Real-time implementation of neural optimal control and state estimation for a linear induction motor

• Real-time implementation.
• Stability proof on the basis of Lyapunov theory.
• Real-life application.
• Experimental results.
• Application to a wide class of discrete-time nonlinear systems.

A reduced order state estimator based on recurrent high-order neural networks (RHONN) trained using an extended Kalman filter (EKF) is designed for the magnetic fluxes of a linear induction motor (LIM). The proposed state estimator does not need the mathematical model of the plant. This state estimator is employed to obtain the unmeasurable state variables of the LIM in order to use a state feedback nonlinear controller. A neural inverse optimal control is implemented to achieve trajectory tracking for a position reference. Real-time implementation results on a LIM prototype illustrate the applicability of the proposed scheme.