High-speed on/off valves (HSVs) are core components in digital hydraulics and are commonly driven by composite-voltage sequences (high/low/negative/zero) to realize millisecond-level switching. Under such high-frequency excitation, the soft magnetic materials (SMMs) in HSVs exhibit pronounced magnetic hysteresis and eddy-current losses, leading to non-negligible electromagnetic hysteresis in electrical-to-magnetic energy conversion that is not captured by existing quasi-static solenoid models. This paper develops a dynamic HSV modeling framework that explicitly accounts for magnetic hysteresis and eddy-current effects. A Jiles–Atherton (J–A) constitutive relation is formulated to describe domain pinning, while a magnetic-inductance term is introduced to represent eddy-current dynamics. Flux-path reluctances are partitioned by geometry and surrounding media, and the five J–A parameters are identified from experimentally measured SMM B–H characteristics. The resulting electric–magnetic–mechanical coupled model predicts electromagnetic hysteresis time during valve opening and closing. Analysis indicates that excessively high excitation voltages aggravate hysteresis, whereas pre-establishing an initial magnetic flux can mitigate it. A modified multi-voltage driving method is experimentally implemented, reducing the opening/closing electromagnetic hysteresis times from 0.56/0.40 ms to 0.12/0.29 ms, respectively. This work enables HSV drive design with reduced switching delay and improved repeatability, supporting higher-bandwidth and more energy-efficient digital hydraulics.
Electrical steel selection is a key lever in electric drive performance and resource efficiency, as material properties directly influence machine performance. Conventional design approaches typically define the machine geometry first and select the material afterward, potentially limiting the utilization of the available material properties. When machine geometry and other design parameters are however fixed, material redesign offers a practical opportunity to improve performance without requiring a complete machine redesign. This paper investigates the influence of six industrial non-oriented electrical steel grades with sheet thicknesses of 0.20 and 0.35 mm on the performance of an electric mobility application. A simulation study is conducted based on magnetic characterization, including magnetic loss modeling and magnetization modeling and is complemented by extensive material characterization of grain size, alloying content, electrical resistivity, and crystallographic texture to establish links between the measured material parameters and the simulated machine performance. The objective is not merely to compare the performance of individual steel grades, but to assess what conventional material classifications reveal—and where they fail to capture application-relevant behavior. The results show that drive-cycle performance depends strongly on the underlying material parameters and the resulting distribution of individual loss components. In particular, sheet thickness becomes increasingly important at high-speed operation, while grain size and alloying content provide additional levers for balancing magnetic performance and processing effort. The findings provide a basis for application-oriented material selection beyond nominal grade classifications, offering machine designers opportunities to identify performance improvements in existing designs and supporting more targeted material design and improved resource utilization in electric machine design.
To address the high computational cost associated with 3-D air-gap magnetic field analysis of permanent-magnet spherical motors, a fast calculation method combining 2-D finite-element slicing and spherical harmonic reconstruction is proposed. First, the 2-D magnetic vector potential problem is solved on several representative cross sections, from which air-gap magnetic field samples are extracted. The local magnetic field components are then transformed into a unified spherical coordinate system, and the spherical harmonic coefficients are identified using a weighted regularized least-squares method to reconstruct the 3-D air-gap magnetic field. Comparisons with the analytical method, the two-dimensional penalty-coupling method, and the 3-D finite element method (FEM) are conducted to investigate the effects of the number of slices, sampling points, and truncation order on the reconstruction accuracy. The results show that the proposed method achieves a relative error of 4.72% in the magnetic flux density magnitude, while requiring only approximately 2.43% of the computation time of the 3-D FEM, thereby providing a favorable tradeoff between magnetic field prediction accuracy and computational efficiency.
This paper proposes a 5-kW reduced-rare-earth hybrid-magnet flux-intensifying asymmetric spoke-type IPM (HFI-ASIPM) motor for electric vehicle traction. The central design strategy compensates for PM-torque loss from ferrite substitution by enhancing the reluctance torque, rather than primarily by increasing ferrite volume. A symmetric all-NdFeB FI-SIPM, an all-NdFeB asymmetric FI-ASIPM, and hybrid NdFeB-ferrite variants are evaluated as successive design stages. An extended equivalent magnetic-circuit (EMC) model quantifies the PM flux linkage produced by rare-earth (RE) and ferrite (FE), while frozen-permeability finite-element analysis (FEA) separates the PM and reluctance torque components. Magnet displacement and rotation are used to realize magnetic-axis shifting and enhance inverse saliency, followed by multi-objective genetic algorithm (MOGA) optimization of the hybrid rotor. The selected design reduces the N38SH PM area by 30% while maintaining an average torque of approximately 23 Nm with 10.8% torque ripple. The peak PM-torque component decreases from approximately 23 to 18 Nm, whereas the peak reluctance-torque component increases from approximately 2 to 6 Nm. The EMC-predicted total PM flux linkage differs from FEA by 2.87%. The peak rotor von Mises stress remains below 180 MPa at 7000 rpm, and no irreversible demagnetization is observed under the evaluated conditions. Prototype measurements of back EMF, torque-current, and torque-angle characteristics validate the predicted electromagnetic behavior.
Classical parallel plate capacitor (PPC) model, the most used analytical model for calculating the foil capacitance, does not account for the edge effect leading to severe underestimation of the static capacitance calculated for foil conductors having a high aspect ratio (e.g., relatively higher isolation distance to foil height ratio in a HV winding). To ameliorate this drawback, the present work first improves the accuracy of the traditional PPC model by accounting for the edge effect through a simple correction factor, derived using RA based on different aspect ratios of foil conductor and series-connected dielectrics. Comparison with 2-D FEM results shows that for wide aspect ratios 0.0167-2.083 of foil conductor, the proposed RA-based model improves the accuracy of static capacitance calculations by 48.2%-74.9% compared to the traditional PPC model. Second, this work further improves the modeling of dynamic capacitance in a multilayer/multidisc foil winding transformer by incorporating the conjoint effects of incremental turn length of coils and the coil-to-coil potential distributions based on different winding schemes. This is verified using 3-D finite element method results derived based on a 1.2 kW, 0.24/2.1 kV, 20 kHz transformers with wave/U and flyback/Z winding schemes. This validates that the proposed model can predict within 16.3% of 3-D FEM results, indicating a relative improvement of 43.7% with respect to the traditional PPC model. The proposed RA-based PPC model offers an improved yet computationally efficient analytical model for capacitance calculation of multilayer/multidisc foil winding transformers.
