Page 195 - Distributed model predictive control for plant-wide systems

8

Local Cost Optimization Based

Distributed Predictive Control with

Constraints

8.1 Introduction
In the second part, the unconstrained distributed model predictive control (DMPC) is intro-
duced given the concept of each DPMC coordination strategy and helps the readers to com-
prehensively understand the essential characteristics of each kind of coordination strategies of
DMPC. There are three kinds of DMPC strategies presented in Part Two. Let us first briefly
review these three kinds of coordination strategies.

• The local cost optimization (LCO)-based DMPC, where each local controller minimizes its
own subsystem’s cost and uses the state prediction of the previous time instant to approxi-
mate the state sequence at the current time instant in computing the optimal solution. If the
iterative algorithm is employed, the Nash optimality of closed-loop system can be achieved.
• Cooperative-based DMPC, where each subsystem-based MPC optimizes the cost of overall
system to improve the global performance. While computing the optimal solution, it also
uses the state prediction of the previous time instant to approximate the state sequence at the
current time instant. This strategy could achieve a good global performance in some cases,
but it reduces the flexibility and increases the communication load. We refer it as global
cost optimization based DMPC here, and the Pareto optimality of the closed-loop system is
obtained by this method.
• Networked DMPC with information structure constraint. In an effort to achieve a trade-off
between the global performance of the entire system and the computational burden, an intu-
itively appealing strategy is provided in Chapter 7, where each subsystem-based MPC only
considers the cost of its own subsystem and those of the subsystems it directly impacts on.

The application areas of all these approaches are complementary. Each method possesses
its own strengths and weaknesses. The practitioner, using knowledge and experience, must
choose the control algorithm that is more appropriate for the problem at hand.

Distributed Model Predictive Control for Plant-Wide Systems, First Edition. Shaoyuan Li and Yi Zheng.
© 2015 John Wiley & Sons (Asia) Pte Ltd. Published 2015 by John Wiley & Sons (Asia) Pte Ltd.