This article proposes a decoupled fault-tolerant model predictive current control for a dual three-phase permanent magnet synchronous motor with harmonic compensation. The harmonics caused by nonsinusoidal back electromotive force can be well controlled under the open-circuit fault scenario by designing a harmonic closed-loop scheme. First, a decoupled predictive model and fault-tolerant voltage vectors are deduced based on a reduced-dimension matrix, which can eliminate the coupling issues between the harmonic and fundamental subspaces. Afterward, a uniform harmonic-free virtual vector is constructed by rearranging the irregular voltage vectors. More importantly, a virtual null vector is designed in the scheme, where the virtual null vector is effective in the harmonic subspace and has no components in the fundamental subspace. So, the harmonics can be further controlled without affecting torque and flux generation. Extensive experimental results verify the effectiveness of the proposed method.

This paper proposes an improved fault-tolerant model predictive torque control (MPTC) for a five-phase permanent magnet synchronous motor based on deadbeat solution, which offers the reduced computation burden, good steady-state performance and simple control structure. Under the open-circuit fault condition, the virtual voltage vectors (VVV) which can restrain the harmonics work as the candidate vectors. The deadbeat solution is adopted to predict the reference voltage vector. The optimal VVV can be selected directly and rapidly according to the sector of the reference voltage vector. Hence the computation burden can be reduced largely because the process of testing all candidate VVVs is avoided. In order to reduce the error between the reference voltage vector and the selected VVV, a null vector is introduced by a simple geometric principle. Then, the duration of the selected VVV can be obtained. Therefore, the steady-state performance is enhanced because of the deadbeat solution and the multi-vectors in each control period. The experiments are presented to verify the proposed fault-tolerant MPTC.

The conventional duty ratio modulation based direct torque control (DR-DTC) for the dual three-phase permanent magnet (DTP-PM) motor reduces the torque ripple using duty ratio modulation in torque control. However, the flux was controlled by a hysteresis comparator, resulting in deficient steady-state performance. This paper proposes a novel DR-DTC strategy to improve torque and flux simultaneously. Firstly, a switching table with the multiple vector selection is newly established. Two adjacent virtual voltage vectors and a null vector are selected in each control period. Then, the duty ratios of selected adjacent virtual voltage vectors are optimized with the dual objective modulation considering torque performance and flux performance. In addition, the combination of adjacent virtual voltage vectors for the DTP-PM motor generates non-standard symmetrical switching sequences. Two switching sequences generated methods are further investigated to address the issue. So, the switching sequences of the proposed strategy can also be generated easily. Finally, the conventional strategy and the proposed strategy are compared through experiments, and the experimental results verify the effectiveness of the proposed strategy.

Deep learning theory is widely used in face recognition. Combined with the needs of classroom attendance and students’ learning status monitoring, this article analyzes the YOLO (You Only Look Once) face recognition algorithms based on regression method. Aiming at the problem of small target missing detection in the YOLOv3 network structure, an improved YOLOv3 algorithm based on Bayesian optimization is proposed. The algorithm uses deep separable convolution instead of conventional convolution to improve the Darknet-53 basic network, and it reduces the amount of calculation and parameters of the network. A multiscale feature pyramid is built, and an attention guidance module is designed to strengthen multiscale fusion, detecting different sizes of targets. The loss function is improved to solve the imbalance of positive and negative sample distribution and the imbalance between simple samples and difficult samples. The Bayesian function is adopted to optimize the classifier and improve the classification efficiency and accuracy, ensuring the accuracy of small target detection. Five groups of comparative experiments are carried out on public COCO and VOC2012 datasets and self-built datasets. The experimental results show that the proposed improved YOLOv3 model can effectively improve the detection accuracy of multiple faces and small targets. Compared with the traditional YOLOv3 model, the mean mAP of the target is improved by more than 1.2%.

Position estimation error caused by the cross-saturation effect seriously affects the performance of signal injection based sensorless control, especially for the surface permanent-magnet synchronous motor (SPMSM) with load. In this paper, a position estimation error compensation method is proposed to improve the accuracy and stability of sensorless control for a dual three-phaseSPMSM. Based on the space vector signal injection, the initial position is obtained by the demodulation of high-frequency current response. By analyzing the relationship between the stator flux and stator current, the optimal compensation angle can be achieved. The position estimation error can be dynamically compensated under load conditions with the combination of initial position and compensation angle. Due to the convergence of compensation angle, the compensation process has weak dependence on prior knowledge of the motor. Simulation and experimental results are presented to validate the proposed method.