The single current sensor operation (SCSO) has attracted widespread attention in the motor drive system because of its ability to reduce cost and volume. However, when the SCSO is used with the conventional space vector puleswidth modulation (SVPWM) technique, the current reconstruction unobservable area in the sector boundary and low modulation area will degrade the accuracy of sampling data. In this article, a mixed SVPWM (MSVPWM) control scheme is proposed to eliminate the unobservable area. With the proposed scheme, the zero vectors ( ${{{\bm V}}_{{\rm {0}}}}$ and ${{{\bm V}}_{{\rm {7}}}}$ ) are replaced by the complementary effective vectors to achieve the same effect in the sector boundary and low modulation area, and the SVPWM is applied to the remaining space vector plane. These inserted effective vectors make the current observation window time prolonged to ensure that the dc-link current data can be accurately acquired in the entire space vector plane. Hence, the high-quality reconstruction of the phase current can be realized with less impact on the output of current ripples and reconstruction errors. Moreover, the relationship between the rotation speed and the increase in switching losses under different switching frequencies is quantitatively analyzed. The effectiveness of the MSVPWM reconstruction algorithm was verified by experimental results obtained from a permanent-magnet synchronous servo motor.

In this paper, the magnetic integration of resonant reset forward (RRF) converter and magnetic-thermal coupling analysis of integrated magnetic (IM) components are explored. The formulas of magnetic reluctance, winding turns, and magnetic flux density are obtained based on the equivalent magnetic circuit and minimum current pulsation scheme. Meanwhile, circuit-field coupling model of RRF converter and magnetic-thermal coupling model of IM components are built with the finite-element analysis (FEA) method. The simulation results illustrate that the core of the integrated RRF converter works in an unsaturated condition, the temperature rise of IM components is about 20 °C, and the integrated RRF converter has almost the same power efficiency as the conventional discrete RRF converter. Furthermore, hardware prototypes are fabricated and tested to verify the simulation accuracy and design validity. The conclusion of this paper will provide a reference for improving magnetic integration efficiency in the future design of RRF converter.

An accurate analytical model for litz-wire (LW) winding loss calculation of power electronic transformers (PETs) was deduced. First, the skin effect and proximity effect losses at bundle levels of a single straight LW were introduced. Then, the proximity loss of an air-core LW winding was studied by adopting the coordinate transformation technique (CTT) considering helical winding (HW) configurations and twisting configurations (TCs). Last, a 2-D analytical model for LW windings considering magnetic materials, LW configurations, and winding configurations (WCs) was proposed with the help of CTT, Dowell method, and Roth's method. This model is helpful for the design of LW windings considering more details such as the choices of WCs and LW configurations.

A numerical modeling method is proposed for the melting process of Titanium metals of Titanium alloys powder preparation used for 3D printing. The melting process simulation, which involves the tight coupling between electromagnetic field, thermal field and fluid flow as well as deformation associated during the melting process, is conducted by adopting the finite element method. A two-way coupling strategy is used to include the interactions between these fields by incorporating the material properties dependent on temperature and the coupling terms. In addition, heat radiation and phase change are also considered in this paper. The arbitrary Lagrangian–Eulerian formulation is exploited to model the deformation of Titanium metal during the melting process. The distribution of electromagnetic flux density, eddy current density, temperature, and fluid flow velocity at different time can be determined by utilizing this numerical method. In a word, the method proposed in this paper provides a general way to predict the melting process of electrode induction melting gas atomization (EIGA) dynamically, and it also could be used as a reference for the design and optimization of EIGA.

A numerical simulation, considering the two-way interactions of the electromagnetic, thermal, and fluid fields with deformation associated during a levitation melting process, is performed using the finite-element method. Material properties dependent on temperature are applied for the simulation. In addition, surface-to-ambient radiation and phase change are considered, and the arbitrary Lagrangian-Eulerian formulation is employed to model the deformation of a metal sample during the melting process. The oscillations of the metal sample at different times are also simulated. The oscillations, melting time, deformation, and distributions of the electromagnetic-thermal-fluid fields are obtained at different moments. A general way to dynamically predict the melting process of induction heating and its performance evaluations are provided in this paper.

Core losses of nanocrystalline alloy high frequency transformer (HFT) are calculated by adopting the finite element method with magnetic hysteresis effects considered. A normal Preisach model is used to represent the magnetic hysteresis during the numerical simulation. The normal Preisach model features with simple formulations and easy parameters identification. A dynamic circuit model of laminated magnetic cores for high frequency operation is used to calculate the core losses. Results are compared between the finite element method (FEM) and magnetic circuit method to prove the accuracy of finite element method, and it also provides a more accuracy way and general to compute the core loss of magnetic cores.