With the increasing demand for high torque density and strong overload capability in aircraft propulsion systems, partially or fully ironless-stator permanent magnet synchronous machine (PMSM) topologies have attracted growing attention due to their lightweight potential and excellent overload performance. To investigate their suitability for various aircraft mission profiles, this paper presents a comprehensive comparative study of slotless, coreless, yokeless, and conventional slotted PMSMs. First, a multiphysics design methodology for a high-torque-density propulsion motor, including the electromagnetic topology, mechanical structure, and cooling system, is presented using a yokeless-stator PMSM as an example. Subsequently, the four motor topologies are optimized through a multi-objective optimization approach to ensure a fair comparison. Their performances are then systematically evaluated in terms of flux density and inductance characteristics, torque density and overload capability, torque ripple considering pulse-width modulation (PWM) effects, loss and efficiency, and thermal behavior. Based on these analyses, the advantages, limitations, and application prospects of different topologies for electric aircraft propulsion are discussed. Furthermore, quantitative performance evaluations over mission profiles with different takeoff-to-cruise power ratios (TCPRs) are conducted. The results show that the coreless PMSM exhibits advantages in high-TCPR aircraft applications due to its superior overload capability, while the yokeless PMSM achieves excellent overall performance aiming at both torque density and overload capability, making it suitable for a wide range of TCPR conditions. Finally, a 100-kW yokeless-stator PMSM prototype for aircraft propulsion is developed and experimentally validated.
The replacement of synchronous machines by converter-interfaced generation lowers grid inertia and makes parallel grid-forming inverters (GFMs) prone to active-power interaction oscillations when virtual inertia and line impedances are mismatched. This work proposes a distributed fuzzy adaptive inertia (DFAI) control that adjusts each GFM's virtual inertia in real time using a consensus of neighboring frequencies and a Mamdani fuzzy inference rule base. By replacing direct-gain multiplications with fuzzy rules, DFAI mitigates error amplification due to magnitude disparities and measurement noise. A small-signal model of multi-GFM systems, including filter and line dynamics, is derived. Root-locus and eigenvalue analysis reveals how parameter inconsistencies excite oscillations and how DFAI enlarges the stable region and increases damping. Closed-loop state-space equations with DFAI are obtained to formalize stability, and a polytopic (vertex-LMI) Lyapunov certificate establishes exponential stability over the entire admissible inertia range for arbitrarily fast adaptation. Hardware-in-the-loop experiments with intentionally mismatched inertia and line impedances validate the analysis: DFAI effectively dampens interaction oscillations, improving decay time and overshoot relative to a conventional virtual synchronous generator (VSG), while maintaining robustness to communication noise, delay, and packet loss. The results indicate that consensus-driven fuzzy inertia adaptation is an effective and implementation-ready means to coordinate parallel GFMs under heterogeneous network and control parameters.
With the increasing penetration of renewable energy sources, maintaining grid stability requires grid-forming (GFM) converters to preserve their GFM characteristics under both steady-state and fault conditions. Although existing fault ride-through (FRT) strategies for GFMs can handle conventional voltage sag faults, they often lead to synchronization instability and current overlimit issues during grid phase jumps due to the lack of a fast synchronization mechanism. To address this challenge, this paper proposes a transient phase control strategy based on a virtual active power synchronization mechanism, which achieves fast response to various grid faults—including voltage sags and phase jumps—without relying on a phase-locked loop (PLL) or grid parameters, making it particularly suitable for weak-grid scenarios. By designing a dynamic calculation method of virtual power, the proposed strategy enables rapid and stable control of the transient power angle, offering strong fault adaptability and engineering feasibility. Simulation and hardware experimental results verify that the proposed method significantly enhances the transient synchronization performance and fault ride-through capability of GFM converters.
