Adaptive neural network control for permanent magnet synchronous motor with input nonlinearity

Abstract

This study aims to design a new adaptive control method for permanent magnet synchronous motors (PMSMs) using neural networks (NNs). In comparison to traditional motor backstepping control designs, this research introduces a command filtering strategy to effectively address the common issue of “complexity explosion” in traditional methods. Additionally, considering the potential input hysteresis nonlinearity in practical applications, we introduce a hysteresis inverse operator to mitigate its adverse effects on control. Furthermore, by employing a finite-time control strategy, we ensure rapid convergence of tracking errors within a finite time frame. Moreover, an adaptive NN controller is designed to approximate unknown continuous nonlinear functions of the system. Finally, the stability and convergence of the closed-loop system are analyzed using the direct Lyapunov method.