Abstract
Carbon anode is one of the key components for the electrolytic production of aluminum. It is mainly composed of calcined petroleum coke, coal tar pitch, and recycled carbon materials. The impurities in the raw materials, which are mainly by-products of different industries, influence significantly the quality of anodes. Usually, no well-known mathematical relationship exists between the various physical and chemical properties of raw materials and the final anode properties. In such situations, the artificial neural network (ANN) methods can serve as a useful tool to predict anode properties. In this study, published data have been used to show the proficiency of different artificial neural networks using the MATLAB software. The average error between the predicted and experimental values is around 6 %. The artificial neural network was also used to identify the effect of impurities such as, vanadium, iron, sodium, and sulfur on the CO2 reactivity of anodes. ANN also showed the effect of pitch percentage and coke porosity on the CO2 reactivity of anodes. The effect of CO2 and air reactivities of coke on the CO2 reactivity of anode was also studied. The predictions were found to be in good agreement with the results of other studies in the literature.
Similar content being viewed by others
References
Berezin AI, Polaykov PV, Rodnov OO, Klykov VA, Krylov VL (2002) Improvement of green anodes quality on the basis of the neural network model of the carbon plant workshop. Light Metals 605–608
Chmelar J (2006) Size reduction and specification of granular petrol coke with respect to chemical and physical properties. Dissertation, Norwegian University of Science and Technology
Milewski J, Świrski K (2009) Modelling the SOFC behaviours by artificial neural network. Int J Hydrogen Energy 34(13):5546–5553
Biedler P, Banta L, Dai C, Love R, Tommey C, Berkow J (2002) Development of a state observer for an aluminum reduction cell. Light Metals 1091–1098
Meghlaoui A, Bui RT, Thibault J, Tikasz L, Santerre R (1998) Predictive control of aluminum electrolytic cells using neural networks. Metall Mater Trans B 29(5):1007–1019
Li J, Zhou H, Guo T (2010) Research on fault diagnosis method based on modified elman neural network. 2nd international conference on information science and engineering (ICISE), pp 1456–1459
Boadu KD, Omani FK (2010) Adaptive control of feed in the Hall–Héroult cell using a neural network. JOM 62(2):32–36
Bhattacharyay D, Kocaefe D, Kocaefe Y, Morrais B, Gagnon M (2013) Application of the artificial neural network (ANN) in predicting anode properties. Light Metals. doi:10.1002/9781118663189.ch201
Bhattacharyay D, Kocaefe D, Kocaefe Y, Morrais B (2015) Comparison of linear multivariable, partial least square regression, and artificial neural network analyses to study the effect of different parameters on anode properties. Light Metals. doi:10.1002/9781119093435.ch189
Parthiban T, Ravi R, Kalaiselvi N (2007) Exploration of artificial neural network [ANN] to predict the electrochemical characteristics of lithium-ion cells. Electrochim Acta 53(4):1877–1882
Pang G, Xu W, Zhai X, Zhou J (2004) Forecast and control of anode shape in electrochemical machining using neural network. Adv Neural Netw, Lecture Notes in Computer Science 3174:262–268
Shang GQ (2008) Application of BP neural network for predicting anode accuracy in ECM. Int Symp Form Sci Eng 2:428–432
Wang L, Apelian D, Makhlouf M, Huang W (2008) Predicting compositions and properties of aluminum die casting alloys using artificial neural network. Metall Sci Technol 26(1):16–21
Bhagavatula YS, Bhagavatula MT, Dhathathreyan KS (2011) Application of artificial neural network in performance prediction of PEM fuel cell. Int J Energy Res 36(13):1215–1225
