Position sensors are indispensable in robotic joint servo systems for acquiring mechanical positions, yet their installation inevitably occupies an axial space and increases system complexity, limiting their applicabi...Position sensors are indispensable in robotic joint servo systems for acquiring mechanical positions, yet their installation inevitably occupies an axial space and increases system complexity, limiting their applicability in compact robot design where spatial constraints and integration efficiency are critical. Sensorless control reduces mechanical and circuit complexity through hardware simplification, but inherently estimates only the electrical instead of mechanical rotor position information, thus remaining constrained in robot joint control applications. Based on the previously proposed dual-gap dualpole composite machine(DDCM), this paper systematically analyzes the causes of mechanical position estimation errors and proposes a correction method that utilizes a correction coefficient to reduce these errors and enhance estimation accuracy. Furthermore, this paper derives the applicability constraints of the proposed scheme, demonstrating that its requirements for electrical angle position errors are not stringent, thus enabling wide applicability in conventional sensorless control scenarios. The effectiveness of the proposed method is verified by conducting experiments on a 0.75 kW prototype.展开更多
Inductance asymmetry,which is brought by inherent asymmetric parameters,manufacture tolerance,winding fault,cables with unequal lengths,etc.,of permanent-magnet synchronous machines(PMSMs)can cause current harmonics a...Inductance asymmetry,which is brought by inherent asymmetric parameters,manufacture tolerance,winding fault,cables with unequal lengths,etc.,of permanent-magnet synchronous machines(PMSMs)can cause current harmonics and inaccurate position estimation.This paper proposes an enhanced fundamental model based sensorless control strategy for PMSMs with asymmetric inductances.The proportional-integral-resonant current regulator is introduced to reduce the second-order harmonics of currents,but there are still negative sequence components in the estimated back-electromotive forces(EMFs),which can cause the position estimated error.Differing from conventional methods in which negative sequences are filtered out before the phase-locked loop(PLL)module,the proposed method directly applies the estimated back-EMF with negative sequences as the reference input of PLL.An improved PLL with a bi-quad filter is proposed to attenuate the arising second harmonic position error and heighten the steady-state accuracy.Then,this position error is used for asymmetric inductance identification and its result is utilized to update the observer model.Furthermore,the dynamic performance is improved by the output limitation on the bi-quad filter as well as the implementation of a fast-locking technique in the PLL.The effectiveness of the proposed scheme is verified by experimental results.展开更多
基金supported in part by the National Natural Science Foundation of China under Grants 52277057 and U22A20217in part by the Shandong Youth Innovation Team under Grant 2022KJ150。
文摘Position sensors are indispensable in robotic joint servo systems for acquiring mechanical positions, yet their installation inevitably occupies an axial space and increases system complexity, limiting their applicability in compact robot design where spatial constraints and integration efficiency are critical. Sensorless control reduces mechanical and circuit complexity through hardware simplification, but inherently estimates only the electrical instead of mechanical rotor position information, thus remaining constrained in robot joint control applications. Based on the previously proposed dual-gap dualpole composite machine(DDCM), this paper systematically analyzes the causes of mechanical position estimation errors and proposes a correction method that utilizes a correction coefficient to reduce these errors and enhance estimation accuracy. Furthermore, this paper derives the applicability constraints of the proposed scheme, demonstrating that its requirements for electrical angle position errors are not stringent, thus enabling wide applicability in conventional sensorless control scenarios. The effectiveness of the proposed method is verified by conducting experiments on a 0.75 kW prototype.
基金supported in part by the National Key R&D Program of China under Grant 2019YFB1503700in part by the National Natural Science Foundation of China under Grant 51977191。
文摘Inductance asymmetry,which is brought by inherent asymmetric parameters,manufacture tolerance,winding fault,cables with unequal lengths,etc.,of permanent-magnet synchronous machines(PMSMs)can cause current harmonics and inaccurate position estimation.This paper proposes an enhanced fundamental model based sensorless control strategy for PMSMs with asymmetric inductances.The proportional-integral-resonant current regulator is introduced to reduce the second-order harmonics of currents,but there are still negative sequence components in the estimated back-electromotive forces(EMFs),which can cause the position estimated error.Differing from conventional methods in which negative sequences are filtered out before the phase-locked loop(PLL)module,the proposed method directly applies the estimated back-EMF with negative sequences as the reference input of PLL.An improved PLL with a bi-quad filter is proposed to attenuate the arising second harmonic position error and heighten the steady-state accuracy.Then,this position error is used for asymmetric inductance identification and its result is utilized to update the observer model.Furthermore,the dynamic performance is improved by the output limitation on the bi-quad filter as well as the implementation of a fast-locking technique in the PLL.The effectiveness of the proposed scheme is verified by experimental results.
