back Back to all Seminars

Modeling and control of feedforward uncertain nonlinear systems subject to hysterisis inputs

Start Date: November 19, 2014 3:30 PM
End Date: November 19, 2014 3:30 PM

​​Professor Salim Ibrir​

King Fahd University of Petroleum and Minerals

Abstract: Hysteresis phenomena are ubiquitous in mechanics, electric drives, aerodynamics, ferromagnetic materials, ferroelectric materials, as well as in the deformation of some materials such shape-memory alloys. Hysteresis has been also identified in many other fields, including economics and biology. The interest in hysteresis phenomena begun in the early of 1980's. Since then, a great deal of interest has been continuously growing and many contributions to modeling and control of hysteresis have been appeared. Nowadays, there exists a set of static and dynamic models to describe the behaviors of hysteretic phenomena like Prandtl-Ishlinskii, Preisach, Dahl, Bouc-Wen, Duhem, and discontinuous models. When the hard input is associated to any system dynamics the overall system becomes a non-smooth system. Unfortunately, in those situations, the input cannot be expressed in affine manner to cancel any kind of system nonlinearities. State transformation to some normal forms becomes not feasible. Besides these inconveniences, the uncompensated input nonlinearities limit considerably the performance of the overall system if the feedback has not been conceived to overcome the input nonlinearity and the uncertainty influence. In this seminar a general framework is proposed to stabilize a class of feedforward nonlinear systems subject to hysteresis inputs. It is assumed that the system dynamics may contain constant or time-varying uncertainties. The control design consists in building an adaptive nonlinear feedback with only two adaptive parameters. It is shown that the nonlinear-parameterized control action achieves a practical stability of the overall system with arbitrary small bound of the system states. This practical stability is guaranteed under the assumption that the upper bounds of the system and the input uncertainties are known. Additionally, it is shown that the dynamic state-dependent controller gain is issued from the solution of an Algebraic Riccati Equation that has an explicit solution whatever the form of the system nonlinearities. The employed feedback has some interesting features like the robustness against system and input uncertainties, the design simplicity, and the applicability to other types of hysteretic systems. Real-time simulations of the pendulum-cart system are shown to demonstrate the efficacy of the control design and its straightforwardness in controlling an under-actuated non-minimum phase hysteretic system whose dynamics is not naturally given in feedforward form.
Bio: Dr. Salim Ibrir received his B. Eng. degree from Blida Institute of Aeronautics, Algeria, in 1991, his M.Sc. from INSA de Lyon, France, in 1994 and his Ph.D. degree from Paris-11 University, in 2000. From 1999 to 2000, he was a research associate (ATER) in the department of Physics of Paris-11 University. Dr. Ibrir held a 3-years post-doctoral position in Concordia University and many short research visiting positions in diverse North American universities before joining The University of Trinidad and Tobago as Associate Professor. From 2011 to 2013 Dr. Ibrir was with the department of Electrical and Computer Engineering of the University of The West Indies, Saint Augustine Campus. In September 2013, Dr. Ibrir joined the electrical engineering department of King Fahd University of Petroleum and Minerals, Saudi Arabia. His current research interests are in the areas of nonlinear control and estimation, adaptive control of non-smooth nonlinear systems, fuel-cell systems, robust system theory and applications, time delay systems, hybrid systems, convex optimization, intelligent and applied controls, ill-posed problems in estimation, hybrid systems and Aero-Servo-Elasticity.

More Information:

​Contact: Professor Meriem Laleg (​)