Suboptimal Nonlinear Predictive Control with Structured Neural Models

Ławryńczuk, Maciej

doi:10.1007/978-3-540-74695-9_65

Cited by 4 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process under consideration is a polymerisation reaction taking place in a jacketed continuous stirred tank reactor (Doyle et al, 1995) ] of the monomer is a disturbance. Because both steady-state and dynamic properties of the polymerisation reactor are nonlinear, it is frequently used as a benchmark for comparing nonlinear control strategies (Doyle et al, 1995;Ławryńczuk, 2007a;2007b;Tatjewski, 2007).…”

Section: Methodsmentioning

confidence: 99%

“…Among many structures of empirical models, neural networks (Haykin, 1999) can be effectively used online in different versions of MPC algorithms (Åkesson and Toivonen, 2006;Liu et al, 1998;Ławryńczuk, 2007a;2007b;Ławryńczuk and Tadej, 2008;Nørgaard et al, 2000;Parisini et al, 1998;Piche et al, 2000;Pottmann and Seborg, 1997;Tatjewski, 2007;Tatjewski and Ławryńczuk, 2006;Trajanoski and Wach, 1998;Yu and Gomm, 2003). This is because neural networks are universal approximators (Hornik et al" 1989), and hence they may be used to approximate nonlinear behaviour of 234 M. Ławryńczuk technological dynamic processes (Hussain, 1999;Nør-gaard et al, 2000).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Nonlinear Predictive Control Based on Structured Neural Models

Ławryńczuk¹

2009

International Journal of Applied Mathematics and Computer Science

Self Cite

View full text Add to dashboard Cite

This paper describes structured neural models and a computationally efficient (suboptimal) nonlinear Model Predictive Control (MPC) algorithm based on such models. The structured neural model has the ability to make future predictions of the process without being used recursively. Thanks to the nature of the model, the prediction error is not propagated. This is particularly important in the case of noise and underparameterisation. Structured models have much better long-range prediction accuracy than the corresponding classical Nonlinear Auto Regressive with eXternal input (NARX) models. The described suboptimal MPC algorithm needs solving on-line only a quadratic programming problem. Nevertheless, it gives closed-loop control performance similar to that obtained in fully-fledged nonlinear MPC, which hinges on online nonconvex optimisation. In order to demonstrate the advantages of structured models as well as the accuracy of the suboptimal MPC algorithm, a polymerisation reactor is studied.

show abstract