Adaptive backstepping control for permanent magnet linear synchronous motor servo drive

This paper presents a position servo control approach for a permanent magnet linear synchronous motor. The non-linear motor dynamics is expressed in the backstepping control scheme on which a recursive designing procedure is carried out. Based on the desired motion trajectory, the magnetic thrust force is first calculated and then treated as the control objective for the next subsystems. The command voltages to stabilise the whole system are established concerning the electric properties of the magnetic windings. To overcome the impacts of system uncertainties, an adaptive neural network is exploited to estimate the uncertainty and provide necessary compensation in the control effort. Based on the Lyapunov functional analysis, the adaptive laws for online tuning the parameters of the neural networks are derived so that the precision of position servo control can be improved. Compared with the conventional current regulated control scheme, this investigation introduces a voltage-controlled pulse-width modulation with a complete theoretic base, including the mechanical and electrical dynamics. The effectiveness of the proposed approach is verified by the experimental results and a comparison study with a recent work developed in the robust fuzzy PI control scheme.