Generalized predictive control with adaptive model based on neural networks

  • Authors:

  • Petr Pivoňka;Petr Nepevný

  • Affiliations:

  • Department of Control and Instrumentation, Faculty of Electrical Engineering and Communication, Brno University of Technology, Brno, Czech Republic;Department of Control and Instrumentation, Faculty of Electrical Engineering and Communication, Brno University of Technology, Brno, Czech Republic

  • Venue:
  • NN'05 Proceedings of the 6th WSEAS international conference on Neural networks
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Generalized Predictive Control (GPC) is well known control algorithm. If we put together predictive strategy of GPC and Neural Network model, which is adaptive, then we obtain new controller with many advantages. Neural model is able to observe system changes and adapt itself, therefore regulator based on this model is adaptive. Algorithm was implemented in MATLAB-Simulink with aspect of future implementation to Programmable Logic Controller (PLC) B&R. It was tested on mathematical and physical models in soft-real-time realization. Predictive controller in comparison with classical PSD controller and it's advantages and disadvantages are shown.