Creator:
Tzafestas, Spyros G. ; Kyriannakis, Efthimios ; Anthopoulos, Yiannis
Contributor:
Korbicz, Józef - red. ; Uciński, Dariusz - red.
Title:
Model-based predictive control of large-scale systems using a neural estimator
Group publication title:
Subject and Keywords:
sterowanie ; sterowanie-teoria ; sztuczna inteligencja ; matematyka stosowana ; informatyka
Abstract:
An approach to the design of discrete-time decentralized control systems based on model-based predictive control (MBPC) and neural estimation is proposed. The class of interconnected large-scale systems (LSS) is considered, and a model is used at each control station to predict the corresponding subsystem output over a long time period. ; In the case of subsystems with m-step delay information patterns the non-locally available interaction trajectories are estimated by a multi-layer neural network trained on-line with a modified backpropagation-type algorithm. Representative computer simulation results are provided and compared for a set of illustrative examples. The proposed control scheme shows better performance than the other schemes, and also covers the important case where the subsystems' interactions are nonlinear.
Publisher:
Zielona Góra: Uniwersytet Zielonogórski
Date:
Resource Type:
Pages:
Source:
AMCS, volume 8, number 3 (1998) ; click here to follow the link