This article proposes a fault-tolerant finite control set model predictive control for a five-phase permanent-magnet (PM) motor drive, which offers reduced computation burden and simplified control model. The virtual voltage vectors synthesized from two basic vectors are used to reduce the computation burden. Meanwhile, the steady-state performance is improved. Combining with a reduced decoupling matrix, the discrete model of the five-phase PM motor before and after fault remains unchanged. So, the reconfiguration of the control structure is minimal. Then, a control set and corresponding switching sequence are proposed, which are very suitable for the real-time system. Finally, the validity of the proposed fault-tolerant control is proved by experiments.

Deep neural networks (DNNs) have been shown to be vulnerable to adversarial attacks in the image domain. Recently, 3D adversarial attacks, especially adversarial attacks on point clouds, have elicited mounting interest. However, adversarial point clouds obtained by previous methods show weak transferability and are easy to defend. To address these problems, in this paper we propose a strong point cloud attack method named AOF which pays more attention to the low-frequency component of point clouds. We combine the losses from point cloud and its low-frequency component to craft adversarial samples and focus on the low-frequency component of point cloud in the process of optimization. Extensive experiments validate that AOF can improve the transferability significantly compared to state-of-the-art (SOTA) attacks, and is more robust to state-of-the-art 3D defense methods. Otherwise, compared to adversarial point clouds generated by other adversarial attack methods, adversarial point clouds obtained by AOF contain more deformation than outlier. Code is available at: https://github.com/code-roamer/AOF .

This paper proposes and designs a new five-phase modular flux-switching permanent-magnet (M-FSPM) machine for high reliability applications. The key is the new machine topology which incorporates the concept of fault-tolerant teeth (FTT) to provide the desired decoupling among phases. The electromagnetic performances of the newly designed M-FSPM machines having 18 and 19 rotor poles are quantitatively analyzed including the flux, back-EMF, cogging torque as well as unbalanced magnetic force (UMF). The results show that the proposed M-FSPM machine not only retains the merits of stator-PM machines and multiphase machines, but also offers lower cost and higher fault-tolerant capability. Especially, the 10/19-pole one incorporates the merits of symmetric back-EMF and reduced cogging torque, while suffers from UMF. Two experimental M-FSPM machines are designed and built for exemplification. Experimental results of the prototypes are given to confirm the validity of the proposed machines.

This paper proposed a new fault-tolerant direct thrust force control (DTFC) for a dual inverter fed open-end winding linear vernier permanent-magnet (LVPM) motor. First, the topology and structural characteristics are briefly presented. Second, the mathematical model of the LVPM motor is derived. Then, a new fault-tolerant DTFC is proposed, in which inverter fed open-end winding technique is adopted. When inverter switch fault occurs in open-end winding driving system, the connection point between the motor port and the fault switch is cut off, and motor port is connected to power source neutral point. An improved space vector pulse width modulation is reconfigured by the rest of switches to implement DTFC. Finally, experimented results are given to verify the proposed fault-tolerant control strategy.

The conventional five-phase finite control set (FCS) model predictive control (MPC) suffers from heavy computational burden. Meanwhile, the control set with finite vectors inevitably leads to deteriorated operation performance. This paper proposes a novel FCS model predictive current control (MPCC) with continued modulation. A cascaded optimization procedure is proposed, which includes main virtual voltage vector (VV) selection, determination of optimal combination of adjacent two VVs, and amplitude optimization. In this cascaded optimization procedure, the phase angel and the amplitude of the synthetized vector can be obtained in different steps. Also, there are two duty cycles calculated in different steps independently, which avoid overflow. By using this cascaded procedure, the continued modulation can be realized without using a modulator. Meanwhile, a vector selection method is introduced to reduce computational burden. This vector selection method avoids evaluation of all the candidate vectors without any negative effects. Moreover, the merits of conventional FCS-MPCC can be preserved. The experimental results verify the effectiveness and superiority of the proposed FCS-MPCC method.