A distributed model predictive control approach for the control of irrigation canals

Abstract

Water networks are large-scale systems, consisting of many interacting components. They are currently typically operated by local decentralized controllers which receive setpoints from human operators. We discuss how communication among the local controllers can be included and in particular propose the use of distributed model predictive control for enabling the local controllers to determine setpoints autonomously using communication and coordination. We consider the control of a particular class of water networks, viz. irrigation canals. A simulation study on a 7-reach irrigation canal illustrates the potential of the proposed approach.