Model-based event-triggered predictive control for networked systems with communication delays compensation

This paper addresses the model-based event-triggered predictive control problem for networked control systems (NCSs). Firstly, we propose a discrete event-triggered transmission scheme on the sensor node by introducing a quadratic event-triggering function. Then, on the basis of the aforementioned scheme, a novel class of model-based event-triggered predictive control algorithms on the controller node is designed for compensating for the communication delays actively and achieving the desired control performance while using less network resources. Two cases, that is, the value of the communication delay of the first event-triggered state is less or bigger than the sampling period, are considered separately for certain NCSs, regardless of the communication delays of the subsequent event-triggered states. The codesign problems of the controller and event-triggering parameter for the two cases are discussed by using the linear matrix inequality approach and the (switching) Lyapunov functional method. Furthermore, we extended our results to the NCSs with systems uncertainties. Finally, a practical ball and beam system is studied numerically to demonstrate the compensation effect for the communication delays with the proposed novel model-based event-triggered predictive control scheme. Copyright (C) 2014 John Wiley & Sons, Ltd.