Transient synchronization stability (TSS) of phase-locked loops (PLLs) during asymmetric faults has been widely explored under a simplified assumption of ideal positive and negative sequence current controls. However, for high voltage direct current systems based on modular multilevel converter (MMC-HVDC) with low switching frequency limitations, the bandwidth of the dual-sequence current controls cannot be designed to be too high. In this regard, the simplified assumption may lead to misjudgments of the TSS. Moreover, most existing transient analysis methods are only applicable for low-order PLL systems, leaving the high-order nonlinear systems with dual-sequence current loop dynamics untouched. In this paper, a coupled interaction model of the dual-sequence current control loops and the PLL is first established for MMC-HVDC systems under single line-to-ground faults. Based on the extended normal form theory, a transient stability assessment method under asymmetric faults is proposed. This method decouples the original 10th-order nonlinear system into multiple univariate and bivariate quadratic equations. By analyzing parabolas and basins of attraction in low-dimensional space, the transient stability criteria of the 10th-order system are obtained and the dominant subsystem reflecting loop coupling effects is identified. Furthermore, the scope of applicability of the simplified PLL dynamic model is clarified by depicting the transient stability region formed by the dual-sequence current loop bandwidths. Besides, the influence laws of positive and negative sequence current phases are also analyzed. The bandwidths and references design method of dual-sequence current controls are proposed under asymmetric faults. Finally, hardware-in-the-loop experiments are conducted to validate the theoretical analysis.
The dual-excited synchronous condenser (DESC) can provide reactive power and inertia support to the renewable energy power systems, but its asynchronous operation capability (AOC) is severely limited by rotor overheating caused by damping losses. This paper investigates the enhancement of asynchronous operation capability of a 50 MVar DESC through a novel rotor cooling structure. Firstly, the loss generation mechanism during asynchronous operation is analyzed based on the magnetomotive force-air gap permeance model (MMF-AGPM), clarifying the harmonic orders contributing to damping losses. Secondly, a multi-domain integrated model of the DESC is developed to predict losses and thermal behavior under different slip ratios. Thermal analysis reveals that the rotor hot spot shifts to the slot wedges when slip ratio exceeds 3$\%$, with asynchronous operation time constrained by the 130 $^{\circ }$C limit of aluminum slot wedges. A novel bilateral fin cooling structure is proposed and designed, showing that increasing the number of fins is significantly more effective than increasing fin height. With the designed structure, asynchronous operation time at 5$\%$ slip ratio is extended from 273 s to over 1054 s. A power system model incorporating thermal power and renewable energy is developed to validate the DESC's inertia support capability for the power grid. Finally, a 10 kVar scaled-down DESC prototype is manufactured and grid-connected experiments are completed, with test results verifying the correctness of the proposed methods and models.
A fixed-time extended state observer-based control (FESOC) strategy is devised to maximize tidal energy capture in a permanent magnet synchronous tidal power generation system (TPGS). The system operates in a complex and dynamic marine environment with time-varying ocean currents, external disturbances, and model parameter uncertainties. These factors increase the difficulty of control design. The method aims at efficient maximum power extraction and robust performance under disturbances and uncertainties. The nonlinear TPGS is transformed into a linearized form via feedback linearization. A linear control law is then designed to ensure high-performance operation under time-varying ocean current speed. All perturbations, including disturbances, uncertainties, and nonlinearities, are lumped into an extended state. A fixed-time extended state observer estimates this extended state in real time. This design improves disturbance rejection and robustness. The effectiveness of the FESOC strategy is validated through real-time simulation results.
The contradiction between short-circuit current suppression and weight/size reduction in automatic turn-to-turn short-circuit fault-tolerant permanent magnet synchronous motors hinders the practical feasibility of the automatic fault-tolerance concept for motor applications. To address this contradiction, this paper proposes a novel automatic turn-to-turn short-circuit fault-tolerant permanent magnet synchronous motor. The motor adopts a non-coplanar flux path topology and in-phase-compensation method, which enable most of the additional leakage flux path to be detached from the main flux path and provide multiple parallel branches, thereby alleviating the contradiction. Its operating principle and power transmission mechanism under fault-tolerant conditions are elaborated. Furthermore, voltage equations incorporating asymmetry and a short-circuit current equation taking into account magnetic saturation are derived. To validate the proposed solution, a 1-kW prototype was designed and fabricated. Finally, experimental results, together with a comparative weight and size analysis, demonstrate that the proposed motor effectively alleviates the aforementioned contradiction.
