Journal of Power Electronics最新文献

Pole-changing permanent magnet (PCPM) motors have advantages over traditional permanent magnet synchronous motors (PMSM) in high-speed applications. However, changing the pole number of a motor requires reconstructing the stator winding. This increases both the cost and complexity. To solve this problem, a new type of single-winding PCPM motor with a pole-slot combination satisfying the rule of 2p_a + 2p_b = Z (the sum of the number of poles of the motor before and after the pole-changing is equal to the number of slots) is proposed in this paper. The multi-pole ratio pole-changing of the PCPM motor can be realized by combining the single-winding structure with another single-winding structure that satisfies the p_a + p_b = Z (the sum of the number of pole pairs of the motor before and after the pole-changing is equal to the number of slots) rule. Through this method, there is no need to reconstruct the stator winding when a pole is changed, Only by changing the phase sequence and direction of the armature current, the switching process is rapid, and the online pole-changing can be realized. This single-winding pole-changing method can flexibly realize the conversion of three pole number modes without increasing the additional winding cost and complexity. Finally, the rationality of the proposed pole-changing method is verified by finite element analysis (FEA).

Using gallium nitride (GaN) transistors in motor drive systems presents challenges because of their high dv/dt characteristics. Compared with silicon (Si) transistors, GaN transistors cause a higher leakage current in the bearing. This phenomenon results in a bearing corrosion problem in the motor, consequently shortening its lifespan. In response to these challenges, an output filter placed between an inverter and the motor has been studied recently. This paper compares the leakage current in a permanent magnet synchronous motor (PMSM) drive system driven by a GaN inverter without and with a sine wave filter. By modeling the impedance of the leakage current path, this paper confirms the possibility of reducing leakage current with the application of the sine wave filter. Experimental results show that using the sine wave filter reduces the leakage current by 32.24% at full load.

The carrier modulated sinusoidal pulsewidth modulation scheme has been acknowledged for its fundamental fortification and harmonic reduction in DC–AC conversion. This study is focused on how to get a variable output voltage with the least harmonics by changing the amplitude of the carrier wave vertically instead of modulating the reference wave (conventional SPWM). This paper proposes four control schemes for achieving the maximum fundamental voltage in a three phase voltage source inverter. The four schemes are single edge carrier shifted with (i) fixed reference and (ii) variable reference, and double edge carrier shifted with (iii) fixed reference and (iv) variable reference. In addition, the reverse modulation index and mutual modulation index are introduced as new control variables for examination. These schemes are simulated with a three phase inverter and the performance is compared with conventional sinusoidal pulsewidth modulation using MATLAB R2021. The amplitude of the fundamental voltage and the corresponding THD are reported. Digital implementations of the proposed schemes are premeditated by using VHDL language and ModelSim. The validity of the proposed schemes is experimentally investigated through a laboratory prototype of a three phase inverter module incorporated with an FPGA Spartan 6 device. This prototype is tested with a three phase induction motor as the load. The fundamental voltage and THD for the existing and proposed schemes are reported.

To improve the adaptability of sensorless control in the full-speed range of a dual three-phase permanent magnet synchronous motor (DTP-PMSM), an improved sensorless control strategy based on a variable-gain super twisting algorithm sliding mode observer (VGSTA-SMO) along with harmonic component compensation has been introduced in this paper. First, the super twisting algorithm sliding mode observer (STA-SMO) is introduced to reduce the inherent chattering phenomenon of traditional SMOs. Then, the variable gain of this STA-SMO is designed to improve the observation adaptability at different rotor speeds. Meanwhile, considering that the harmonic components, especially at low rotor speeds, in the stator current can affect the observation accuracy of this VGSTA-SMO, a least mean squares adaptive notch filter (LMS-ANF) is employed to implement online harmonic compensation. Finally, comparative experiments with different methods in the full-speed range are conducted and compared to verify the effectiveness of the proposed novel sensorless control strategy whose observation deviation is about 2% even when the rotor speed is as low as 100 rpm.

This paper presents the design of a 35 kW isolated fuel-cell DC–DC converter (FDC) with a hybrid switch structure. By utilizing a circuit structure and pulse-width modulation (PWM) method that enables the use of insulated-gate bipolar transistors (IGBTs), the price is reduced without a significant efficiency degradation even at frequencies of tens of kHz when compared to using all silicon carbide metal–oxide–semiconductor field-effect transistors (SiC-MOSFETs). The proposed converter is based on a PWM resonant converter, which has good switching characteristics and low-voltage stresses on the switching devices. However, it suffers from high cost owing to heavy current stresses. This problem can be solved using a hybrid switch structure that adopts IGBTs in the leading-leg switches. Considering the wide variations in the input and output voltages, the design of the main parameters, the selection of power semiconductors considering the switching characteristics, and the frequency selection method are presented. The feasibility of the proposed FDC design is verified using a prototype implemented with an output power of 35 kW (input voltage = 330–610 V and output voltage = 450–850 V) and a switching frequency of 40 kHz.

