Hao Yan

Three-phase dual-active-bridge (3p-DAB) dc–dc converters find widespread application in high-power conversion systems due to their high power density, operational efficiency, and inherent galvanic isolation. However, parameter imbalances in the three-phase transformer of a 3p-DAB can lead to unbalanced phase currents, which may result in unbalanced thermal or current stress on the power devices and reduce the lifetime of both the devices and the transformer, thereby decreasing the reliability of the system. This article analyzes the unbalanced phase current behavior in 3p-DAB converters and proposes a phase current balancing control method using a single bus current sensor. Compared with multisensor control methods, the proposed approach is more cost-effective and enables a more compact converter design. The control strategy reconstructs the phase current trajectories in the $\alpha \beta $ frame based on bus current information, then estimates the root-mean-square (rms) values of each phase current to provide feedback for the closed-loop balance control system. The feasibility of the proposed method is verified through simulations and an experimental platform based on a TI TMS320F28379D microcontroller.

Transfer learning is a promising technique used to deal with the issue of domain transfer in fault diagnosis of three phase PWM-VSI. However, there is limited research on partial transfer for VSI, where the data category of the source domain is the complete health conditions and contains the data category of the target domain. Consequently, this paper designs a new weighted mechanism to solve the open-circuit fault diagnosis for VSI in three-phase permanent-magnet synchronous motor under partial transfer scenarios. The distribution difference between the two domains is decreased by minimizing the Wasserstein distance between the weighted source domain data and target domain data, while the weights are obtained in the process. The outlier data in the source domain are assigned with small weights to diminish their effect on the classification results, thereby enhancing the performance of the model for partial transfer diagnostic. Simultaneously, the transferable recognition network is trained by the weighted cross-entropy loss on the source domain data and the conditional entropy loss on the target domain data. A training strategy is proposed to optimize weights and network parameters alternately where one aspect is fixed while the other is updated iteratively. The proposed method is verified by experiments.

The dual three-phase permanent magnet synchronous motor (DTP-PMSM) is characterized by its low-voltage high-power output, high reliability, and high control flexibility, rendering it widely applicable. Conventional control systems for DTP-PMSMs typically necessitate a minimum of four current sensors for closed-loop control, contributing to increased control system costs and volume. This article proposes a method tailored for a zero-phase-shift DTP-PMSM, utilizing two Hall current sensors to sample multiple branch currents and reconstruct phase currents. The article provides a detailed exposition of the principles behind six-phase current reconstruction using zero voltage vector sampling in a DTP-PMSM, along with the corresponding drive system structure. An analysis of the sampling delay, reconstruction dead zone, and inherent reconstruction errors associated with this method is presented. Simulation and experimental results ultimately validate the effectiveness of the proposed control strategy.

As a power electronic converter with bidirectional energy transmission, the Dual Active Bridge (DAB) has a broad application prospect in the electrical power distribution system of hybrid‐electric flight. For the DAB application in the electrical power distribution system, high efficiency and high reliability are envisaged due to the mission‐critical nature of the application. A new pulse width modulation optimization control strategy is proposed, which is based on the feedback temperature fluctuations of the GaN (Gallium Nitride) power transistors, to improve the lifetime of the whole circuit. The complete topology and simulations of the DAB converter are carried out in PLECS. Detailed simulation results manifest the precise lifetime of the GaN power transistors and verify the accuracy of the new optimization control strategy. Meanwhile, Insulated Gate Bipolar Translator is set as a control group to compare the lifetime of conventional silicon‐based power devices with the GaN power transistor. Hardware In Loop platform test is built to further validate the proposed method.