Airborne Wind Energy (AWE) systems using Leading Edge Inflatable (LEI) kites offer promising solutions for accessing consistent high-altitude wind power with reduced material requirements compared to conventional wind turbines. Despite this potential, their development is limited by the lack of emulation platforms that capture both aerodynamic and electromechanical dynamics. This work presented an emulator for a 12 $m^{2}$ LEI kite energy system, including Computational Fluid Dynamics (CFD) simulations in ANSYS Fluent and vector-controlled Permanent Magnet Synchronous Motor (PMSM). The PMSM emulates the kite dynamics and is mechanically coupled with a PMSG, which delivers power to a programmable three-phase resistive load via an AC-DC-AC conversion stage. The presented emulator is implemented in MATLAB Simulink simulations and shows the control of the system performances by Vector-Controlled Symmetrical Optimum Methods (VC-SOM). A distinct operational transition occurs at 35 seconds, when the kite begins effective power generation, with the control system successfully managing the transition from zero to active power generation states. The technique is further verified on a laboratory Hardware-in-the-Loop (HIL) system utilising an OPAL-RT 4520 controller and an OP5033 XG for real-time simulation. The experimental HIL findings show high emulation fidelity, with tracking errors of 2.0% for speed, 4.6% for torque, and 3.1% for power, and an average electromechanical conversion efficiency of 91%. The proposed emulator provides a robust testbed for AWE system development, enabling validation of control algorithms and power conversion strategies before physical deployment, thereby accelerating the technological readiness of kite-based renewable energy systems.
This paper proposes a hybrid excited asymmetric consequent-pole flux reversal permanent magnet machine with dc-biased sinusoidal current (DC-biased HE-ACPFRPMM) to simultaneously achieve high torque production and good flux weakening capability. The proposed machine, featuring an asymmetric CP PM arrangement and DC-biased current, can not only effectively reduce leakage flux but also alter the polarity of the coil flux linkage by applying different polarities of DC excitation. This enables a significant regulating effect on the synthetic flux linkage and achieves flux regulation. In addition, the proposed DC-biased HE-ACPFRPMM can also achieve high torque production through the flux modulation effect. The topology and operation principle of the DC-biased HE ACPFRPMM are first presented. Then, the selection of slot/pole combination and key parameter influence analysis are discussed. To evaluate the merits of the DC-biased HE-ACPFRPMM, the electromagnetic performances of the machine under different excitation states are investigated, and its performance is compared with that of two existing DC-biased machines. The results demonstrate that the proposed DC-biased HE ACPFRPMM can offer higher torque density, better flux weakening capability, and higher efficiency than that of the two contrast machines. Finally, an experimental validation is conducted.
This paper proposes a double-sided 10-X axial-flux variable-reluctance (VR) resolver for the independent estimation of the angular positions of two rotors. In the proposed structure, all excitation and signal windings are located on a common stator, thereby eliminating the need for a rotary transformer and simplifying the mechanical and electrical construction. Twelve toroidal excitation coils are employed together with two indepen-dent sine/cosine signal-winding sets positioned on opposite sides of the stator. A Modified Winding Function Theory (MWFT)-based model is developed to calculate the air-gap functions, excitation inductance and current, flux linkages, and induced signal voltages. The resolver performance is evaluated under healthy operation and static, dynamic, and mixed misalignment conditions. In the healthy simulated case, the maximum position error (MPE) and average absolute position error (AAPE) are 0.039° and 0.0132°, respectively. Under the specified 30% mixed-misalignment condition, the first rotor exhibits an MPE of 0.260° and an AAPE of 0.065°. Experimental tests are performed at approximately 1200rpm using an 18-bit optical encoder as the reference position sensor. In the healthy condition, the experimental MPE and AAPE are 0.042° and 0.015°, respectively, whereas under 50% static misalignment, they are 0.204° and 0.110°. The close agreement of the experimental and FEM results with the corresponding MWFT results verifies the proposed resolver and validates the developed modeling approach. The proposed resolver offers a compact, stator-winding-only position-sensing solution for high-pole-count electrical machines, particularly 20-pole traction drives.
Homopolar linear synchronous motors (H-LSM) represent a viable solution for high temperature superconducting (HTS) maglev requiring low-cost vehicle propulsion systems. However, due to the inherent 3D-field distribution of H-LSM, the necessary 3D finite element analysis (FEA) hinders its rapid design and optimization. This paper proposes for the first time a complete refined hybrid analytical framework model for H-LSM with discrete secondary segments. This model integrates the methods of airgap magnetomotive force-permeance, winding function theory, and conformal transformation, accurately considering the magnetic field modulation effects caused by discrete secondary iron segments and primary slotting, as well as the transverse and longitudinal end effects. The analytical predictions are verified by 3D FEA and a laboratory-scale prototype, demonstrating high agreement, which validates the effectiveness of the proposed model. This model reduces computational time by approximately 98% compared to FEA under the specific conditions, enabling rapid analysis and optimal design of H-LSM with complex 3D-field characteristics under arbitrary pole-slot combinations for HTS maglev applications.
