Industrial Kiln Multivariable Control: MNN and RBFNN Approaches
Artificial neural networks have been recognized as a valuable framework for nonlinear identification and control. In this paper we discuss and compare the use of two types of neural network arquitectures (1) MNN (Multilayer Neural Network) and (2) RBFNN (Radial Basis Function Neural Network) for modelling a second order nonlinear chemical process — a lime kiln in the pulp and paper industry. The simulation results showed that MNN performs better in this practical case. Therefore, it was used in an IMC (Internal Model Control) strategy. The neurocontroller was analysed with regards to performance and robustness against disturbances.
KeywordsRadial Basis Function Input Space Inverse Model Radial Basis Function Neural Network Radial Basis Function Network
Unable to display preview. Download preview PDF.
- Barto, A. G. in Neural Networks for Control. Miller III, W. T., Sutton, R. S., Werbos, P. J. (eds.). Cambridge MA 02142: The MIT Press 1990.Google Scholar
- Hunt, K. J., Sbarbaro, D., Zbikowski, R., Gawthrop, P. J.: Automatica, 28 6, 1083 (1992).Google Scholar
- White, D. A., Sofge, D. A. (eds): Handbook of Intelligent Control: Neural, Fuzzy, and Adaptive Approaches. New York: Van Nostrand Reinhold, 1992.Google Scholar
- Trentelman, H.L., Willems, J. C. (eds.): Essays on Control: Perspectives in the Theory and its Applications. Boston: Birkhäuser, 1993.Google Scholar
- Ribeiro, B., Correia in Energy Efficiency in Process Technology. A. D., Pilavachi, P. A., (eds.). London: Elsevier Science Publishers LTD, 1993.Google Scholar