Robust Model Predictive Control Of Time-delay Systems

With the working conditions of modern production systems becoming more complex and their requrement increasing, It exists in control systems that uncertainty, delay and control the complex requirements of high performance features, which make control problems become more complex. Moreover, the uncertainty and delay which exist in the system may affect performance of control systems seriously, eventually may lead to unstablity for the system. On the other hand, due to the executing agency for the physical constraints and process conditions on the system, production safety and other considerations, control input and state variables of the system are bounded in the actual physical systems.Thus, it is in the process of carrying out control system analysis and control synthesis that is necessary to consider the actual system when the input constraints. As the model predictive control (MPC) have outstanding advantages of hard and soft constraints in handling system. Furthermore, the control input constrained uncertain time-delay system robust model predictive control methods are studied in the paper. The main works of this article can be summarized as follows:(1) By using of min-max model predictive control strategy combining the linear matrix inequality techniques, a delay-dependent robust model predictive control method with a time-delay compensation is given for the convex polyhedron uncertain time-delay system in which control input is bound. A time-delay compensation term is introduced into the model predictive controller in the method, which increases the degree of freedom of control, thereby enhancing the system control performance. Meanwhile, the feasibility analysis of the control algorithm and proof of closed-loop system robust stability are also given in the paper. Finally, simulation results verify that the proposed control method is effective.(2) By combining model predictive control with H_∞control method, a robust H_∞model predictive control method based on free weighting matrix is presented for a uncertain time-delay system with the control input bound and external disturbance. Receding horizon optimization theory is applied in the method to meet the constraints, while the design control algorithms can not only guarantee the feasibility of control algorithm and the stability of closed-loop system, but also guarantee the impact of an uncertain external disturbance resulted in closed-loop system is restrained at a certain level of inhibition.