Boron nitride (BN), characterised by its wide band gap and robust dielectric properties, has emerged as a promising filler for polymer-based dielectric composites. Nevertheless, its practical implementation in polymer matrices has been hindered by poor dispersibility and inadequate interfacial adhesion. This study addresses these challenges by developing hyperbranched polymer (HBP)-functionalised BN (HBP-BN) and fabricating HBP-BN/cellulose nanofiber (CNF) composite films via vacuum-assisted self-assembly. The HBP-BN/CNF films demonstrate a low dielectric loss (tan delta < 0.016) and significantly improved volume resistivity (10(9) Omega & centerdot;m) and high thermal stability (55.36% residual mass at 800 degrees C). Notably, the surface flashover voltage is elevated from 13.82 to 15.61 kV, demonstrating superior insulation capability. All these results highlight the promising potential of HBP-BN/CNF composite films for use as advanced dielectric materials, providing a novel approach to develop eco-friendly high-performance materials for high-voltage insulation systems in next-generation electronics and electrical devices.

The linear induction generator, recognized for its highly efficient, structural simplicity, and operational reliability, has garnered significant attention across various domains, including magnetic levitation systems, rail transportation, and wave energy conversion. This study presents an investigation into the power output performance of linear induction generators employed in rail vehicles, utilizing Matlab/Simulink for systematic analysis. The simulation model integrates a double closed-loop vector control system to address electromechanical coupling. Outer speed loop uses a PI regulator to track the given speed, while the inner cur-rent loop implements field-oriented control. The simulation achieves the modeling of train operation under constant speed conditions. Simulation results indicate that the linear induction motor reliably transition into the power generation state with the power generation exhibiting a non-monotonic relationship with the slip |s|. The power generation capacity shows an initial increase followed by a decrease within the range of 0.02 ≤ |s| ≤ 0.09, reaching its maximum value at |s| = 0.05.

Electrodynamic suspension (EDS) has a broad application prospect in high-speed magnetic levitation transportation due to its advantages, such as strong self-stabilization ability and simple control. However, in high-speed application scenarios, flat-plate permanent magnet electric suspension has a high magnetic drag force and requires a large amount of weight of the installed permanent magnets. To ameliorate these problems, this article addresses the plate-type superconducting EDS system with a higher lift-to-drag ratio. First, the structure and principles of the superconducting EDS system are presented. Second, a 3-D analytical model of the electromagnetic force considering the transverse end effect is established by the magnetic vector potential equation and the boundary conditions at the end of the conductor plate. Among them, the source magnetic field of the superconducting magnet array required by the boundary conditions is solved by the coil discretization idea. Then, the reliability of the proposed analytical model is verified by comparing the computational results of the analytical model with the finite element simulation results. Finally, based on the 3-D analytical model, the suspension stiffness characteristics of the superconducting EDS system and the influence of specific parameters on the system's suspension performance are analyzed.

This paper presents a fast and robust online identifying method for current sensor faults in permanent magnet synchronous motor (PMSM) drive systems using a noise-enhanced two-dimensional convolutional neural network (NE-2DCNN). The model is optimized and quantized via TensorRT for lightweight edge deployment, significantly reducing computational complexity and enabling low-latency identification on Jetson platforms. To enhance noise robustness, a noise enhancement strategy is introduced by injecting Gaussian white noise within the specific SNR range of 19.02–25 dB during the dataset generation stage, while refined noise fault labels are designed in the output layer for low SNR conditions (<19.02 dB), enabling direct identification of noise-related faults. A structured input feature map integrating key electromechanical signals is constructed to improve the fault feature extraction capability of the CNN. The proposed method achieves a breakthrough by enabling unified identification of six types of current sensor faults—including broken-line, stuck, gain, zero-offset, saturation, and noise faults. It significantly surpasses the conventional capability of identifying only 3–4 fault types. Experimental results demonstrate that the approach offers rapid identifying speed, high accuracy, and strong noise robustness under diverse operating conditions.

