Shoujun Song

Switched reluctance motors (SRMs) are are increasingly favored in high-demand applications such as more electric aircraft (MEA) due to their robust construction, high reliability, robustness against extreme environments, and wide speed regulation range. This paper provides a comprehensive review of reliability research pertaining to SRM drive systems. It systematically examines the principal failure modes affecting the motor body, power converters, and sensors, and surveys a spectrum of reliability assessment methodologies — including structure-based static model, statistical model based on stochastic process, and intelligent model based on data-driven. The review also discusses reliability improvement strategies for each component of SRM drive systems. Finally, this paper identifies the current challenges in reliability research and prospects future development directions. This work serves as a valuable reference for advancing reliability-centered research and development in SRM drive systems.

In distributed rear-wheel-drive electric vehicles (DRWDEVs), traction torque balance between the two driving wheels is usually achieved through chassis-level coordinated control of the dual permanent magnet synchronous motors; however, such coordination imposes certain demands on controller communication, vehicle motion sensors, and computational resources. To simplify control and provide safety redundancy under emergency conditions, this article proposes a motor-level coordinated control strategy. A single observer based on differential motor states replaces the traditional dual-observer scheme, and a hybrid extended state observer is developed. Experimental comparisons with a sliding-mode extended state observer demonstrate the superior performance of the proposed method. Furthermore, a relationship between the traction torque difference and the observer gain is established via a specially designed smooth nonlinear function, achieving a balanced improvement in dynamic and steady-state performance. The proposed approach attains traction torque balance during single-wheel slip in straight-line DRWDEV operation using only bottom layer motor control, without the need for vehicle motion sensors, vehicle dynamic models, or a hierarchical control architecture, thereby greatly reducing control complexity. Experimental results verify the effectiveness and superiority of the proposed control strategy in traction torque balancing.

The precise control of speed and torque in switched reluctance machines (SRMs) depends largely on high-precision, real-time position information acquired from coaxially mounted mechanical position sensors. However, in the shaft-line-embedded starter/generator (ESG) system, the operating environment temperature exceeds $350~^{\circ }$ C, where mechanical position sensors could not only increase the overall size of the drive system, but also reduce its reliability under high-temperature conditions. To explore the application of SRMs in the integrated starter/generator systems of more-electric aero-engines and remove mechanical position sensor, this article proposes a position sensorless control method based on linear inductance residual calculation combined with delay-compensated second-order generalized integrator frequency-locked loop with soft start-up (DCSOGI-FLL-WSS) to improve the position estimation accuracy at room and high-temperature environments. The performance of position estimation, system efficiency, and torque ripple under rated load, 1.5 times overload, and speed regulation conditions were also compared and analyzed. Experimental results show that the starting process of ESG can be easily achieved by the proposed method under different temperatures conditions. Due to the changes in the parameters of the SRS/G caused by the high-temperature environment, the accuracy of position estimation under the working condition of $350~^{\circ }$ C decreases. However, the proposed method demonstrates advantages in terms of position estimation accuracy, system efficiency, and torque ripple reduction for SRS/G systems.

This article proposes a novel five-level power converter with only 12 switches and 12 diodes for a four-phase 8/6-pole switched reluctance motor (SRM), combined with a model predictive torque control (MPTC) strategy to suppress torque ripple. To overcome the high computational burden caused by the multilevel design, a candidate voltage vectors (CVVs) selection method is designed, reducing the number of CVVs to 9 or 5 at each control period. The effectiveness of the proposed topology and control method is evaluated experimentally under different operating conditions. Experimental results show that the converter effectively maintains the midpoint average voltage balance, and the CVVs optimization method reduces runtime by at least 33.89% compared with the strategy without optimization. Furthermore, the system achieves a torque ripple reduction of at least 32.29% compared with the conventional asymmetric half-bridge power converter (AHBPC). Although the proposed converter is inferior to the AHBPC in terms of efficiency, cost, size, reliability, and weight, it still offers strong potential for applications requiring low torque ripple and high robustness.

In this paper, a sensorless finite-control set model predictive torque control (FCS-MPTC) strategy is proposed for switched reluctance motor (SRM) drives. To minimize the switching vector candidates and relieve the computing pressure in traditional MPTC methods, the flux linkage closed loop is adopted to assist candidate vector screening. In the controller, the flux linkage of each phase is calculated via the flux integral with the terminal voltage and phase currents. Through alpha belta transformation, the rotating flux angle and rotational speed can be estimated via phase locked loop (PLL). As the rotor position angle is synchronizing with the rotating flux angle, thus can be estimated directly by compensating the angle bias between them. The estimated position angle can be used for modeling the flux linkage and deriving the predicted torque, which is utilized in constructing the cost function in the MPTC method. Based on the aforementioned control process, the position sensorless control can be fully achieved. The experiments are performed under a 12/8 SRM test-bed, which verify the validity of the proposed method

In this paper, a current measurement and reconstruction control scheme for a four-phase switched reluctance motor (SRM) drive system with two current sensors is proposed. To implement the proposed scheme, following steps are included. Firstly, a compact power converter topology with modified current sensor measurement connection structure is developed, which utilized only 6 power switches and two current sensors. Secondly, in order to suppress the slow demagnetization and excessive negative torque caused by the shared reuse of power switches in the converter, two pulse injection schemes have been developed. Thirdly, based on the pulse injection scheme, a simplified four phase current reconstruction logic has been developed to ensure reliable estimation of phase currents from the measured currents of two current sensors. Detailed analysis and experiments are implemented with an 8/6 structure SRM prototype system, which fully verify the effectiveness of the proposed converter system topology and control schemes.

In this article, a deadbeat predictive control (DPC) method with low measurement effort and the capability to effectively suppress torque ripple is proposed for switched reluctance machine (SRM). First, to enhance prediction model accuracy, a flux-linkage model based on second-order Bézier curves combined with full-position interval interpolation is proposed. This method only requires determining three parameters—aligned-position incremental inductance, unaligned-position incremental inductance, and unaligned-position saturation inductance to accurately describe flux-linkage characteristics. Next, Varignon’s principle is applied to analyze the mechanism by which phase current harmonics influence torque. On this basis, the beetle antennae search algorithm, which enables rapid optimization, is used to optimize the current waveform, resulting in a novel harmonic-optimized current profile. The developed enhanced DPC methodology incorporating pulsewidth modulation demonstrates significant torque ripple reduction under diverse operating conditions, minimal offline measurement requirements, and high estimation accuracy, thus providing an efficient and practical solution for achieving high-performance SRM control in industrial applications.

To address the challenge that the accuracy of rotor position estimation in switched reluctance machines (SRMs) struggles to meet the requirements of high-performance control, this article proposes an optimization-based position estimation technique leveraging the mapping relationship between current prediction errors and rotor positions. By constructing polynomials for predicted current and rotor position using flux linkage information from the conducting phase, the method intelligently expands data samples and identifies the optimal estimated position through comparative analysis of predicted versus actual current deviations. Experimental results demonstrate that, compared to conventional nonconducting phase pulse injection-based approaches, the proposed method reduces the mean absolute value of error (MVE) ranging between $\text{0.65}^{\circ}$ and $\text{0.99}^{\circ}$ while exhibiting strong robustness.