Conventionally considered as detrimental to torque density, flux-leakage contains components that can improve output capability. Due to the nonlinear permeability of the core material, flux-leakage flowing into the bridge is more beneficial than that flowing into the air. The useful component is defined as utilized flux-leakage (UFL), with the remainder termed nonutilized flux-leakage (NUFL). Building on this distinction, a novel motor design strategy named recycling flux-leakage is proposed. It is divided into three key steps: 1) inlet control; 2) path selection; and 3) outlet guidance. The inlet control step ensures that all flux-leakage enters the bridge. The path selection step minimizes the flux-leakage losses when passing through the bridge and maximizes the proportion of UFL. The outlet guidance step aims to guide partial UFL into the air-gap, converting it into a unique effective flux defined as the continuous pole flux (CPF). The proposed design strategy is applied to a conventional spoke-type motor. The resulting improved rotor features elements such as square barriers and L-shaped bridges. Through simulation and experimental verification, the proposed rotor can simultaneously improve PM utilization and torque density.

The linear primary permanent-magnet vernier machine (LPPMVM) is a promising candidate for long-stroke applications owing to its high thrust density and cost-effectiveness. Nevertheless, significant thrust ripple persists due to rich spatial harmonics in the air-gap magnetic field. To address this limitation, this paper proposes a novel linear primary permanent magnet vernier machine incorporating a reverse Halbach array (RH-LPPMVM). The core innovation lies in the synergistic use of reversed magnetisation segments and a deliberate permeance phase shift to selectively attenuate specific harmonic orders in the armature flux density distribution. Analytical and finite-element studies reveal that the 2nd and 22nd harmonic components dominate electromagnetic force fluctuations. By suppressing these critical harmonics through the proposed reverse-Halbach design, the RH-LPPMVM achieves a significant reduction in thrust ripple while maintaining the average thrust. Experimental validation on a prototype confirms the overall effectiveness of the proposed method, though discrepancies between predicted and measured ripple are discussed and attributed to practical manufacturing factors.

Magnetic screw motors (MSMs) facilitate the conversion of rotary motion into linear motion through magnetic coupling. Nevertheless, relative motion errors are inherently present within the magnetic transmission process. To acquire the relative motion error in real time during the operation of the MSM, this article proposes a relative motion error observer based on the MSM transmission structure. The rotary component can be calculated by the magnetic transmission ratio of the MSM and the reference linear component of the mover. Then the calculated rotary component is compared with the actual rotary component obtained from sensors, thereby the real-time determination of the relative motion error can be realized. To observe relative motion errors, the rotor position, linear speed, and force of the mover are respectively designated as control targets, and the corresponding closed-loop control systems are established. The observation of relative motion error facilitates the analysis of the MSM transmission characteristics and the implementation of various control strategies. For instance, relative speed compensation can provide a foundational approach for sensorless control of linear position. Furthermore, to improve the dynamic performance of the closed-loop system without increasing the complexity of the system, the adaptive fuzzy PI controller is implemented in the control system.

The Magnetic Screw Motor (MSM) has emerged as a promising technology for precision linear drives, offering non-contact operation, high efficiency, and low maintenance. However, MSMs face challenges in maintaining performance stability due to transmission ratio fluctuations caused by structural flexibility. Traditional control strategies, such as deadbeat direct torque and flux control (DB-DTFC), offer fast and precise torque and flux tracking, making them effective in many motor control applications. However, in MSMs, these strategies are limited by their inability to effectively reject disturbances arising from the variable transmission ratio. To address this limitation, we propose a hybrid control strategy integrating Active Disturbance Rejection Control (ADRC) with Deadbeat Direct Torque and Flux Control (DB-DTFC). Furthermore, a Dynamic Flux Compensation (DFC) strategy is incorporated for hierarchical flux adaptation. By leveraging ADRC’s disturbance estimation and compensation capabilities alongside DFC, the proposed method significantly improves disturbance rejection and system robustness. Simulation results validate the effectiveness of the proposed approach in enhancing MSM performance under varying operational conditions.

