Development Of A Neural Network Coupled With A Mathematical Model To Optimize The Secondary Cooling Zones In The Continuous Casting Of Steel Billets [acoplamento De Técnicas De Redes Neurais E Modelagem Matemática Da Solidificação Para Aotimização Do Resfriamento Secundário No Processo De Lingotamento Contínuo De Aços]

This work presents the insertion of an artificial intelligence technique (neural network) applied to continuous casting billet process. It was developed a neural network algorithm, which works linked to a mathematical model to determine the optimum values for the water flow rates in the secondary cooling zones. Firstly, the results obtained by the mathematical model were validated against experimental data (industrial) and a good agreement has been observed, in a second step these results were used by the neural network algorithm for determining optimum settings of water flow rates in the different sprays zones, which are conducive to the best quality of the solidified ingot. With the new set of values for flow rates, another simulation was carried out analyzing the solidification process during continuous casting in order to attain metallurgical restrictions concerning the reheating of strand surface temperature and metallurgical length.31853194Russel, S.J., Norvig, P., (1990) Artificial Intelligence: A Modern Approach, , Prentice Hall, Upper Saddle River, New JerseyBhadeshia, H.K.D.H., Neural Networks in Materials Science (1999) ISIJ International, 39 (10), pp. 966-979Kominami, H., Neural Network System for Breakout Prediction in Conditions Casting Process (1991) Nippon Steel Technical Report, (49). , AprWatanabe, T., Omura, K., Konishi, M., Watanabe, C., Furukawa, K., Mold Level Control in Continuous Casting by Neural Network Model (1999) ISIJ, 39 (10), pp. 1053-1060Zietsman, J.H., Kumar, S., Meech, J.A., Samarasekera, I.V., Brimacombe, J.K., Taper J.P.Design in Continuous Billet Casting Using Artificial Neural Networks (1998) Ironmaking and Steelmaking, 25 (6), pp. 476-483Mukherjee, A., Artificial Neural Networks for the Prediction of Mechanical Behavior of Metal Matrix Composites (1995) Acta Metallurgical, 43 (11), pp. 4083-4091. , FebSowka, E., Dikhoff, P.S., Harder, J., Munscher, F., Beirer, G., Breakout Avoidance System, BASYS, for Continuous Slab Casting (1999) Iron & Steel Engineer, pp. 30-35. , MayToledo, G.A., Lainez, J., Cirión, J.C., Model Optimization of Continuous Casting Steel Secondary Cooling (1993) Materials Science & Engineering A, 173, pp. 287-291Brimacombe, J.K., Baptista, L.A., (1980) Spray Cooling in the Continuous Casting of SteelThomas, B.G., Samarasekera, I.V., Brimacombe, J.K., Mathematical Model of the Thermal Processing of Steel Ingots: Part 1 Heat Flow Model (1987) Metallurgical Transactions, 18 B, pp. 119-130Brimacombe, J.K., Argarwal, P.K., Baptista, L.A., Hibbins, S., Prabhakar, B., Spray Cooling in the Continuous Casting of Steel (1980) Steelmaking Proceedings of the NOH-BOS Conference (ISS-AIME), pp. 109-123Santos, C.A., Spim Jr., J.A., Ierardi, M.C.F., Garcia, A., The Use of Artificial Intelligence Technique for the Optimization of Process Parameters Used in the Continuous Casting of Steel (2002) Applied Mathematical Modeling, 26 (11), pp. 1077-109