Halbach permanent magnet (PM) machines have many attractive advantages such as higher average torque and lower torque ripple than conventional surface-mounted PM (SPM) machines. However, due to more PM segments per pole and repelling force between adjacent PMs, rotor tolerances during manufacturing in Halbach PM machines are more crucial. In this article, six types of rotor tolerances in Halbach machines are evaluated with focus on the influence of rotor back-iron, including PM thickness, width, remanence, magnetization angle, circumferential position, and radial position tolerances. A novel worst-scenario cogging torque identification method is proposed by combining the genetic algorithm and phasor diagram to reduce the computational burden. Based on this method, the worst-scenario cogging torque in iron- and air-cored Halbach PM machines is identified and evaluated. It is found that compared to the air-cored machines, cogging torque of iron-cored machines is more sensitive to rotor tolerances, although their ideal cogging torques without rotor tolerances are similar. Besides, among these tolerances, PM thickness and radial position tolerances are more critical, while PM remanence and magnetization angle tolerances have less impact. Finally, the iron- and air-cored Halbach PM machines are fabricated to verify the finite element analysis.
To improve the voltage stabilization control performance, a high dynamic cooperative voltage stabilization control is proposed for the generator and energy storage based hybrid electrical power system in the aircraft applications, which can effectively suppress the millisecond class transient periodic and non-periodic interference due to the electric actuator and the pulse load. The robust electrical power system (REPS) topology is proposed, which consists of the three-stage brushless synchronous generator system and the energy storage system (ESS). The generator is used to supply the steady electrical power, while the ESS is used to compensate for the transient electrical power. The coupling power supply behavior of the REPS is then thoroughly analyzed, which can provide a guideline for the voltage stabilization control design. The load current feedforward based voltage control is proposed to suppress the millisecond class transient non-periodic load interference, while the load current periodic component feedforward based voltage control is proposed to improve the REPS performance with the millisecond class transient periodic load interference. Furthermore, the notch compensation based current control is proposed to improve the dynamic response performance of the ESS, which can eliminate the LC resonance issue due to the bidirectional DC-DC converter. Finally, the effectiveness of the proposed voltage stabilization control is verified on a 40kW REPS experimental platform. The resulting REPS has an excellent voltage stabilization control performance even under the millisecond class transient periodic and non-periodic load disturbance conditions.
Minimum state-of-charge (SoC) violations caused by estimation uncertainty pose safety risks and accelerate battery degradation in energy storage systems. This letter proposes a risk aware time-adaptive conformal quantile regression (RT-CQR) framework for SoC interval estimation. An interval-induced upper bound on the expected minimum-SoC violation risk is established, linking the lower-tail behavior of prediction intervals to safety critical events. Building on this bound, RT-CQR unifies risk aligned quantile learning with violation-weighted time-adaptive conformal calibration to construct risk-aware SoC intervals. Case studies on diverse lithium-ion battery datasets demonstrate that RT-CQR reduces minimum-SoC violation risk with competitive interval sharpness and empirically calibrated coverage.
In this article, a cascaded brushless power-fed permanent magnet adjustable speed coupler (CBPF-PMASC) is proposed to satisfy the requirements for the speed regulation and energy saving of large power fan and pump loads. The coupler comprises two axially cascaded sub-systems: a power sub-system and a control sub-system, whose axial dimensions can be appropriately configured according to their respective power requirements. Firstly, the fundamental structure and speed regulation mechanism of the proposed coupler are illustrated and explained in detail. Subsequently, based on the equivalent circuit models of permanent magnet synchronous generators and wound-rotor induction machines, an integrated equivalent circuit model for the coupler is developed. The power flow within the coupler is then analyzed to demonstrate that the slip power from both the input and output shafts can be efficiently fed back into the grid through the stator winding. A power-sizing equation is derived to establish a theoretical basis for the electromagnetic design of the proposed coupler. The finite element analysis (FEA) predictions and experimental results for a 2.7 kW prototype corroborate the validity of the theoretical analysis and demonstrate the superiority of the proposed coupler.
