Adaptive Neural Asymptotic Tracking Control for PMSM Systems Under Current Constraints and Unknown Dynamics
Jianyi Zhang, Wei Ren, Jingjie Li, Xi‐Ming Sun
Abstract
In this brief, an adaptive neural asymptotic tracking control (ANATC) scheme is developed for permanent magnet synchronous motor (PMSM) systems under current constraints and unknown dynamics. More specifically, a system transformation strategy is first introduced to handle the current constraint, and is applied to embed the constraint condition into the transformed system through nonlinear mapping. In this way, it is unnecessary to study the current constraint independently, thus facilitating the design of the control scheme. In addition, the neural networks (NNs) are applied to approximate the unknown dynamics, and only require updating one parameter online. Based on the system transformation strategy and the NN approximator, an ANATC scheme is developed to establish the asymptotic tracking performance without steady-state error. Finally, hardware experiments are presented to illustrate the performance of the ANATC scheme.