Neural network real-time IP position controller online design for permanent magnet linear synchronous motor

This paper presents a real-time IP position controller realized by a neural network for permanent magnet linear synchronous motor (PMLSM) servo system. In the paper, the proposed neural networks configuration is simple and reasonable and its weight has definite physical meaning and rapidly adjustable character in order to obtain real-time control. The mover mass, damping coefficient and disturbance force are estimated by the proposed estimator, which is composed of a recursive least-square (RLS) estimator and a disturbance observer. The observed disturbance force is fed forward, to increase the robustness of PMLSM drive system.