Tecnologia em Metalurgia, Materiais e Mineração
Tecnologia em Metalurgia, Materiais e Mineração
ABM Week 2018



Gabriela Araújo Gois, Leonard Carvalho, Moacir Andretti, Paulo dos Santos Assis

Downloads: 1
Views: 48


Este artigo propõe o uso de redes neurais artificiais para a previsão do consumo total de combustível no alto-forno. Para tanto, foi considerado um conjunto de dados contendo 270 registros, com 19 variáveis de entrada, com base nos dados históricos médios de operação de um alto-forno no período de janeiro/2016 a junho/2017, e verificou-se que o modelo apresentou bons resultados com coeficiente de correlação de 0,837, rmse de 11,8 (treino) e 12.7 (teste). O modelo final apresenta uma camada de entrada com 19 neurônios, camada intermediária com 19 neurônios e camada de saída com 1 neurônio.


Redes neurais artificiais; Consumo de combustível; Alto-forno.


This paper proposes the use of artificial neural networks for the prediction of fuel rate in the blast furnace. For this purpose, a dataset of 270 records, with 19 input variables were considered, based on the historical mean data of operation from the jan/2016 to jun/2017 of a blast furnace, and it was verified that model presented good results with correlation coefficient of 0.837 and rmse of 11.8 (train) and 12.7 (test). The model consisting of an input layer with 19 neurons, intermediate layer with 19 neurons and output layer with 1 neuron.


Artificial neural networks; Fuel rate; Blast furnace.


1 Zhou CQ. Final Technical Report: Project No. DE-FG36-07GO17041. Minimization of Blast Furnace Fuel Rate by Optimizing Burden and Gas Distribution. Washington, D.C.: U.S. Department of Energy, American Iron and Steel Institute; 2012.

2 Haapakangas, J. Coke properties in simulated blast furnace conditions. [tese]. Finlândia: University of Oulu; 2016.

3 Janjua R. Energy use in the Steel Industry. Brussels: World Steel Association; 2014. (Report Word Steel Association).

4 Babich AI, Gudenau HW, Mavrommatis KT, Froehling C, Formoso A, Cores A, et al. Choice of technological regimes of a blast furnace operation with injection of hot reducing gases. Revista de Metalurgia. 2002;38(4):288-305.

5 Zhang SJ, Yu AB, Zulli P, Wright B, Austin P. Numerical simulation of solids flow in a blast furnace. In: Commonwealth Scientific and Industrial Research Organisation. Proceedings of the 2nd International Conference on CFD in the Minerals and Process Industries; 1999 Dezembro 6-8; Melbourne, Austrália. Camberra, Austrália: CSIRO. 1999.

6 Haykin S. Redes neurais: princípios e prática. Trad. Paulo Martins Engel. 2nd ed. Porto Alegre: Bookman; 2001.

7 Bulsari A, Saxen H. Classification of blast furnace probe temperatures using neural networks. Steel Research. 1995;66(6):231-236.

8 Singh H, Sridhar N, Deo B. Artificial neural nets for prediction of silicon content of blast furnace hot metal. Steel Research. 1996;67(12): 521-527.

9 Ge AX. A neural network approach to the modeling of blast furnace [dissertation]. Massachusetts: Institute of Technology; 1999.

10 Pitambare DP. Survey on optimization of number of hidden layers in neural networks. International Journal of Advanced Research in Computer and Communication Engineering. 2016;5(11): xx-xx [[Q1: Q1]] [[Q1: Q1]].

11 Ngia LS, Sjoberg J. Efficient training of neural nets for nonlinear adaptive filtering using a recursive LevenbergMarquardt algorithm. IEEE Transactions on Signal Processing. 2000;48(7): xx-xx [[Q2: Q2]] [[Q2: Q2]].

12 Yu H, Wilamowski BM. Levenberg-marquardt training. Industrial Electronics Handbook. 2011;5(12):1-15.

13 Ripley BD. Pattern recognition and neural networks. Oxford: Cambridge University Press; 2007.

14 Chen C, Wang Y, Chang Y, Ricanek K. Sensitivity analysis with cross-validation for feature selection and manifold learning. In: Advances in Neural Networks. Proceedings of the 9th International Symposium on Neural Networks, ISNN 2012; 2012 July 11-14; Shenyang, China. Berlin, Heidelberg: Springer; 2012.

15 Faleiro RMR, Velloso CM, Castro LFA, Sampaio RS. Statistical modeling of charcoal consumption of blast furnaces based on historical data. Journal of Materials Research and Technology. 2013;2(4):303-307.

16 Gasparini VM, Castro LFA, Quintas ACB, Souza Moreira VE, Viana AO, Andrade DHB. Thermo-chemical model for blast furnace process control with the prediction of carbon consumption. Journal of Materials Research and Technology. 2017;6(3):220-225.

17 Zhao H, Liu R, Zhao Z, Fan C. Analysis of energy consumption prediction model based on genetic algorithm and wavelet neural network. In: Institute of Electrical and Electronic Engineers. Proceedings of the 3rd International Workshop on Intelligent Systems and Applications; 2011 May 28-29; Wuhan, China. Piscataway, New Jersey: IEEE; 2011. p. 1-4.

5d7fdb800e8825090cbbebff tmm Articles
Links & Downloads

Tecnol. Metal. Mater. Min.

Share this page
Page Sections