Three-Tank Process Level Control based on a Nonlinear Observer

Abstract

This paper investigates the design and the hardware implementation of the control level for the three-tank system in modeling error and unavoidable disturbances. The main objective is to design a new controller such that the closed-loop meets the two reference levels using the only measured output available in sampling instants. A filtered continuous–discrete high-gain observer is designed to estimate the missing state. The estimated, treated as the exact state, is then used in the combined flatness-PI feedback control law. The proposed controller is very efficient in simulation and in real-time implementation.

KEYWORDS:

• Filtered high-gain observer
• Flatness control
• Nonlinear continuous–discrete-time observer
• PID control
• Three-tank system

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Mondher Amor

Mondher Amor received his BS and MS degree in automatic and industrial computing from ENSIT, Tunis, Tunisia, in 2000 and 2017, respectively. Then he received a PhD degree in electrical engineering from the National Engineering School of Monastir in 2021. His research interests are focused on nonlinear control, nonlinear observer design and implementation, and fault-tolerant control.

Taoufik Ladhari

Taoufik Ladhari received the electrical engineering diploma in 2003 from INSAT (Tunis); in 2004, he received his DEA in automatic systems from UPS-LAAS Toulouse, France. He received a PhD degree in process engineering in 2007 from the National High School of Mines of Saint-Etienne, France. He is currently assistant professor at ENIM. His research interests are in the control and state estimation of nonlinear systems, optimization of bioprocess, and biomedical system. Email: [email protected]

Salim Hadj Said

Salim Hajd Said received his MS degree in automatic and signal processing and PhD degree in electrical engineering from National Engineering School of Tunis, in 2004 and 2009, respectively. He is currently an assistant professor of automatic control at the Preparatory Institute for Engineering Studies of Monastir, Tunisia. His research interests include state observation, predictive and adaptive control of linear and nonlinear systems. Email: [email protected]

Faouzi M’Sahli

Faouzi M’Sahli received his BS and MS degrees from ENSET, Tunis, Tunisia in 1987 and 1989, respectively. In 1995, he obtained his PhD degree in electrical engineering from ENIT, Tunisia. He is currently a professor of electrical engineering at National School of Engineers, Monastir, Tunisia. His research interests include modeling, identification, predictive and adaptive control of linear and nonlinear systems. He has published over 80 technical papers and co-author of a book Identification et commande numerique des procedes industriels, Technip Editions, Paris. Email: [email protected]