Page 305 - Distributed model predictive control for plant-wide systems

13

Operation Optimization of

Multitype Cooling Source System

Based on DMPC

13.1 Introduction
With the rapid growth of energy consumption, various kinds of cooling and storage devices
are widely used in air-conditioning system in giant buildings so as to realize energy saving and
emission reduction. The main characteristic of the subsystem is a diversity of cooling devices,
which generally include conventional electric refrigerators, ice storage system, ground-source
heat pump system, and combined cooling heating and power.The load features of these cooling
devices are not identical and the types of energy they consume are complex.
To improve energy utilization efficiency and reduce cost, there is an urgent need to provide a
new kind of economic scheduling strategy that makes full use of the time-of-use pricing, which
is currently widely used in China. The conventional electric refrigerators and the ice storage
tank can satisfy at least 90% of cooling load. The ice storage systems are widely applied
because of the autocontrol of ice storage and ice-making. Cooperating between the conven-
tional electric refrigerator and the ice storage tank is the key to reduce the cost and improve
the economic effect.
This chapter proposes a hierarchically distributed MPC strategy, which builds an economic
model of electric refrigerator and ice storage tank, and gets steady power states and optimal
set-point of each electric refrigerator and ice storage tanks under optimal conditions by using
mixed integer programming. Then it uses DMPC to make sure that each electric refrigerators
and ice storage tanks can track the upper optimal set-point as soon as possible with guarantee-
ing that the total power can track predictive load.

13.2 Structure of Joint Cooling System
The typical cooling system includes dual mode refrigerator, ice storage tank, exchanger, and
some other components. Normal electric refrigerator just has the air-condition mode. Dual

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.