|WAHLIN, BRIAN - WEST CONSULTANTS, AZ
Submitted to: Journal of Irrigation and Drainage Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/28/2005
Publication Date: 6/1/2006
Citation: Wahlin, B.T., Clemmens, A.J. 2006. Automatic downstream water-level feedback control of branching canal networks: theory. Journal of Irrigation and Drainage Engineering. 132:(3):198-207
Interpretive Summary: Water is becoming a scarce resource, and irrigation districts are under pressure to use water more effectively. Computerized automatic control of irrigation canals has the potential of improving the operation efficiency of irrigation districts. Until now, all of the effort to automatically control irrigation canals has been applicable to a single, in-line canal. However, it is desirable to automatically control an entire branching canal network instead of just one canal. A methodology was developed by which existing automatic control algorithms can be applied to control branching canal networks. This methodology was tested on a simple example. These results will be of use to the Bureau of Reclamation, irrigation districts, and consultants.
Technical Abstract: Over the last 40 years researchers have made various efforts to develop automatic feedback controllers for irrigation canals. However, most of this work has concentrated on feedback controllers for single, in-line canals with no branches. In practice it would be desirable to automate an entire canal network and not just one of the branches. Because the branches in a network are hydraulically coupled with each other, a branching canal network cannot be controlled by designing separate controllers for each branch and then letting them run simultaneously. Changing the gate position in one pool on one branch can affect the water levels in pools on other branches. Because of this effect, the controllers designed for each of the in-line branches of the network will interfere with each other and potentially create instabilities in the branching canal network. Thus, the controller must be designed for the network as a whole and the branching flow dynamics must be explicitly taken into account during the controller design process. This paper presents simulation results using two different types of feedback controllers on a simple branching canal network. The first controller is a Linear Quadratic Regulator (LQR) and the second controller is Model Predictive Control (MPC). Both of these algorithms were able to effectively control a simple branching canal network.