11 October 2019 in IEEE Sensors Journal
14 August 2018 in IEEE Transactions on Power Electronics
In magnetically suspended rotor (MSR) system of the flywheel, harmonic current produced mainly by mass imbalance and sensor runout would cause undesirable harmonic vibration. Resonant control (RSC) is one of the remarkable performance alternatives for rejecting a set of selected frequency components of the periodic disturbance in a control system. Based on the intrinsic relationship between the RSC and repetitive control which is deduced and clarified, this paper proposes a novel quasi-resonant control (QRSC) scheme for a baseline pre-stable MSR system to suppress harmonic current. By introducing a damping factor, it achieves a good balance between stability and dynamic performance. With phase hybrid compensation, the proposed QRSC provides system absolute stability, and expands the stability margin for the entire operational speed range. Further, as the domination of harmonic current, low-order frequency components are suppressed by parallel QRSCs. A stability criterion with rigorous proof for the plug-in QRSCs-controlled MSR system is addressed as well. Experimental results obtained with a magnetically suspended flywheel prototype corroborate the theoretical approach.
8 November 2019 in IEEE Sensors Journal
In magnetically suspended rotor (MSR) system supported by active magnetic bearings, the mass imbalance of the rotor and the sensor runout lead to multi-frequency vibration of the rotor system, which seriously affects the operation accuracy. In order to suppress the harmonic vibration in the MSR system, an improved second-order repetitive control (SORC) method based on parameter optimization is proposed. Firstly, the dynamic model of MSR system is established. The necessity of restraining the current stiffness force can be confirmed through the division and further analysis of the bearing force. Then, an improved SORC structure diagram is constructed, and the coupling relationship between control gain and weight factor is determined by system stability criterion. Next, for the rapidity, stability and steady-state accuracy in the convergence process, parameters are optimized separately. Then, stage integration can achieve a better balance of performance impacted by parameters. Finally, the simulation results show that the proposed scheme can effectively suppress the multi-frequency vibration of the MSR system.