During the excitation of high-temperature superconducting (HTS) magnets, traditional contact-type lead power supply must traverse from room temperature to cryogenic temperatures, generating significant Joule heat and forming a heat leakage source, which reduces the thermal stability and reliability of the magnet. To address this issue, this paper leverages the inherent zero-resistance/flux-flow resistance transition property of HTS materials and proposes a wireless power supply method for HTS magnets based on self-driven bridge synergy, with a focus on studying its through-wall excitation performance. A finite element model of a normal-conducting/superconducting hybrid electromagnetic coupling mechanism was constructed, and the coupling efficiency between the primary and secondary sides was improved by incorporating magnetic conductive materials. And an experimental platform for wireless power supply of a levitated HTS magnet was established, achieving an excitation current of 52 A across a 12 mm metal container wall, validating the effectiveness of the proposed power supply method. This work lays a theoretical foundation for advancing wireless power supply technologies for HTS magnets.

To overcome the challenges of cross-domain interference, few-shot learning limitations, and edge deployment constraints in conventional deep learning, this paper introduces a novel fault diagnosis framework for motor couplings based on frequency-domain physics-constrained diffusion and wavelet distillation. First, a single-step Markov forward diffusion mechanism, penalized by the acoustic frequency-band energy envelope, enhances feature robustness under low signal-to-noise ratios. Second, an end-to-end learnable wavelet convolutional layer is designed to adaptively capture non-stationary transient impulse signatures. Third, an asymmetric knowledge distillation architecture transfers these physics-informed features to a lightweight network, ensuring stable training while significantly lowering computational cost. Extensive experiments validate that the proposed approach maintains consistent classification performance under severely limited samples and deteriorated cross-domain scenarios. Operating with merely 0.03 M parameters and an inference latency of 0.76 ms, the designed model achieves an outstanding balance between cross-domain diagnostic accuracy and hardware efficiency, offering a practical and scalable solution for health monitoring in edge-based intelligent systems.

The 200 km/h permanent magnet direct-drive (PMDD) inboard bearing bogie represents a novel lightweight solution for railway vehicles. However, its direct-drive architecture can induce significant electromechanical coupled vibrations in the transmission hollow shaft, thereby posing risks to structural reliability. Therefore, investigating the electromechanical coupled dynamics of the bogie is crucial for the design of the transmission system. This study proposes a novel PMDD inboard bearing bogie structure and establishes an electromechanical coupled dynamic model that integrates vehicle rigid-flexible coupled dynamics with electrical traction drive control. The model is validated through bench and field tests. Good agreement is observed between the simulated and measured frequency characteristics of the stator current, torque ripple, and vibrations in the bench tests, while the frequency characteristics of electromechanical coupled vibrations are further verified by the field test. Subsequently, the electromechanical coupled responses are analyzed to elucidate the vibration mechanism. The results show that inverter modulation introduces (6n±1)-th harmonic currents, generating 6n-th harmonic torques. These harmonic torques act as the primary excitation sources, inducing significant torsional vibrations in the motor stator, rotor, and hollow shaft. Notably, vibration amplification occurs when the harmonic torque frequencies approach the torsional natural frequency of the hollow shaft (1050 Hz). Consequently, compared with conventional models, the proposed coupled model effectively captures the high-frequency vibration behavior induced by electrical excitation.

Electrodynamic suspension (EDS) technology employs superconducting magnets for passive levitation and propulsion in high-speed transportation, offering enhanced energy efficiency. A critical challenge, however, lies in the eddy current losses induced in the on-board magnet's cryogenic components by time-varying magnetic fields from ground-based propulsion coils during nonsynchronous operations like start-up or braking, which generate heat that compromises thermal stability. This article establishes a combined numerical and experimental framework to address this issue. First, a 3-D numerical model based on the A-formulation is developed to predict eddy current loss distribution within a novel, lightweight high-temperature superconducting (HTS) magnet system featuring an all-aluminum-alloy cryostat. Subsequently, a static equivalent measurement system is implemented, utilizing a calorimetric method to accurately measure losses under simulated dynamic conditions. The experimental results show good agreement with the simulations, validating the model's predictive accuracy. Key findings include the identification of a distinct loss peak near 100 Hz, attributed to the influence of penetration depth and skin effects in the aluminum structures, clarifying the frequency-dependent loss mechanism. Furthermore, a comparative analysis with a previous magnet generation demonstrates that the all-aluminum-alloy design achieves a significant weight reduction of over 30% and effectively suppresses eddy currents near the critical current leads, despite a moderate temperature rise in the coil cases. These insights provide valuable guidance for the design and optimization of thermally stable and lightweight HTS magnets, contributing to the reliability of next-generation EDS systems.