The standard inverter topology of five-phase permanent magnet synchronous motor (PMSM) has weaker fault-tolerant capability than other complex drives. Additionally, traditional open-switch fault-tolerant strategies demand precise diagnosis methods involving numerous states of different phases and reconfiguration according to fault phase change. To enhance fault-tolerant algorithm applicability, a universal fault-tolerant space vector pulsewidth modulation (SVPWM) strategy without extra hardware is proposed to handle single open-switch fault for five-phase PMSM. The most significant innovation is to achieving fault-tolerant operation for arbitrary phase open-switch fault without requiring adjustments based on fault phase location. Owing to this, corresponding diagnosis process can be simplified. In diagnosis process, the positive and negative relationship of the third harmonic currents within faulty half cycle is adopted to obtain the overall diagnosis result without requiring precise identification. In fault-tolerant algorithm, postfault basic voltage vectors are first reconstructed based on switching state analysis. Then, to minimize the influence of harmonic subspace and improve SVPWM control precision, virtual voltage vectors are synthesized based on geometric meaning of basic voltage vectors. Furthermore, through establishing vector sets involving five phases and summarizing their intersection regions, universal virtual voltage vectors are determined. Finally, the effectiveness of proposed method is verified by experiments.

Conventional fault-tolerant methods for open-switch faults (OSFs) in five-phase permanent magnet synchronous motors (PMSMs) tend to exhibit a certain degree of complexity. To reduce the complexity of fault-tolerant control, this article proposes a model predictive fault-tolerant control strategy based on discrete space vector modulation (DSVM). The core of the method lies in utilizing reconstructed voltage vectors to expand the candidate vector set and designing a cost function based on voltage error to simplify the vector selection process. Furthermore, dynamic segmented current references are employed according to the location of the faulty switch, reducing torque ripple during fault-tolerant operation. By eliminating the need for modifications to the motor model or the use of reduced-order matrices within the control process, this approach significantly simplifies the control structure. Experimental results demonstrate that the proposed method achieves satisfactory fault-tolerant operation capability and maintains good dynamic response performance under single OSF conditions.

Domain adaptation approaches have been applied to motor rolling bearing fault detection. Nevertheless, these approaches are constrained by the assumption of label space consistency between source and target domains. In complex industrial application scenarios, it is more common that the target domain label space is a subset of the source domain label space. To address such partial domain adaptation problems, a multi-scale feature calibration partial transfer network is proposed in this paper. This method extracts discriminative features from fault signals through a multi-scale convolutional feature extraction network to improve the accuracy of transfer classification. A domain shift calibration module is designed to explicitly learn cross-domain differences, achieving fine-grained alignment at the feature level. Furthermore, a domain-class joint distribution alignment mechanism is developed. On one hand, this mechanism enhances the transferability of shared classes and suppresses the negative transfer of outlier classes through weighted class alignment. On the other hand, the feature-level and task-level joint distribution alignment mechanisms work synergistically to minimize cross-domain discrepancy. Experiments on three bearing datasets verify the effectiveness and superiority of the proposed method.

An iron loss prediction method that combines the energy-based hysteresis model with the statistical theory of loss and the field separation approach is proposed. The prediction method aims to provide a more physically realistic description of magnetic energy dissipation. First, a static energy-based hysteresis model is established based on thermodynamic principles, drawing inspiration from both the Jiles-Atherton model and the Preisach model. The static model accurately predicts hysteresis loss. Subsequently, a dynamic iron loss model is developed by incorporating statistical loss theory into the static energy-based hysteresis model through the field-separation approach. Eddy current loss and excess loss are taken into account by introducing an additional field strength associated with dynamic iron losses into the total magnetic field strength. Finally, an Epstein frame testing platform is constructed to measure iron loss over a range of frequencies. Comparisons between analytical results and measured results show good consistency. The proposed model serves as a promising approach for iron loss prediction, offering a sound physical foundation and strong scalability.

