This monograph focuses on the design of optimal reference governors using model predictive control (MPC) strategies. These MPC-based governors serve as a supervisory control layer that generates optimal trajectories for lower-level controllers such that the safety of the system is enforced while optimizing the overall performance of the closed-loop system.
The first part of the monograph introduces the concept of optimization-based reference governors, provides an overview of the fundamentals of convex optimization and MPC, and discusses a rigorous design procedure for MPC-based reference governors. The design procedure depends on the type of lower-level controller involved and four practical cases are covered:
PID lower-level controllers;
linear quadratic regulators;
relay-based controllers; and
cases where the lower-level controllers are themselves model predictive controllers.
For each case the authors provide a thorough theoretical derivation of the corresponding reference governor, followed by illustrative examples.
The second part of the book is devoted to practical aspects of MPC-based reference governor schemes. Experimental and simulation case studies from four applications are discussed in depth:
control of a power generation unit;
temperature control in buildings;
stabilization of objects in a magnetic field; and
vehicle convoy control.
Each chapter includes precise mathematical formulations of the corresponding MPC-based governor, reformulation of the control problem into an optimization problem, and a detailed presentation and comparison of results.
The case studies and practical considerations of constraints will help control engineers working in various industries in the use of MPC at the supervisory level. The detailed mathematical treatments will attract the attention of academic researchers interested in the applications of MPC.