J. Jarmulak, E.J.H. Kerckhoffs, L. Rothkrantz, “Universal approach to neural process control illustrated on a biomass growth model”, in: Artificial Intelligence in Agriculture: A Postprint Volume from the 2nd IFAC/IFIP/EurAgEng Workshop, A.J. Udink ten Cate, R. Martin-Clouaire, A.A. Dijkhuizen, and C. Lokhorst (Eds.), Elsevier Science, 1995, ISBN 0-08-042597-6; pp. 243-248? (pages are from the pre-print version).
- pdf – sorry, the layout is messed up a bit by WordPerfect X4 that had problems with a 13 year old file
Abstract: The article presents a control scheme, based on use of neural networks, which can be applied to a control of complex nonlinear systems. The emphasis is on producing methodology and appropriate software which could later be used to construct control systems in a standard way. As usual when neural networks are used many experiments are required before desired results are reached. Therefore, a NeuroControl Workbench has been constructed which allows one to construct numerous experiments with a minimum of effort and to find the most promising control architecture. This article presents some results of control experiments with a bioreactor simulation.
Keywords: adaptive control, artificial intelligence, identifiers, model-based control, neural networks, nonlinear systems, predictive control, software tools.