With its powerful feature extraction and early fault identification capabilities, deep learning-based diagnosis of interturn short circuit (ITSC) faults has gained increasing attention to ensure the reliable and safe operation of induction motors. However, its practical deployment in industrial settings faces a significant challenge: performance degradation due to distribution shifts in current signals under different working conditions. To overcome this issue, we propose a deep multiscale subdomain adaptive network (DMSAN) for cross-condition ITSC fault diagnosis. The framework first employs parallel multiscale convolutional kernels with residual connections to capture multiscale fine-grained information from raw current signals. Subsequently, a subdomain adaptation layer integrated with multikernel local maximum mean discrepancy (MK-LMMD) is applied to align feature distributions across domains. Furthermore, a dynamic weight strategy is designed to optimize the classification loss and subdomain loss, thereby ensuring model convergence stability. DMSAN demonstrates superior performance and stability across fifteen cross-condition fault diagnosis transfer tasks, achieving an accuracy of 99.59%. This represents a substantial improvement compared to existing methods, including conditional domain adversarial network (CDAN) (91.63%), generative adversarial networks (GANs) (87.76%), domain adversarial neural network (DANN) (88.05%), and deep adaptive network (DAN) (82.83%).

This article proposes a fault-tolerant control (FTC) strategy applicable to a multimode dual stator permanent magnet synchronous motor under two-phase open-circuit fault (OCF) conditions. Traditional FTC research primarily focuses on motors with a single set of windings. While these methods can be directly applied to a dual-stator motor, they fail to fully exploit the multimode operation characteristics of such a motor. Based on the principle of maintaining constant magnetomotive force before and after the fault, the fault-tolerant reference currents and the reduced-order decoupling matrices are derived for both adjacent and nonadjacent two-phase OCF, respectively. An FTC strategy based on the coordinated injection of d-q axis currents into the inner and outer stators is proposed, aiming to minimize copper loss. Under different operating modes, the torque equation and copper loss formula are used to construct a Lagrangian equation, from which the optimal reference values for the current distribution coefficients injected into the inner and outer stators are derived. Meanwhile, based on the copper loss discriminant, the optimal operating mode is selected. The proposed fault-tolerant strategy achieves minimum copper loss control across the full torque range and demonstrates excellent dynamic response. Experimental results validate the effectiveness of the strategy.

Existing ITSC fault diagnosis methods for electric machines still face challenges in terms of incipient fault detection, noise robustness, cross-motor type applicability and multiple operating conditions. To this aim, this article proposes a novel dynamic sparse convolutional residual network fault diagnosis model for detecting incipient ITSC fault and estimating severity levels in electric machines. First, a novel dynamic sparse convolution paradigm is proposed in the diagnostic model to extract local features and long-range periodic dependencies from time-series signals. Second, an improved squeeze-and-excitation mechanism is introduced to adaptively enhance the representation of fault-related discriminative features. Extensive experiments carried out across three distinct electric machine platforms demonstrate that the proposed method achieves superior diagnostic accuracy and noise robustness compared with the state-of-the-art models, while maintaining low computational cost. The proposed method also exhibits good generalization performance for unseen machines. Furthermore, visualization analysis is conducted to enhance the model’s interpretability.

Permanent magnet synchronous motors are widely used in various industrial drive systems. Accurately diagnosing inter-turn short-circuit (ITSC) faults in windings is a crucial prerequisite for ensuring the safe and stable operation of the drive system. Hence, this article proposes a spatiotemporal multibranch graph convolutional network (ST-MGCN) for the diagnosis of incipient ITSC faults. To address the challenge of weak fault features under complex conditions, ST-MGCN employs a three-branch architecture to simultaneously extract spatial, temporal, and global features from multichannel current data, thereby enriching the representation of fault features. A novel causality-driven fusion module adaptively reweights branch contributions and feature significances, preserving subtle incipient fault representations. Experimental validations on two distinct datasets exhibit the superior diagnostic accuracy and noise robustness of the proposed ST-MGCN. It demonstrates exceptional sensitivity to the weak features of the incipient ITSC fault. It maintains high performance under complex conditions, with varying speeds and load levels, outperforming the state-of-the-art methods.