In a large-scale orthogonal-winding planar induction motor (OWPIM) with a segmented primary, the magnetic-circuit discontinuity introduced by primary-core segmentation significantly influences its electromagnetic performance. To quantitatively investigate this effect, an analytical multi-subdomain model that explicitly accounts for the core segmentation is developed based on the subdomain method. Combined with three-dimensional finite-element analysis (3-D FEA), the proposed model is used to investigate the effects of segmentation on the air-gap magnetic field, secondary eddy currents, and electromagnetic forces, with particular emphasis on the x-direction segmentation effect, while the associated influence on the z-direction electromagnetic characteristics is also evaluated. The accuracy of the proposed model is validated through locked-mover tests, which demonstrate a maximum error of 4.13% and an average error below 3.5%. This work provides a quantitative elucidation of the impact of core segmentation on the electromagnetic behavior of OWPIMs. The findings provide a fundamental theoretical basis for the subsequent optimal design of such motors and the development of active electromagnetic force control strategies.
Synchronous reluctance motors (SynRM) have the characteristic of a large range of inductance variation. To ensure the signal-to-noise ratio (SNR) of the high-frequency (HF) current when estimating the rotor position using HF voltage injection method, the amplitude of the HF voltage is selected according to the maximum inductance, resulting in increased audible noise, HF losses, and torque ripple. To address these issues, a sensorless control of SynRM based on adaptive amplitude HF square-wave voltage injection is proposed. First, the online inductance identification method based on an adaptive affine projection algorithm is embedded in HF voltage injection. Then, based on the real-time changes of inductances, the amplitude of HF voltage can be adaptively adjusted to ensure the SNR of HF current, to effectively reduce audible noise, HF losses, and torque ripple. In addition, due to the d-axis inductance being greater than the q-axis inductance for SynRM, and SynRM has no true torque axis and excitation axis, injecting HF voltage into the q-axis instead of injecting it into the d-axis reduces the amplitude of the injected voltage when generating HF currents of the same amplitude. The effectiveness of the studied algorithm is verified on a 1.5 kW SynRM experimental test bed.
This paper takes the induction planar motor used in microgravity simulation as the research object. Considering the complex magnetic field distribution characteristics in the air gap, an equivalent magnetic network model based on the adaptive connection of cross-magnetic permeance in the full air gap area is proposed, which can perform adaptive magnetic permeance connection according to the grid nodes. To further improve the accuracy of electromagnetic field prediction, a hybrid analytical model is proposed by coupling the improved subdomain model and the adaptive equivalent magnetic network model, which taking advantage of the subdomain method to consider the effect of eddy current and the equivalent magnetic network method to deal with nonlinear problems, respectively. The air-gap magnetic field distribution and motor performance characteristics are studied. Finally, the accuracy of the proposed method is verified through three-dimensional finite element analysis and prototype experiments.
The braking torque of permanent magnet synchronous motor (PMSM)–based elevator systems obtained using contactor-based or conventional electronic short-winding braking decreases significantly at medium and high speeds, which may lead to hazardous operating conditions in the event of mechanical brake failure. To address this limitation, this paper proposes two model predictive control (MPC)–based electronic short-winding braking strategies that enhance braking torque without any additional hardware. Unlike conventional approaches that permanently short-circuit the stator windings by continuously enabling either the upper or lower inverter switches, the proposed methods fully exploit the admissible space voltage vectors under the same switch-disabling constraints while complying with the Safe Torque Off (STO) requirement. A maximum torque predictive control strategy is first developed to maximize the braking torque under the restricted voltage vector set, at the expense of increased torque ripple. To mitigate this effect, a target torque predictive control strategy is further proposed, which significantly reduces torque ripple with a slight reduction in average braking torque. Simulation and experimental results demonstrate that both strategies achieve substantially improved braking performance compared with conventional electronic short-winding braking methods.
Aiming at the output performance degradation in three-phase permanent magnet synchronous machines (PMSMs) caused by asymmetrical winding (ASW) due to specific slot-pole combinations, this article establishes the criterion for the approximate symmetrization design of ASW layout structures based on the compensation principle of stator fundamental MMF deviation vector. Based on this, two approximate symmetrization design methods accounting for constraints such as thermal balance and slot space are proposed. These two methods utilize the genetic algorithm to seek optimal solutions that alter the layout of the slot conductors of phase B and phase C, aiming to make the stator fundamental MMF deviation vector approach zero as closely as possible. This approach achieves a better symmetry of the three-phase ASW layouts without compromising the average torque and efficiency essentially. Meanwhile, the obtained winding schemes can achieve the balance of stator's circumferential thermal load and no obvious local hot spots. Finally, three 21-slot and 18-pole three-phase surface-mounted PMSMs with different winding layouts are manufactured and tested. Both finite-element analysis and experimental results validate the effectiveness of the two proposed approximate symmetrization design methods for the ASW layout structures.