To reduce the cost and size of the power converters, techniques of reconstructing three-phase current through a single current sensor have been employed in permanent-magnet synchronous motors. In existing studies, the reconstruction precision is largely affected by current reconstruction dead zones in the space vector pulsewidth modulation (PWM) plane, which requires additional algorithms to compensateeither by modifying PWM modulation strategy or by phase-shifting of PWM signal. In this article, a phase current reconstruction method for three-phase inverter with the modification of a twelve sector space vector pulsewidth modulation (TS-SVPWM) is proposed. By arranging a single current sensor on a circuit branch between two power switches, the proposed method can reconstruct the three-phase current of the voltage source inverter and avoid the dead zones located at the sector boundary region. This method can be realized not only with a single Hall-effect current sensor but also with a single shunt resistor, which can expand its application. The proposed method can be a fault-tolerance solution when there are phase current sensor faults. Beside, the three-phase current can be reconstructed when there are open-circuit faults in the power switches. Compared with previous methods, the proposed TS-SVPWM can reduce the phase current harmonics. The effectiveness of the proposed method is validated by experiments.

The aim of this article is an attempt to provide a more credible control mode or a fault-tolerant control strategy for permanent magnet synchronous motor (PMSM) drives when an unexpected failure occurs to the current or position sensor. Considering that, a position sensorless control method that uses only one current sensor is proposed. Different from a traditional position sensorless control scheme that utilizes additional current sensors to obtain information of multiple phase currents, only a single current sensor (SCS) arranged on one of the current branches is included among the whole control circuit. Three-phase currents are reconstructed by sequentially sampling the SCS under specific voltage vectors, and directly employed for motor position estimation. A nonlinear position observer established according to the gradient descent is adopted. Focusing on the special current-acquisition pattern in this article, a step-size optimization method based on golden section search is proposed to minimize position estimation error. The experimental results reveal that the method in this article is comparable to the traditional full-current-sensor method in terms of position estimation, which confirms the fault-tolerant capability as one certain sensor of the control structure malfunctions.

In single current sensor (SCS) methods, different phase currents should be acquired sequentially under the duration time of specific voltage vectors within each switching period. The sampling asynchronization caused due to sampling delay will inevitably result in the increase of phase current reconstruction error and deteriorate the control performance. According to the fundamental expression formula of three-phase currents in permanent magnet synchronous motor drives, the sampling-delay error (SDE) occurring in the traditional SCS techniques is analyzed in this article. It is demonstrated that the error oscillates at the stator electrical frequency, with the amplitude affected by both the switching frequency and motor speed. Correspondingly, a novel compensation method based on a δ–γ reference frame is presented. There is a simple proportional relationship between δ- and γ-axis error component after filtering the ac component by a notch filter. As a result, the SDE can be easily eliminated without extra complex compensation algorithms introduced. The prominent feature of the proposed method is that it is developed without necessity of motor parameters when compared with previous compensation techniques, making it a more robust solution. Several experimental results are provided to verify the effectiveness of the proposed method.

In this article, we present an improved current reconstruction technique with current prediction for a permanent magnet synchronous motor drive system. In some conventional single current sensor techniques, current fluctuations occur at the sector-switching period (SSP) due to current holding between two adjacent sampling intervals, which can cause current distortion and introduce many current harmonics, especially when the motor speed is very high. Focusing on that, a current prediction method is proposed in this article to prevent the current fluctuation happening at SSP. The SSP is divided into two independent parts for the accurate calculation of the output voltage vector and the actual three-phase currents at SSP are determined on account of both the predicted dq -axis currents and the measurable ones. As a result, compared with the previous current-holding strategy, the current fluctuations can be fundamentally eliminated under the proposed method with less current harmonics introduced, further improving the control performance of the whole system. The accuracy and practicability of the proposed method are verified by the experimental results.

Electric variable speed drives (VSDs) based on two VSDs connected to a multiphase machine are an attractive solution to replace high-power mechanic and hydraulic systems in many sectors of industry and transportation because they present high performance with reduced cost, volume and weight. Among the causes which affect the reliability of dual VSDs, the common-mode current flowing through the machine bearing is an important issue. This paper faces the mitigation of the common-mode current by reducing the common-mode voltage (CMV) generated by the operation of a dual VSD. The CMV reduction is carried out without introducing any extra device and/or passive filtering method. This CMV reduction is performed by applying a specific phase-displacement between the modulation strategies of each single inverter drive. The proposed technique has been evaluated in a down scaled experimental setup in order to test its effectiveness.