There are few studies on two open-switch diagnosis and fault-tolerance for standard five-phase permanent-magnet synchronous motor. To improve the reliability of the motor system, this article proposes a vector-reconstruction-based fault diagnosis method and fault-tolerant space vector pulsewidth modulation (SVPWM) strategy. By analyzing and comparing the reconstructed voltage vectors under different two open-switch fault types, distinctive fault characteristics of vector deficiency magnitude and direction are identified, which serve as the basis for establishing diagnosis criteria. Through evaluating the average amplitude of the reference voltage vector and the angle of synthesized reference voltage vector in one electrical cycle, the initial identification of the fault type and subsequent further determination of the fault location can be achieved. The diagnosis process is simple and convenient. To handle the dislocation and dissymmetry of the postfault voltage vectors, the proposed SVPWM strategy employs geometric principles (vertical line) to synthesize virtual voltage vectors, thereby minimizing harmonic voltage components. Finally, the effectiveness of proposed methods is validated by experiments. Comparative analysis further indicates that the impact of two same-side open-switch fault on the adjacent phase is the most severe among same level two open-switch fault types.

This article aims to analyze and design a Halbach-array dual permanent magnet fault tolerant vernier machine (HEDPM-FTVM) that achieves “high-torque, low-loss, and robust fault tolerance.” Unlike conventional surface-mounted permanent magnet fault tolerant vernier machine (SMPM-FTVM), the proposed design employs an alternating-pole rotor structure and embeds Halbach-arrayed permanent magnets (PMs) within the split stator teeth. These innovations enable the machine to attain higher torque while significantly reducing PM eddy current losses. The study proceeds as follows: The topology and operating principles are introduced, followed by building a magnetomotive force (MMF) permeance model to investigate torque enhancement and PM eddy current loss reduction mechanisms. Critical design parameters are analyzed via finite element analysis (FEA) and optimized holistically through multiobjective optimization. Finally, prototype fabrication and testing validate the design and analytical findings.

This article proposes a fault-tolerant sensorless control strategy based on a frequency adaptive extended state observer (ESO) with complex coefficient filter (FAESO-CCF) for a 3x3 -phase permanent magnet (PM)-assisted synchronous reluctance motor (PMA-SynRM) with single-phase open-circuit fault (OCF). By controlling the cooperative operation of each module, the torque ripple can be significantly reduced, thereby realizing the fault-tolerant operation under single-phase OCF. In order to meet the requirements of fault tolerance, the input voltage and current of the position sensorless are severely distorted, resulting in a decrease in the estimation accuracy of the rotor position. To solve this problem, the frequency adaptive ESO (FAESO) with bandpass filter properties is proposed. Moreover, a complex coefficient filter (CCF) is introduced to extract the third and fifth harmonics in the active flux estimated by FAESO, so that the bandwidth of FAESO can be appropriately relaxed and the response speed of the system can be improved. Meanwhile, a type-3 phase-locked loop (PLL) is designed to overcome the defect of type-2 PLL that there is a steady-state error during ramp acceleration, which further improves the accuracy of system. Finally, the accuracy and effectiveness of the proposed observer are evaluated by experiments.

