Skip to Main Content
This study introduces a multivariable adaptive decoupling control strategy using multiple models and neural networks for a class of discrete-time nonlinear dynamic systems. The adaptive decoupling switching control strategy based on generalised predictive control is composed of a linear adaptive decoupling controller, a neural-network-based nonlinear adaptive decoupling controller and a switching mechanism. First, the linear adaptive decoupling controller is designed to ensure the boundedness of the input and output signals of the closed-loop system. Second, the neural-network-based nonlinear adaptive decoupling controller is developed to improve the transient performance of the closed-loop system. Third, the stability and convergence of the closed-loop system are achieved simultaneously by using multiple-models-based switching mechanism. Simulation studies are provided for a numerical example and a forced-circulation evaporation process of an alumina production system so as to demonstrate the effectiveness of the proposed method.