Control of an irrigation branch canal using model predictive controller.

Sustainable use of the available water resource is an important challenge for food security. An irrigation canal system consists of a number of pools connected to each other by control structures. Unsteady flow Saint-Venant equations are used to model the flow in the canal system and these equations are solved using implicit finite difference methods. The canal system is represented as linear time invariant system and the resulting equations are the dynamic equations in state space form. These linearized equations are modelled in Matlab® environment. The control strategy adopted in this research is the downstream control method of operation and the control structures adopted are radial gates and modified radial gates. The unsteady flow model is applied to the chosen study area and the results obtained from the model are used to simulate the dynamics of the canal system using model predictive controller toolbox. The results show that any closed loop controller will work efficiently with the hydraulic model developed. [ABSTRACT FROM AUTHOR]