1: \begin{abstract}

2: This paper presents a dual-level model predictive control (MPC) scheme for two-timescale dynamical systems subject to input and state constraints, with the scope to enforce closed-loop separable dynamics. A novel dual-level MPC (i.e., D-MPC) algorithm is initially presented. At the high level of the control structure, a stabilizing MPC regulator  minimizes the deviation of the output and its setpoint at a slow time scale. A shrinking horizon MPC is designed at the low level to refine the computed control actions in the basic time scale so as to generate satisfactory short-term transient of the output associated with the ``fast" dynamics.

3: To further improve the closed-loop control performance, an \emph{incremental} D-MPC algorithm is also proposed, via introducing at the high level of the D-MPC an integral action and an explicit design of the ``fast" output reference. 

4: The proposed algorithms are not only suitable for systems characterized by different dynamics, but also capable of imposing  separable closed-loop performance for dynamics that are non-separable and strongly coupled. The recursive feasibility and convergence properties of the D-MPC and \emph{incremental} D-MPC closed-loop control systems are proven. The simulation results concerning the use of the proposed approaches for the control of a Boiler Turbine (BT) system, {\color{black}including the comparisons with a decentralized PID controller and a multirate MPC,} are reported to show the effectiveness of the proposed algorithms in imposing closed-loop separable dynamics and the advantages in generating satisfactory control performance.

5: \end{abstract}