Ray M, Ganguly S, Das M, Datta S, Bandyopadhyay NR, Hossain SM (2008) Artificial neural network (ANN)-based model for in situ prediction of porosity of nanostructured porous silicon. Mater Manuf Process 24(1):83–87
Panda DK, Bhoi RK (2005) Artificial neural network prediction of material removal rate in electro discharge machining. Mater Manuf Process 20(4):645–672
Fruhwirth RK, Filzwieser A (2007) Computational intelligence—neural computation in a copper refinery. EMC 2007:1–14
Piuleac CG, Rodrigo MA, Cañizares P, Curteanu S, Sáez C (2010) Ten steps modeling of electrolysis processes by using neural networks. Environ Model Softw 25(1):74–81
Vega JJ, Reynoso R (2008) Learning limits of an artificial neural network. F´ISICA S 54(1):22–29
Dursun E, Kilic O (2011) The Levenberg-Marquardt neural network model of the PEMFC’s MEA. In: 10th international conference on environment and electrical engineering (EEEIC), pp 1–4
Liu W (2012) Introduction to modeling biological cellular control systems, vol 6. Springer, Milan
Shmueli G, Patel NR, Bruce PC (2007) Data mining for business intelligence: concepts, techniques, and applications in Microsoft office excel withXLminer, 2nd edn. Wiley, New Jersey
Tosta MRJ, Inzunza EM, Delgado LA (2009) Boron salt inhibitors of air reactivity of prebaked carbon anodes: literature review and laboratory studies. Light Metals 1173–1176
Bensah YD, Foosnaes T (2010) The nature and effect of sulphur compounds on CO2 and air reactivity of petrol coke. ARPN J Eng Appl Sci 5(6):35–43
Eidet T, Sorlie M, Thonstad J (1997) Effects of sulphur, nickel and vanadium on the air and CO2 reactivity of cokes. Light Metals 436–437
Lee JM, Baker JJ, Rolle JG, Llerena R (1998) Characterization of green and calcined coke properties used for aluminum anode-grade carbon. ACS Div Fuel Chem 43(2):271–277
Perruchoud RC, Meier MW, Fischer WK (1996) Bath Impregnation of carbon anodes. Light Metals 673–679
Liu FQ, Liu YX, Mannweiler U, Perruchoud R (2006) Effect of coke properties and its blending recipe on performances of carbon anode for aluminium electrolysis. J Central South Univ Technol 13(6):647–652
Tran K (2011) Influence of raw material properties and heat treatment temperature on the reactivity of carbon anodes. PhD Thesis. The University of Queensland
Pietrzyk SS, Thonstad J (2012) Influence of the sulphur content in the anode carbon in aluminium electrolysis—a laboratory study. Light Metals. doi:10.1002/9781118359259.ch113
Chevarin F, Lemieux L, Ziegler D, Fafard M, Alamdari H (2015) Air and CO2 reactivity of carbon anode and its constituents: an attempt to understand dusting phenomenon. Light Metals. doi:10.1002/9781119093435.ch192
Fang N, Xue J, Li X, Lang G, Gao S, Xia B, Jiang J, Bao C (2015) Effects of coke types and calcining levels on the properties of bench-scale anodes. Light Metals. doi:10.1002/9781119093435.ch193
Bonnetain L, Hoynant G (1965) Les Carbones. Tome 2, Masson et Cie, chap. XVII, pp 298–306
Coste B, Schneider JP (2013) Influence of coke real density on anode reactivity consequence on anode baking. Light Metals. doi:10.1002/9781118647745.ch11
Acknowledgments
The technical and financial support of Aluminerie Alouette Inc. as well as the financial support of the National Science and Engineering Research Council of Canada (NSERC), Développement économique Sept-Îles, the University of Québec at Chicoutimi (UQAC), and the Foundation of the University of Québec at Chicoutimi (FUQAC) are greatly appreciated.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bhattacharyay, D., Kocaefe, D., Kocaefe, Y. et al. An artificial neural network model for predicting the CO2 reactivity of carbon anodes used in the primary aluminum production. Neural Comput & Applic 28, 553–563 (2017). https://doi.org/10.1007/s00521-015-2093-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-015-2093-7