Adaptive composite learning dynamic surface control for chaotic fractional-order permanent magnet synchronous motors

Abstract

This paper aims to address the tracking problem of uncertain fractional-order permanent magnet synchronous motors with parametric uncertainties. To guarantee the system stability and offset the effect of parametric uncertainties, an adaptive backstepping composite learning neural control scheme based on interval excitation is presented. Moreover, dynamic surface technique is exploited to overcome the technical limitation of “explosion of complexity” caused by standard backstepping framework. In virtue of stability analysis and illustrative simulation, it is confirmed that the proposed control scheme not only attenuates the tracking error as small as possible, but also achieves satisfactory parametric convergence with high estimation precision.