Articles
All articles | Recent articles
Modelling the Wear of MC 98/15 Refractory Material in the Slag Spout Zone of an Oxygen Converter with the Use of Artificial Neural Networks
W. Zelik1, R. Lech2, S. Sado1, A. Labuz3, A. Lasota3, S. Lis3
1 Zaklady Magnezytowe "ROPCZYCE" S.A. Research and Development Centre of Ceramics Material
2 AGH in Cracow
3 ArcelorMittal Poland Dabrowa Gornicza Division
received February 17, 2020, received in revised form August 4, 2020, accepted August 9, 2020
Vol. 11, No. 2, Pages 81-90 DOI: 10.4416/JCST2020-00009
Abstract
In oxygen converters, the high temperature and aggressive alloys affect the refractory lining, which is most frequently made of MgO-C materials. The refractory lining of a converter is worn zonally along with the progress of the campaign. This paper describes a trial conducted to forecast the wear of the refractory material depending on the selected parameters of the metallurgical process. Based on the results of industrial measurements, a wear model of the magnesia-carbon refractory material for the converter slag spout zone was developed. For the construction of the model, multilayer artificial neural networks were used. The accuracy of the forecast of the refractory material wear in the wear classes achieved in the experiment equals 63.9 % with the network architecture consisting of the following numbers of the neurons 17: h(20,10,5) :10, where the parameter h(20,10,5) contains the number of neurons in individual hidden layers. The calculations were made i.a. in the R language and environment.
Download Full Article (PDF)
Keywords
Refractories, MgO-C, BOF converter, artificial neural network, statistical analysis
References
1 Lis, T.: Temporary methods of obtaining steel, Politechnika Slaska. 2nd Ed., Gliwice (2000).
2 Piech, J.: Refractory lining of furnaces and heat units in steel industry, Ropczyce (2012).
3 Nadachowski, F., Kloska, A.: Refractory wear processes. Cracow (1997).
4 Potschke, J., Deniet, T.: The corrosion of refractory castables, Interceram: Ref. Man., 5 – 18, (2005)
5 Vollmann, S.: Investigation of rotary slagging test by computational fluid dynamics calculations, Taikabutsu Overseas, 30, 10 – 18, (2010).
6 Harmuth, H., Vollmann, S., Rossler, R.: Corrosive wear of refractories in steel ladles – Part I: Fundamentals, UNITECR, art. 1-E-12. Kyoto (2011).
7 Vollmann, S., Harmuth, H., Kronthaler, A.: Corrosive wear of refractories in steel ladles – Part II: Influence of fluid flow, UNITECR Conference art. 1-E-13. Kyoto (2011).
8 Lech, R., Zelik, W.: Slag corrosion of MgO-C materials in attitude of dimensional analysis, Materiały Ceramiczne, 68, [3], 224 – 229, (2016).
9 Zelik W., Lech R.: Forecasting of the wear of selected refractory material of the MgO-C type in the slag zone of a steel ladle using dimensional analysis, Proc. UNITCR 15C9, (2019).
10 Hastie T., Tibshirani R., Friedman J.: The elements of statistical learning. Data mining, inference, and prediction, 2nd Ed., Springer Science + Business Media LLC, 309, 389 – 416, (2009).
11 Von König, G., Weidemüller, Ch., Pietzko, G.: Wear of refractory blast furnace linings, new findings and their application, (in German), Stahl Eisen, 91, [2], 63 – 69, (1971).
12 Chollet, F., Allaire, J.J.: Deep learning work with R environment and keras library, Helion (2019).
Copyright
Göller Verlag GmbH