Comparative Study on Harmonic Current Suppression of Dual three-phase PMSM Based on LMS Adaptive Linear Neuron and Resonant Controller

In this paper, the effects of the harmonic current suppression for dual three-phase permanent magnet synchronous motor (DTP-PMSM) by adaptive linear neuron (ALN) controller and resonant controller are systematically investigated. Although the ALN controller and the resonant controller have been extensively employed for the suppression of harmonic currents in the machine drive system, their suppression effect is not comparatively studied and the relationship between them is not revealed. The study shows that the transfer function of the least-mean-square (LMS)-based ALN controller in the frequency domain is equivalent to an ideal resonant controller, therefore, the ALN controller can be regarded as a resonant controller with ideal discretization method and has no resonant pole displacement and phase deviation to the desired resonant frequency. Compared with the conventional resonant controller with discretization methods, such as Tustin transformation and Tustin with pre-warping transformation, the ALN controller exhibits better harmonic suppression performance. Moreover, the design principle of the learning factor in the ALN controller is proposed by referring to the resonant gain in the conventional resonant controller. Finally, the effects of harmonic current suppression for DTP-PMSM by the ALN controller and the resonant controller are validated and compared by experiments.