The design, modeling, fabrication, and characterization of three types of printed circuit board (PCB)-integrated solenoid inductors using air core and two embedded magnetic cores are demonstrated in this paper. The fabrication applies a double-side PCB process to form the inductor windings. A cavity for accommodating the magnetic cores is created in the middle of the PCB. Two types of magnetic cores fabricated using in-house developed proprietary processes are applied as magnetic cores for the inductors. One of these cores is made out of FeSiAl magnetic powder, while the other is made out of laminated NiFe thin films. The size of the inductor is 5 mm × 3 mm × 1.56 mm. The quality factor of the FeSiAl powder core inductor reaches its peak value of 36.6 at 40 MHz. The NiFe multilayer thin-film core inductor obtains the highest inductance (22.48 nH) at 10 MHz but has a lower quality factor compared with the two other inductors. The tested saturation current of both magnetic core inductors is greater than 3 A. The inductors are also tested in a Buck converter switching at 10 MHz with an input voltage of 4 V, output voltage of 1 V, and load current of 1 A. The FeSiAl powder core inductor has the lowest loss of 115 mW, thereby suggesting that embedding the pre-made powder cores or multilayer thin-film cores is a good option for manufacturing PCB-integrated inductors. The powder core approach tends to yield an excellent high frequency performance, while the multilayer thin-film core option allows the integration of the magnetic thin-film process into the PCB fabrication flow to reduce costs and improve reliability for volume production.

With the widespread application of EMI filters, a method to improve power density has become a main concern in switch mode power supplies. Many previous works have been done for improving the EMI filtering performance of chokes. This article investigates an integrated choke structure that can minimize the leakage magnetic flux, enlarge the DM inductance, and improve the DM noise suppression performance. Compared to discrete EMI filter inductors, the proposed stacked choke scheme can integrate CM and DM chokes into toroidal cores by applying a combination of different types of toroidal cores. The proposed inductor can exhibit better DC magnetic saturation characteristic. Finite Element Method (FEM) simulations, impedance measurement, insertion loss and conducted EMI noise spectrums (150 kHz–30 MHz) are used to validate the accuracy of the theoretical analysis and the proposed equivalent magnetic reluctance models.

The traditional torque sharing function (TSF) strategy can lead to large torque ripple and copper consumption in switched reluctance motors (SRMs) due to the limitation of the voltage in the phase change interval. In this paper, an improvement scheme is proposed to address this problem. First, a segmented non-linear correction TSF (SN-TSF) is proposed. Then with the torque ripple and copper consumption of the SRM as optimization objectives, the proposed SN-TSF values are optimized using the velocity control particle swarm optimization (VCPSO) algorithm based on the particle velocity–boundary relationship. The optimal parameters are used to obtain the optimal SN-TSF curves with different loading torques and speeds. Experimental results show that the proposed SN-TSF strategy reduces the torque ripple by 40.17% and 66.5%, and the peak phase currents by 5.058 A and 5.371 A under two different operating conditions, which demonstrates the effectiveness of this strategy.

When a long cable is included in power conversion systems, it causes adverse effects, such as voltage spikes and ringing at load terminals. These nonideal voltages can break the insulation of electric machines, such as transformers and motors, and reduce their lifespan. To estimate such voltage characteristics, cable impedance should be modeled on the basis of the cable length. In this paper, two cable impedance models, a multiphysics model and an equivalent circuit model, are introduced. The multiphysics model using Ansys Q3D Extractor is suggested in consideration of the structure, material, and length of a practical cable. Meanwhile, the equivalent circuit model can be quickly utilized to examine voltage spikes and frequency. The accuracy of the proposed models is verified through simulation and the experimental results based on a motor drive system equipped with 30 and 100-m cables.

FCSMPC is a classical converter predictive control algorithm whose control performance is affected by the prediction error of the prediction model. In classical predictive control theory, the feedback correction mechanism is used to compensate for such prediction error. However, when this strategy is directly applied to the FCSMPC algorithm, the prediction error cannot be easily calculated. To address the prediction error compensation problem of FCSMPC, this paper proposes a prediction error compensation method based on neural network. A neural network prediction model is also constructed based on the timing characteristics of prediction error signals. The prediction error of this prediction model at the next moment is calculated by the designed neural network model, and then the output of the prediction model is compensated at the current moment. To improve the anti-interference performance of FCSMPC, the MRSVD algorithm is used to filter the prediction error sample data and the neural networks are trained by these sample data. The adaptability of the prediction error calculation is further improved by combining offline training with the online calculation of the neural network. A simulation model of the proposed method is then constructed using MATLAB, and simulation results show that the control performance of the FCSMPC algorithm is improved and that the effectiveness and feasibility of the proposed method are verified.