PSO neural inverse optimal control for a linear induction motor

Abstract

In this paper, a discrete-time inverse optimal control is applied to a three-phase linear induction motor (LIM) in order to achieve trajectory tracking of a position reference. An online neural identifier, built using a recurrent high-order neural network (RHONN) trained with the Extended Kalman Filter (EKF), is employed in order to model the system. The control law calculates the input voltage signals which are inverse optimal in the sense that they minimize a cost functional without solving the Hamilton-Jacobi-Bellman (HJB) equation. Particle Swarm Optimization (PSO) algorithm is employed in order to improve identification and control performance. The applicability of the proposed control scheme is illustrated via simulations.