The track harmonic magnetic field acts as the excitation source of the linear generator and provides the power supply for superconducting electrodynamic suspension (EDS) trains. However, external disturbances can induce vibrations that deteriorate the operating conditions of the harmonic generator. Therefore, vibration suppression is essential to improve power supply conditions and operational safety. In this paper, the topology and equivalent circuit of a harmonic linear generator with onboard integrated coils are first introduced, in which power collection and electromagnetic damping functions are integrated. Afterwards, an electromagnetic model of the EDS system, including the harmonic linear generator, is established and verified to accurately characterize the electromagnetic interaction between the onboard integrated coils and null-flux coils. Finally, the dynamic characteristics of the suspension bogie are investigated, and a vertical-lateral synergistic vibration suppression strategy is carried out by additionally controlling d-axis current in the propulsion coils. The results show that the proposed strategy effectively makes up for the deficiency of existing control strategy by the onboard integrated coils, and the vibrations have been effectively suppressed under the excitations of vertical and lateral irregularities at different operating speeds. It lays a theoretical basis for the development of electromagnetic damping and contactless power supply technologies of superconducting EDS trains.

No-insulation (NI) high-temperature superconducting (HTS) magnets are promising for high-field applications due to their inherent self-protection capability. Nevertheless, the reliable detection of localized quench (thermal runaway) remains a crucial challenge for operational safety. This paper presents a non-invasive quench detection method utilizing a distributed Hall-sensor array, which monitors magnetic field variations resulting from current redistribution during a quench. The method is fundamentally grounded in the whole-turn current-sharing mechanism, elucidated by a circuit-based theoretical analysis. This mechanism explains that a local thermal event initially induces current to redistribute across adjacent turns, generating a spatially extended and azimuthally uniform magnetic field perturbation. Consequently, a limited number of external sensors can effectively capture the quench signal. Experimental studies on an NI HTS magnet under local thermal disturbances confirm the method's reliability. The method’s principal advantage is its non-intrusive nature, which preserves the magnet’s intrinsic integrity by avoiding the disturbances to turn‑to‑turn contact inherent to invasive sensors. Furthermore, simulation results delineate the operational stability limits: NI HTS magnets exhibit robust stability under high-temperature, low-current conditions due to insufficient loss generation, whereas under low-temperature, high-current operation, a high turn-to-turn contact resistivity (on the order of hundreds of µΩ·cm² or higher) significantly increases the risk of an irreversible quench. To ensure safety against local critical current degradation, a design current margin of 30% is recommended. The proposed detection strategy and comprehensive analysis provide critical insights for the design and protection of NI HTS magnets, facilitating their practical application.

In recent years, the silicon carbide (SiC) power devices have been widely applied in permanent magnet synchronous motor (PMSM) drives. However, their application in three-level inverters introduces critical design challenges, including low stray inductance busbar configurations and balanced thermal distribution. To overcome these challenges, this article presents a collaborative design of conductor plate and switching pattern for SiC three-level active neutral-point-clamped (ANPC) inverters. First, a four-layer laminated conductor plate is developed to minimize switching overshoot and reduce stray inductance. The conductor plate geometries are optimized for balanced thermal distribution. Then, to mitigate common-mode current and enhance dielectric reliability, a multimode space vector pulse width modulation (SVPWM) is proposed. Through the switching patterns optimization, the common-mode voltage and switching frequency are suppressed simultaneously. Additionally, the optimized switching pattern achieves precise neutral-point voltage balance by dynamically adjusting the dwell times of four trapezoidal-shaped vectors. The experimental verification with PMSM drive and resistive-inductive load demonstrates the efficiency and robustness of the designed configuration and the optimized modulation scheme.