The broken rotor bar (BRB) fault is one of the typical faults of induction motors (IMs). The accurate fault diagnosis of BRB can effectively reduce economic losses and enhance the stability of the system. Aiming at the problems of limited data collection and the difficulty in estimating the severity of faults, a double-branch improved residual shrinkage network (DB-IRSN) for BRBs in IMs was proposed, where both transformed current and vibration signals are utilized. First, the Hilbert transform is utilized to enhance fault features in current signals. Subsequently, the current and vibration signals are input into the IRSN simultaneously to filter the noise component of the signal. Moreover, the dense connection mechanism is used to connect the residual shrinkage module to avoid the loss of effective information. Then, regularization methods are used to avoid the overfitting phenomenon under small samples, and the bidirectional gated recurrent unit (BiGRU) is used to fuse spatiotemporal features. DB-IRSN is capable of fully extracting the fault characteristics from two signals to achieve good performance in diagnosis. Finally, the experiments are carried out on two datasets obtained from Sao Paulo University and the experimental platform under different severities of BRBs and loads. When the training set accounts for 0.1, the model achieves an accuracy of 97.98% and 97.65%, respectively, with performance reaching 99% as the proportion of the training set increases. In addition, experimental results in noisy environments and unbalanced voltage (UV) indicate that the proposed DB-IRSN demonstrates satisfactory performance under small samples. Comparative experiments show that DB-IRSN has higher accuracy compared to existing networks.

This paper aims to analyze the electromagnetic vibration of the permanent magnet vernier machine (PMVM) with dynamic eccentricity effect. Firstly, the topology of the PMVM and the definition of the dynamic eccentricity effect are introduced. Then, the magnetic field of the PMVM is calculated. The PM MMF and armature reaction MMF of the PMVM with dynamic eccentricity effect can be transformed by the eccentric air-gap permeance to change the magnetic field. And it can generate the unbalanced magnetic force (UMF) that can have bad impact on the electromagnetic vibration. What is more, the flux density and the UMF are calculated by finite element method (FEM) and analytical method. And the result shows that the electromagnetic vibration of the PMVM with dynamic eccentricity effect is worse. Finally, a new winding structure has been proposed to reduce the UMF of the PMVM with dynamic eccentricity effect. The result shows that it can reduce the electromagnetic vibration without scarifying electromagnetic performance.

This article presents a new sensorless fault-tolerant control approach for five-phase permanent magnet synchronous motor (PMSM) using a sliding mode observer (SMO) due to its strong robustness and stability. Asymmetric operations under open-circuit faults (OCFs) cause high harmonics and dc components in back EMFs, making it challenging to estimate speed and position. Thus, in this article, a cross-coupling second-order sequence filter (CC-SOSF) works for extracting the symmetrical sequence components of back EMFs by eliminating the harmonic. In addition, a dc mitigation circuit is added to this filter to eliminate dc bias. To ensure frequency adaptation and avoid the issue of the estimated frequency feedback, an open-loop frequency estimator (OLFE) is adopted. Finally, a phase-locked loop calculates the estimated rotor position and speed. Simulations and experimental tests confirm the effective elimination of both harmonics and dc components, leading to accurate position and speed estimation in the presence of OCFs.

There is rare research on open-switch fault tolerance for standard five-phase permanent-magnet synchronous motor (PMSM). To enhance the reliability of the motor system, this article proposes a vector-deficiency-based fault diagnosis method and two different fault-tolerant space vector pulsewidth modulation (SVPWM) strategies. They are all grounded on the reconstruction of voltage vectors postfault. By applying the distinctive characteristic of reconstructed-vector deficiency direction to monitor the length and angle of the reference voltage, the fault location can be determined, which is simple and convenient. To handle the dislocation and dissymmetry of the vectors postfault, one of two SVPWM strategies employs geometric principles (vertical line) to synthesize virtual voltage vectors. While the other one applies the Cramer rule to realize 4-D calculation in both fundamental and harmonic subspaces. Both the SVPWM strategies can reduce harmonic voltage components. Furthermore, to improve motor efficiency, the virtual signal injection (VSIC) maximum torque per ampere (MTPA) method is successfully extended to open-switch fault circumstance by adjusting the inverter voltage postfault within the algorithm. Finally, the effectiveness and superiority of the proposed methods are validated by experiments. Besides, two fault-tolerant SVPWM strategies exhibit distinct advantages and disadvantages in torque performance, current quality, noise, dynamic response, and MTPA tracking.