lines. In METEC & ESTAD conference. 
  3. chine learning and data science tools with python. In Lecture Notes in Computer Science, volume 9605 LNCS, pages 435–480. 
  4. Branca, T. A., Fornai, B., Colla, V., Murri, M. M., Streppa, E., and Schröder, A. J. (2020). The chal lenge of digitalization in the steel sector. Metals, 20(2). 
  5. Brandenburger, J., Colla, V., Nastasi, G., Ferro, F., Schirm, C., and Melcher, J. (2016). Big data solution for quality monitoring and improvement on at steel production. IFAC-PapersOnLine, 49(20):55–60. 
    Cateni, S. and Colla, V. (2016a). The importance of variable selection for neural networks-based classi- cation in an industrial context. In Smart Innovation, Systems and Technologies, volume 54, pages 363–370. 
    Cateni, S. and Colla, V. (2016b). Variable selection for e cient design of machine learning-based models: E cient approaches for industrial applications. In Communications in Computer and Information Science, volume 629, pages 352–366. 
  6. Cateni, S., Colla, V., and Vannucci, M. (2009). Gen eral purpose input variables extraction: a genetic 
    algorithm based procedure give a gap. In 2009 Ninth International Conference on Intelligent Systems Design and Applications, pages 1278–1283. IEEE. 
    Cateni, S., Colla, V., and Vannucci, M. (2010). Vari able selection through genetic algorithms for clas si cation purposes. In Proceedings of the 10th IASTED International Conference on Arti cial Intelligence and Ap plications, AIA 2010, pages 6–11. 
  7. Cateni, S., Colla, V., and Vannucci, M. (2017). A fuzzy system for combining lter features selection meth ods. International Journal of Fuzzy Systems, 19(4):1168– 1180. 
  8. Colla, V., Cateni, S., Maddaloni, A., and Vignali, A. (2020). A modular machine-learning-based ap proach to improve tensile properties uniformity along hot dip galvanized steel strips for automotive 
    applications. Metals, 10(7):1–23. 
  9. Dai, H. and Zuo, X. (2020). Industry 4.0 boosts the steel industry into a new era of digitalization and smart technology. In AISTech - Iron and Steel Technology Conference Proceedings, volume 3, pages 1829–1841. 
    Herrmann, K. and Irle, M. (2009). Impoc: An online
  10. material properties measurement system. Flat-Rolled Steel Processes: Advanced Technologies, pages 265–269. Jolfaei, M. A., Shen, J., Smith, A., Zhou, L., and Davis, 
    C. L. (2018). Em sensor system for characterisation of advanced high strength strip steels. Studies in Applied Electromagnetics and Mechanics, 43:49–56. 
    Kebe, T., Dürr, W., Goureev, D., and Reimann, G. (2011). Einsatz des impoc-verfahrens zur bestim mung mechanisch-technologischer eigenschaften von stahlband. Proceedings of the DGZfP, pages 1–8. 
  11. Kingma, D. P. and Ba, J. L. (2015). Adam: A method for stochastic optimization. In 3rd International Confer ence on Learning Representations, ICLR 2015 - Conference Track Proceedings. 
  12. Lee, H. and Song, J. (2019). Introduction to convolu tional neural network using keras; an understanding from a statistician. Communications for Statistical Ap plications and Methods, 26(6):591–610. 
  13. Licen, S., Cozzutto, S., Angelucci, M., and Barbieri, P. (2018). Self-organizing map algorithm as a tool for analysis, visualization and interpretation of elec tronic nose high dimensional raw data. Chemical Engineering Transactions, 68:313–318. 
  14. Matino, I., Dettori, S., Colla, V., Weber, V., and Salame, S. (2019). Forecasting blast furnace gas production and demand through echo state neural network based models: Pave the way to o -gas optimized 
    management. Applied Energy, 253. 
  15. Mehmood, T., Liland, K. H., Snipen, L. G., and Sæbø, S. (2012). A review of variable selection methods in partial least squares regression. Chemometrics and Intelligent Laboratory Systems, 118:62–69. 
  16. Mocci, C., Vannucci, M., Colla, V., Van Den Berg, F., and Schmidt, R. (2018). Som analysis of impoc data in relation to process parameters in strip steel pro duction. In Proceedings of the 12th European Conference on Non-Destructive Testing ECNDT. 
  17. Nastasi, G., Mocci, C., Colla, V., Van-Den-Berg, F., Mulder, R., Beugeling, W., Kebe, T., and Steel, T. (2016). Chill marks e ect detection algorithm for plant impoc data. In Proceedings of the 19th World Conference on Non-destructive Testing WCNDT. 
  18. Pearson, K. (1895). Notes on regression and inheritance in the case of two parents. Proceedings of the Royal Society of London, 58:240–242. 
  19. Reddi, S. J., Kale, S., and Kumar, S. (2018). On the convergence of adam and beyond. In 6th Interna tional Conference on Learning Representations, ICLR 2018 - Conference Track Proceedings, . 
    Scheppe, D. (2009). Impoc - increasing production yield by online measuremenent of material proper 
    ties. SEAISI Quarterly (South East Asia Iron and Steel Institute), 38(3):41–46. 
  20. Van Den Berg, F., Fintelman, D., Yang, H., Mocci, C., Vannucci, M., and Colla, V. (2021). The use of ad vanced data analytics to monitor process-induced changes to the microstructure and mechanical prop erties in at steel strip. In Advances in Intelligent Systems and Computing, volume 1338, pages 78–91. 
    Springer Cham. 
  21. Van Den Berg, F., Kok, P., Yang, H., Aarnts, M., Beugeling, W., Meilland, P., Kebe, T., Stolzenberg, M., Krix, D., Peyton, A., et al. (2017). Product uni formity control (puc): How 15 european research institutes contribute to improve the in-line charac terisation of microstructure and mechanical proper ties in the manufacturing of steel strip. In In-line Measurement and Control for Metals Processing. IOM Communications. 
  22. Vannucci, M. and Colla, V. (2018). Self-organizing maps based undersampling for the classi cation of unbalanced datasets. In Proceedings of the Interna tional Joint Conference on Neural Networks. 
  23. Windt, K. (2019). The steel industry and plant engi neering: From digitalization to a digital eco-system. MPT Metallurgical Plant and Technology International, pages 28–34.