This study focuses on the development of a high-efficiency computational model for predicting the magnetic field of high-temperature superconducting (HTS) magnets, which is crucial for the design and optimization of HTS-based devices. A fast-computational model for magnetic field was proposed, leveraging the feature extraction capabilities of the deep residual neural network. To validate the effectiveness and reliability of the proposed model, a prototyped HTS magnet system was employed for experimental verification. The comparison between the calculation results and the experimental data demonstrated a high degree of consistency, confirming the practical applicability of the model. Subsequently, an enhanced linear adaptive genetic algorithm was introduced for the optimization design of a 3 T magnetic resonance imaging magnet utilizing rare-Earth barium copper oxide superconducting tapes. The optimized magnet is composed of 60 double-pancake coils, operating at a cryogenic temperature of 30 K and featuring a coil inner diameter of 600 mm. An active shielding technique was adopted, which involves the strategic arrangement of additional coils to counteract the stray magnetic fields. Through this approach, the fringing field was effectively suppressed. In terms of field homogeneity, the magnet achieved 74 parts per million within a 250 mm diameter spherical volume.

The dynamo-type flux pump is a non-contact superconducting magnet excitation device based on the principle of electromagnetic induction, with its output characteristics significantly influenced by the number of rotor permanent magnets and the system structure. Our previous research indicates that as the number of rotor permanent magnets increases, the output voltage of the flux pump initially rises and then decreases, revealing the impact of magnetic field strength and distribution on the output performance. To further investigate this phenomenon, this study conducted simulations to analyze the effect of adding an iron yoke on the superconducting stator side on the DC output characteristics of the flux pump, and examined how changes in the number of rotor permanent magnets affect the system. Given the introduction of ferromagnetic materials, the H-A method was employed for modeling and simulation. The results show that adding an iron yoke significantly improves the DC output voltage of the flux pump. However, as the number of permanent magnets continues to increase, the magnetic field interference between adjacent magnets intensifies, causing the enhancement effect of the iron yoke on the output voltage to first increase and then weaken. Through an in-depth analysis of the magnetic field distribution, electric field distribution, and current density, this study reveals the mechanism by which the iron yoke influences the DC output voltage of the flux pump, providing theoretical support for further optimizing the performance of dynamo-type flux pumps.

Superconducting electrodynamic suspension (EDS) is one of the core technologies for achieving ultra-high-speed operation of rail transit systems. With advantages such as large levitation gaps, self-stabilizing levitation, and high operational efficiency, it has become a global research hotspot in the rail transit field. This paper first reviews the development history and experiment progress toward engineering deployment of superconducting EDS technologies in Japan and China, and compares achievements of the two countries in technology research and development, test line construction and speed breakthroughs. Next, it presents a systematic analysis of the structural characteristics, performance advantages and disadvantages, and engineering adaptability of three guideway-side topologies in superconducting EDS systems. From the perspective of theoretical analysis, the paper then reviews the research progress and optimization directions for three analytical methods: dynamic circuit method, scalar magnetic potential harmonic analysis method, and vector magnetic potential harmonic analysis method. Focusing on vehicle dynamics, existing research results on modeling methods, model accuracy optimization and vibration reduction control strategies are summarized respectively from three dimensions: single suspension frame, vehicle dynamical modeling and vehicle vibration control. Finally, the paper highlights outstanding challenges in existing superconducting electric levitation technologies in aspects such as electromagnetic coupling modeling, optimization of dynamic characteristics, and reliability in engineering applications, and discusses future development trends.

This paper proposes a robust deadbeat predictive current control (R-DPCC) strategy for permanent magnet synchronous motor (PMSM) drives, aimed at enhancing system tolerance to parameter mismatches and external disturbances. An extended state observer (ESO) is integrated into the control loop to estimate and compensate for lumped disturbances arising from model uncertainties and load variations, thereby improving current prediction accuracy. In addition, a recursive least squares algorithm with a forgetting factor is employed for online parameter identification. A coordinated parameter adaptive updating mechanism is also designed to ensure stable model updating. Experimental results demonstrate that the proposed method effectively reduces current harmonic distortion and torque ripple under parameter mismatch conditions, while maintaining accurate parameter estimation, which confirms its robustness and effectiveness.

