In the present work, a constructive learning algorithm is employed to design an optimal one-hidden layer neural network structure that best approximates a given mapping. The method determines not only the optimal number of hidden neurons but also the best activation function for each node. Here, the projection pursuit technique is applied in association with the optimization of the solvability condition, giving rise to a more efficient and accurate computational learning algorithm. As each activation function of a hidden neuron is optimally defined for every approximation problem, better rates of convergence are achieved. The proposed constructive learning algorithm was successfully applied to identify a large-scale multivariate process, providing a multivariable model that was able to describe the complex process dynamics, even in long-range horizon predictions. The resulting identification model is then considered as part of a model-based predictive control strategy, with high-quality performance in closed-loop experiments.3 24062411Andrietta, S.R., Maugeri, F., Optimum design of a continuous fermentation unit of an industrial plant for alcohol production (1994) Advances in Bioprocess Engineering, , Kluwer Academic PublishersBärmann, F., Biegler-König, F., On a Class of Efficient Learning Algorithms for Neural Networks (1992) Neural Network, 5 (1), pp. 139-144Battiti, R., First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method (1992) Neural Computation, 4 (2), pp. 141-166Camacho, E.F., Bordons, C., (1999) Model Predictive Control, , Springer-Verlag"Clarke, D.W., (1994) Advances in Model Based Predictive Control, , Oxford University PressEdgar, T.F., Himmelblau, D.M., (1988) Optimization of Chemical Processes, , McGraw-HillFriedman, J.H., Stuetzle, W., Projection Pursuit Regression (1981) Journal of the American Statistical Association (JASA), 76, pp. 817-823Geman, S., Bienenstock, E., Doursat, R., Neural Networks and the Bias/Variance Dilemma (1992) Neural Computation, 4 (1), pp. 1-58Haykin, S., (1999) Neural Networks: A Comprehensive Foundation - 2nd Edition, , Prentice HallHornik, K., Multilayer feedforward networks are universal approximators (1989) Neural Networks, 2 (5), pp. 359-366Huber, P.J., Projection Pursuit (1985) The Annal a of Statistics, 13 (2), pp. 435-475Hwang, J.-N., Lay, S.-R., Maechler, M., Martin, R.D., Schimert, J., Regression Modeling in Back-Propagation and Projection Pursuit Learning (1994) IEEE Transactions on Neural Networks, 5 (3), pp. 342-353Jones, L.K., On a conjecture of Huber concerning the convergence of projection pursuit regression (1987) The Annals of Statistics, 15, pp. 880-882Kosko, B., (1997) Fuzzy Engineering, , Prentice HallKosko, B., (1992) Neural Networks and Fuzzy Systems: A Dynamical Systems Approach to Machine Intelligence, , Prentice HallKwok, T.Y., Yeung, D.Y., Constructive Algorithms for Structure Learning in Feedforward Neural Networks for Regression Problems (1997) IEEE Trans. on Neural Networks, 8 (3), pp. 630-645Meleiro, L.A.C., (2002) Design and Applications of Linear, Neural, and Fuzzy Model-Based Controllers, , PhD Thesis (in Portuguese)Meleiro, L.A.C., Campello, R.J.G.B., Maciel Filho, R., Von Zuben, F.J., Identification of a Multivariate Fermentation Process Using Constructive Learning (2002) Proc. SBRN'2002 - VII Brazilian Symposium on Neural Networks, , IEEE Computer SocietyMeleiro, L.A.C., Maciel Filho, R., Campello, R.J.G.B., Amaral, W.C., Hierarchical Neural Fuzzy Models as a Tool for Process Identification: A Bioprocess Application (2001) Application of Neural Networks and Other Learning Technologies in Process Engineering, , Mujtaba, I. M. and Hussain, M. A. (Editors). Imperial College PressNg, G.W., (1997) Application of Neural Networks to Adaptive Control of Nonlinear Systems, , Research Studies Press Ltd., John Wiley & Sons IncRoosen, C.B., Hastie, T.J., Automatic Smoothing Spline Projection Pursuit (1994) Journal of Computational and Graphical Statistics, 3, pp. 235-248Soeterboek, R., (1992) Predictive Control - A Unified Approach, , Prentice HallVon Zuben, F.J., Netto, M.L.A., Unit-growing learning optimizing the solvability condition for model-free regression (1995) Proceedings of the IEEE International Conference on Neural Networks, 2, pp. 795-800Von Zuben, F.J., Netto, M.L.A., Projection Pursuit and the Solvability Condition Applied to Constructive Learning (1997) Proceedings of the International Joint Conference on Neural Networks, pp. 1062-1067. , Houston - USA, 2