Page 41 - Distributed model predictive control for plant-wide systems

Introduction 15

In the third part, Chapters 8–10, we focus on introducing the design methods of the
stabilizing DMPCs with constraints for the advanced readers. In Chapter 8, a design method
for the LCO-DMPC is developed, which is based on a dual mode scheme and is able to
handle input constraints. The feasibility and stability of this method are analyzed. In addition,
Chapter 9 introduces a stabilizing DMPC with constraints, in which each subsystem-based
MPC optimizes the cost of whole system. The consistency constraints, which limit the error
between the optimal input sequence calculated at the previous time instant, referred to as
the presumed inputs, and the optimal input sequence calculated at the current time instant to
within a prescribed bound, are designed and included in the optimization problem of each
local predictive control. The noniterative algorithm for the related fast process is designed
for solving each local predictive control. Both the feasibility and stability of this method
are analyzed. Chapter 10 provides a networked distributed predictive control with inputs
and information constraints, where each local predictive control optimizes not only its own
performance but also that of the systems it directly impacted on. The consistency and stability
constraints are designed to guarantee the recursive feasibility and the asymptotical stability
of the closed-loop system if the initial feasible solution exists.
In the last part, Chapters 11–13, three practical examples are given to illustrate how to imple-
ment the introduced DMPC into the industrial process. At first, the implementation of DMPC
to accelerated cooling processes in heavy plate steel mills is introduced. The control problem,
the system model, the system decomposition, the control strategy, and the performance of the
closed-loop system under the control of DMPC are provided. Then, different from the met-
allurgical process, one example of the speed train control with DMPC is presented and the
technical details are also provided. Finally, a load control of a high building in Shanghai with
multicooling resources system is studied, and the distributed predictive with a scheduling layer
is developed and detailed in Chapter 14.
In conclusion, this book tries to give systematic and latest distributed predictive control
technologies to the readers. We hope this book could help engineers to design their control
systems in their daily work or in their new projects. In addition, we believe that this book is fit
for the graduate students who are pursuing their master’s or doctor’s degree in control theory
and control engineering. We will be very pleased if this book could really do something for
you if you are interested in the control of a plant-wide system or predictive control.