No-insulation (NI) high-temperature superconducting (HTS) coils are widely used in many applications because of their higher current density, excellent mechanical properties, and self-protection capability. However, because their current transmission path is not unique, the modeling of NI coils becomes more complex compared to the insulated one. In this paper, we have established a field-circuit coupling model based on the J-A formulation and validated its accuracy during both transient charging/discharging and external magnetic field exposures. The results indicate that the proposed model is capable of effectively characterizing the overall behavior and local characteristics of the NI HTS coils. Specifically, the local properties are validated against AC loss measurements under external alternating magnetic fields. Based on the developed model, a small-scale pancake NI coil was fabricated to investigate the factors governing the AC losses in such small pancake coils under both radial and axial background magnetic fields with DC transport current. Results reveal that axial fields induce significantly higher AC losses in NI coils compared to radial fields. Notably, under radial magnetic fields, the loss patterns of NI coils exhibit negligible differences from those of insulated coils. These insights contribute to magnetic field configuration optimization and loss management in NI coil applications.

The permanent magnet linear braking (PMLB) has garnered significant attention for its contactless operation and high force density. However, conventional analytical models often exhibit limited accuracy due to oversimplified geometric assumptions and the inability to accommodate the multidegree-of-freedom (DOF) attitude variations of the mover. To address these limitations, this article presents a novel 3-D analytical model based on an equivalent circuit method. Specifically, the permanent magnet (PM) is equivalently represented as a rectangular current-carrying coil using the surface current method. The mutual inductance between this equivalent PM coil and the brake coil is calculated via the discrete Neumann formula. A dynamic equivalent circuit is then constructed to determine the induced currents in the brake coil, and the resulting braking force is evaluated using both the energy method and the method of images. The proposed model is thoroughly validated through finite element method (FEM) simulations. Additionally, the impact of various attitude conditions on braking performance is systematically analyzed. Finally, experimental results obtained from a full-scale prototype and measurement system show excellent agreement with the theoretical predictions, thereby confirming the model's accuracy and practical applicability.

Cryogenic pulsating heat pipes (PHPs) show great potential for cooling high-temperature superconducting (HTS) devices. Conventional HTS current leads employ copper braids and ceramic insulators for conductive cooling, but electrical insulation can introduce substantial contact thermal resistance. To simultaneously achieve efficient cooling and electrical insulation, a quartz–stainless-steel hybrid liquid‑nitrogen PHP is proposed for HTS current‑lead cooling. Unlike transparent adiabatic sections in previous dual-material PHPs that have mainly been used for flow visualization, the quartz section in the present design serves as a functional insulating component, providing both electrical and thermal insulation while suppressing axial heat leakage through the tube wall. A two-dimensional computational fluid dynamics (CFD) model based on the volume of fluid (VOF) method, including solid domains, was developed to investigate the effects of adiabatic-section material and filling ratio (FR) on two-phase flow and heat-transfer performance. The results show that the quartz adiabatic section suppresses axial heat leakage, promotes phase-change-driven heat transport, and establishes a vapor-driven pressure difference of approximately 762 Pa. Compared with the all-stainless-steel configuration, the hybrid PHP exhibits lower thermal resistance and higher effective thermal conductivity. At a heat load of 3 W, its effective thermal conductivity reaches 14,290 W/(m·K). Among the four FRs of 50%, 60%, 70%, and 80% studied in this work, an FR of 70% shows the best overall heat transfer performance. Overall, reducing axial heat conduction in the adiabatic section enhances heat transport driven by fluid oscillation and phase change, providing guidance for PHP-based HTS current‑lead cooling design.

Although permanent magnet synchronous motors offer high efficiency and power density, their performance is often compromised by inherent harmonic currents and torque pulsations, limiting applicability in high-precision motion control. While conventional deadbeat predictive current control (DPCC) improves dynamic response and control performance, it fails to mitigate harmonic-induced vibration due to its limited observation capability for high-frequency disturbances. To address these challenges, this article proposes a harmonic-suppression DPCC with a decoupled compensation structure. First, a parameter-robust extended state observer is designed to actively estimate and compensate for low-frequency disturbances caused by parameter variations. Second, parallel multiloop quasi-resonant controllers are embedded in both the current and speed loops to provide high gain at targeted harmonic frequencies, enabling precise cancellation of periodic torque harmonics. Electromechanical simulations and experiments validate the proposed approach, demonstrating significant harmonic suppression and robust performance. The results confirm its potential for high-precision, low-vibration motion control applications.