Winding currents contain lots of harmonics if motors rotate at the high speed. This paper addresses two key challenges on the harmonic suppression: (1) using the multiple synchronous rotating frame transformations (MSRFTs) to get the primary signals, and (2) proposing the closed-loop detection system (CDS) to extract the harmonics accurately. Firstly, the mathematical model containing the harmonics was built. Secondly, the MSRFTs were introduced, and the high-precision detection method called the CDS was analyzed. The CDS has the higher harmonic detection accuracy than the open-loop method, and can enhance the disturbance rejection capability of the system. Thirdly, the parameter-independent compensation algorithm was proposed to enhance the robustness against the motor parameter variations, which was combined with the compensation calculated by the mathematical model to suppress the harmonics. The effectiveness of the proposed harmonic suppression strategy was verified by the experiments on the turbo molecular pump.

Surge is one of the main factors affecting the stability of the centrifugal compressor. This article investigates surge disturbance suppression of active magnetic bearing (AMB)-rotor systems in magnetically suspension centrifugal compressors. An active disturbance control method is adopted to make the rotor stably levitate under surge disturbance. First, the dynamic model of the magnetic bearing rotor is established and simplified into a single-channel model. Second, a third-order linear extended state observer (LESO) and a linear active disturbance rejection controller are designed. The performance analysis of the controller shows that the estimation error accumulation of the observer is bounded and the estimation errors decrease with the increase of the observer bandwidth. Furthermore, the variable differential parameter added in the differential link of the controller can flexibly adjust the system damping. A method of tuning parameters for linear active disturbance rejection control (LADRC) is proposed to suppress surge disturbance while satisfying the stability of the system. Finally, the results of the experiment demonstrate that the proposed scheme is effective in suppressing the large-scale vibration effect of the surge disturbance on the rotor, and the magnetically suspended centrifugal compressor (MSCC) can run stably with low amplitude under the modified surge and mild surge.

Accurate commutation point is essential to ensure the ideal torque performance and low power consumption for brushless DC (BLDC) motor. In order to achieve the precise sensorless commutation at a constant or the rapidly varying speed, a fast commutation error correction method is proposed in this article. First, it is pointed out that the controlled variable variation, the high rate of commutation error relative to the low speed and the nonlinearity of the correction loop are the main origins of commutation error when the speed varies rapidly. Then, a novel commutation error correction method is proposed through the following three aspects: first, constructing a controlled variable first utilized for commutation correction, which keeps constant at the rapidly varying speed; second, incorporating feedforward compensation for the phase shift caused by low-pass filter; third, using feedback linearization to reduce the nonlinearity of the correction loop. The proposed commutation error correction method not only corrects the commutation error fast at a constant speed, but also achieves precise commutation when the rotor speed changes rapidly. Moreover, the proposed method is based on the principle of closed-loop control, thus, it also has the advantage of anti-interference. At last, the experimental results are shown to validate the effectiveness of the proposed method.

Accurate rotor position is necessary for high-performance control of permanent magnet synchronous motor (PMSM). Compared with the methods using position sensors to detect rotor position, the sensorless methods can simplify the motor structure and enhance the reliability. To improve the estimation accuracy of sensorless methods, the position error correction method based on low-order harmonic suppression is proposed in this paper. The low-order harmonics generated by position error are analyzed in detail, which will cause extra motor losses. The position error signal can be reconstructed by low-order harmonics which are extracted by rotating frame transformation and low-pass filter. And the estimated position can be corrected using the reconstructed signal. The proposed method can not only improve the position estimation accuracy, but also suppress the low-order harmonics, which can reduce motor losses and enhance the motor performance. Based on space vector PWM (SVPWM), the proposed method is independent of motor parameters. It can be applied to the sensorless methods which can obtain voltages. The position error correction method has well dynamic performance and great robustness. The effectiveness is verified by the experiments on the turbo molecular pump.

Commutation error is a crucial factor that reduces the efficiency of brushless dc (BLdc) motors. This article presents a commutation point optimization method for a sensorless BLdc motor. First, the sources of commutation error under position sensorless control are analyzed, and a conclusion is drawn that the phase difference between the back electromotive force vector and phase current vector equals the total commutation error. Second, a fundamental waveform extractor is used to eliminate the influence of harmonics on the equivalent commutation error. Third, the phase difference of the fundamental waveforms is extracted by a combined phase-locked loop, and the real-time commutation error is obtained consequently. Finally, accurate commutation is achieved by compensating for the phase difference, and experimental results on a reaction flywheel prototype evaluate the presented method.