In this paper, a fast and robust diagnostic method for multiple open-circuit (OC) faults is presented for voltage-source inverters. Only two line voltages are used as diagnosis variables, which is economic, simple, and can reduce the influence caused by the failure rate of information sources. Possible magnitudes of these line voltage vectors are characterized in detail both for the healthy and faulty operations. Then, several voltage features are extracted for diagnosis. Single-switch open-circuit (SSOC) faults and double-switch open-circuit (DSOC) faults can be accurately and quickly located through the proposed method. The fastest diagnostic process can be finished in 1/20 of the fundamental period. The diagnostic results of the proposed method are not affected by certain load OC faults, which may lead to false diagnosis of the existing methods. Moreover, the proposed method is robust to different control strategies, filter components, and carrier frequencies. Experiments are carried out on a dSPACE platform, and the effectiveness of the proposed method is verified.

Captured outdoor scene images are easily affected by haze. Most image dehazing methods have limited generalization capabilities for real-world hazy images owing to the complexities of real-world environments and domain gaps in the training datasets. This paper proposes a semi-supervised single-image dehazing network based on disentangled meta-knowledge. The symmetric and heterogeneous design of the disentangled network is conducive to the separation of the content and mask features of hazy images and these features are used as meta-knowledge to guide feature fusion in the dehazing network. Moreover, functions describing constant-color and disentangled-reconstruction-checking losses are designed to ensure the subjective qualities of the generated dehazed images. The results of extensive experiments conducted on synthetic datasets and real-world images indicate that the proposed algorithm outperforms state-of-the-art single-image dehazing algorithms. In addition, the algorithm effectively improves the performance of object-detection tasks. The source code is publicly available at https://github.com/dehazing/DNDM .

Detecting driver fatigue is critical to ensuring road safety. Existing fatigue detection methods typically rely on traditional hand-picked features as inputs. However, these hand-picked features can hardly respond accurately to the driver’s fatigue state due to a certain degree of subjectivity and the extraction of these features requires a long time window, which limits the accuracy and real-time performance of the detection. This paper proposes a novel fatigue detection method based on multi-source information fusion, which relies entirely on neural networks for automatic feature extraction. Through simulated driving experiments, we recorded physiological signals and facial videos from 21 participants for model training and testing. The results show that our model outperforms existing methods in terms of accuracy and real-time performance, achieving a detection accuracy of 93.15% within a 3-second time window (specificity $=$ 94.04%, sensitivity $=$ 91.71%). The visualization results of the model reveal potential relationships between facial regions for the first time, validating the rationality and effectiveness of our method. The practical issues of fatigue detection methods and future research directions are also explored.

The safety and reliability of inverters play a significant role in stable power electronic system operations. Considering a railway vehicle air-conditioning inverter as a case study, this paper presents a generic open-circuit fault diagnosis method that can be applied to various inverter and control schemes. First, a detailed model of three-phase output currents under both normal and fault status is derived, and then the concept of space trajectory is introduced based on the three-phase output currents. The open-circuit faults of the inverter lead to output phase current distortion. Therefore, different types of faults have different space trajectories, and the corresponding shape of the space trajectory is considered to be a fault feature. Then, the fault is located by the direction vector and center-of-mass coordinate. The effectiveness and reliability of the diagnosis method are verified and analyzed using a hardware-in-the-loop test platform.

The inverter is an indispensable part of a power electronics system. Its safety and stability are important indicators for evaluating the reliability of the system. Inverter faults are usually caused by operating faults in the switch elements. This paper therefore proposes a non-intrusive fault-diagnosis method for open-circuit faults in inverter semiconductor power switches. This method requires only a current signal. It is simple and economical. Faults can be diagnosed quickly using the proposed algorithm. First, the phase current waveforms before and after the open-circuit fault are analysed based on the mathematical model. Then, based on the analysis of the phase currents, the fault features are extracted and the fault is located using the proposed integration algorithm. Finally, the effectiveness and reliability of the proposed fault-diagnosis method are verified using a hardware-in-the-loop experiment.