Wiener-Hopf optimal control of a hydraulic canal prototype with fractional order dynamics

Vicente Feliu-Batlle*, Daniel Feliu-Talegon, Andres San Millan Rodriguez, Raúl Rivas-Pérez

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)


This article addresses the control of a laboratory hydraulic canal prototype that has fractional order dynamics and a time delay. Controlling this prototype is relevant since its dynamics closely resembles the dynamics of real main irrigation canals. Moreover, the dynamics of hydraulic canals vary largely when the operation regime changes since they are strongly nonlinear systems. All this makes difficult to design adequate controllers.

The controller proposed in this article looks for a good time response to step commands. The design criterium for this controller is minimizing the integral performance index ISE. Then a new methodology to control fractional order processes with a time delay, based on the Wiener-Hopf control and the Padé approximation of the time delay, is developed. Moreover, in order to improve the robustness of the control system, a gain scheduling fractional order controller is proposed. Experiments show the adequate performance of the proposed controller.
Original languageEnglish
Pages (from-to)130-144
Number of pages14
JournalISA Transactions
Early online date7 Dec 2018
Publication statusPublished - 2018

Bibliographical note

The authors would like to acknowledge the support provided in part by the Consejería de Educación, Cultura y Deportes de la Junta de Comunidades de Castilla-La Mancha (Spain) with the Project POII-2014-014-P, in part by the European Social Fund and in part by the Spanish scholarship FPU14/02256 of the FPU Program of the Ministerio de Educación, Cultura y Deporte.


  • Wiener-Hop filter
  • Fractional order control
  • Hydraulic canal control
  • ISE optimization
  • Gain scheduling control


Dive into the research topics of 'Wiener-Hopf optimal control of a hydraulic canal prototype with fractional order dynamics'. Together they form a unique fingerprint.

Cite this