PhD Thesis

Modeling and Control in Open-Channel Irrigation Systems

Work default illustration

Supervisor/s

Information

  • Started: 01/09/2019
  • Finished: 10/12/2021

Description

This thesis reviews and discusses the existent modelling, estimation, and control strategies that have been reported in the OCIS field. Throughout the review process, it has been identified that due to the complexity of the fundamental models that describe the OCIS dynamics, multiple control-oriented modeling strategies have been reported. These modeling strategies have been classified and discussed, finding the need for control-oriented models that include the following:
• potential energy balances along the channels;
• the nonlinear hydraulic dynamics of the OCIS;
• useful to describe the dynamic behavior of interacting OCIS.
In addition, the need for estimation strategies designed from accurate models that include potential energy balances and nonlinear descriptions of the OCIS dynamics is identified. Also, it is established that the most common control objective in OCIS is to maintain a constant depth at the upstream or downstream end of the channels, and this control objective induces constant losses due to leaks and seepage. From the identified gaps, two control-oriented modeling strategies have been proposed, which include informa- tion about potential energy balances along the channels, nonlinear hydraulic dynamics to describe the OCIS, and are useful to describe the dynamic behavior of interacting OCIS. The modelling strategies have been validated, obtaining that by using approximated mass and potential energy balances (M&PEB) an accurate description of the OCIS dynamics can be reached. Therefore, the proposed control-oriented modelling strategy designed from approximated M&PEB is used in the development of deterministic and stochastic strategies for detection isolation and magnitude estimation of unknown flows such as leaks and seepage.

The estimation strategies have been designed from the enhancement of a moving horizon estimation (MHE) approach with the inclusion of detection and isolation mechanisms. In the control area, the development and implementation of conventional control strategies for OCIS are explored, and a nonlinear control strategy for interacting chan- nels is proposed. Finally, the proposed control-oriented modeling strategy designed from approximated M&PEB is used in the design of an efficient optimization-based control approach for OCIS. This is a nonlinear model predictive control strategy that takes advantage of the control-oriented modelling strategy accuracy, and available information of the users’ demands in the development of an optimisation problem with a finite horizon, which is solved at each time instant with the objective to minimise losses caused by leaks and seepage while the users demands and operational and hydraulic restrictions of the OCIS are satisfied.

The work is under the scope of the following projects:

  • MuYSCA: Modelado y control de sistemas de riego en canal abierto (web)