This article elaborates a novel online full parameter estimation method for permanent magnet synchronous motors (PMSMs). Usually, only two parameters can be estimated using voltage functions in the d-q frame due to the problem of rank deficiency. The α-β frame possesses the inherent advantage that full motor parameters can be estimated using voltage functions during steady and transient states. An algorithm based on recursive least squares in the α-β frame is proposed to estimate full motor parameters, including stator resistance, $d$ -axis and $q$ -axis inductances, and flux linkage simultaneously. Simulation and experimental results on the PMSM drive system are presented to verify the effectiveness of the proposed algorithm. Compared with the method in the d-q frame, the superiority of the proposed method in terms of faster convergence rate, less computational cost, and high accuracy is demonstrated.

In the control of permanent magnet synchronous motors, the active disturbance rejection control (ADRC) has attracted considerable attention due to its intrinsic ability for disturbance rejection. However, the existing ADRC methods have the problem of coupling the reference tracking performance and disturbance suppression performance. To simultaneously improve the current control's dynamic response speed and steady-state control accuracy, a two-degree-of-freedom ADRC current control method based on the improved extended state observer (IESO) is proposed in this article. First, the designed IESO cancels the error correction term in the observed current update law to realize the complete decoupling of the reference tracking performance and disturbance suppression performance. Under the decoupling structure, the observed disturbance update law of the IESO is designed as a proportional-integral-repetitive control structure to suppress periodic and aperiodic disturbances. Then, the system stability is proved by analyzing the distribution of characteristic roots, and a two-step parameter tuning guideline of the IESO is provided. The proposed method realizes the complete decoupling of the reference tracking performance and disturbance suppression performance, reducing the complexity of the controller parameter selection and facilitating the realization of the high-performance current control. Finally, the effectiveness of the proposed method is validated by experimental results.

A novel sensorless control strategy for the brushless dc motor (BLDCM) is proposed in this paper. The proposed strategy is realized by employing new flux-linkage functions and can be applied to drive the sensorless BLDCM in the high-speed and low-speed ranges with highly accurate and reliable commutation. Another attractive feature is that the jumping edges of the flux-linkage functions are utilized to determine the commutation points, so there is no threshold needed as compared with other sensorless control methods. In addition, the three-phase current control method is adopted, and terminal voltages in flux-linkage function expressions can be easily obtained by calculation. By this way, the sample delay and the influence of the floating phase voltage can be eliminated, which can improve the accuracy and reliability of the sensorless control strategy. The effectiveness of the proposed strategy is verified by experimental results.

In this article, a variable structure sliding mode observer (VS-SMO) for load torque observation is designed by adding a decoupling term to the structure of the traditional sliding mode observer. The VS-SMO can remove the coupling relationship between the parameter to be estimated and the derivative of the estimated speed error, as well as the chattering signal in the estimated load torque. Then, a variable structure extended sliding mode observer (VS-ESMO) is designed on this basis, with the moment of inertia used as an extension, to achieve simultaneous identification of the load torque and the moment of inertia. This article analyzes the estimation errors and the impact of measurement noise on estimation errors. The stability of the VS-ESMO is proved by backstepping control theory. An experiment is carried out on a 3.1 kW surface-mounted permanent magnet synchronous motor to validate the identification performance. The experimental results demonstrate that the VS-ESMO has a high estimation accuracy and fast convergence speed, and the identification value does not contain a chattering signal. Using the inertia and load torque identified by VS-ESMO for parameter tuning and compensation can improve system dynamic performance and anti-interference performance to the load.

Based on diode-assisted buck-boost inverter, this paper proposes a new commutation torque ripple suppression strategy for brushless dc motor (BLDCM). Four types of switching vectors are constructed, according to the working pattern of the diode-assisted inverter and the operation mode of the BLDCM. Moreover, the effects of switching vector combination on commutation torque ripple suppression and motor speed regulation are analyzed in the commutation and normal conduction periods, respectively. Based on this analysis, the duration of switching vectors within each modulation cycle is derived and the sequence of vectors is arranged at the same time in these two periods. The proposed method can effectively suppress the commutation torque ripple over the full speed range by unified switching vectors during the commutation period, without needing to switch control strategies according to the speed range. In addition, the increase of the voltage stress of switching devices in the inverter bridge can be avoided by designing the duration and sequence of switching vectors during the commutation and normal conduction periods. The effectiveness of the presented method is validated by the experimental results.

In braking operation, when the unipolar modulation pattern is adopted by brushless DC motor (BLDCM), there exists not only the commutation torque ripple, but also a problem concerning controllability of braking torque. This paper conducts a research into eight unipolar modulation patterns and proposes a smooth braking torque control strategy for BLDCM. First, the controllability principle of braking torque is theoretically analyzed under different modulation patterns. Then, this paper derives and compares the speed ranges of each pattern where the braking torque is controllable and the commutation torque ripple can be reduced. On this basis, according to the torque control performance of each modulation pattern in different speed ranges, an optimized combination modulation method is developed in full speed range. The proposed control strategy ensures that the braking torque is controllable and meanwhile the commutation torque ripple can be reduced effectively, thus obtaining good braking torque control performance. The effectiveness of the proposed strategy is verified by experimental results.