随着直流电网的不断发展,对直流开断技术的要求进一步提高.相较于传统的气体电弧直流开断,真空电弧直流开断具有电寿命高、燃弧空间封闭、介质恢复速度快等特点,但是传统真空电弧电压低,需要通过横向磁场的作用,提升电弧电压完成直流真空电弧的开断.本文通过实验研究了横向磁场作用下不同触头侧部燃弧空间(即触头侧部边缘与真空灭弧室屏蔽罩之间的空间)与开断电流时的开断特性,获得了触头侧部燃弧空间宽度11～41 mm范围内对直流真空开断特性的影响规律.

The objective of this work is to experimentally determine the critical contact separation length Lcritical for DC interruption in vacuum under an external transverse magnetic field (TMF), according the interruption behaviors and vacuum arc characteristics. L is defined as the separation length at the instant of TMF application. Lcritical is the critical and the minimal necessary separation length of the contacts for a successful DC interruption in vacuum. A pair of Helmholtz coils was used to generate the external TMF. A high-speed camera was used to record the vacuum arc appearance. The experimental results quantitatively determine Lcritical for DC vacuum interruption under a certain DC and the external TMF. When the DC ranged from 100 to 500 A under an external TMF of 30 mT, Lcritical ranged from 2.1 to 3.7 mm and did not show linear variation. Second, the extinguishing time of DC vacuum interruption te at various currents significantly increased in a similar pattern under decreasing L in the range of 0.3–1.6 ms. The decrease in L mainly increased the duration of the stable stage and had little apparent influence on the duration of the unstable stage of the DC vacuum arc. Third, the physical mechanism determining Lcritical in DC vacuum interruption under a TMF could be explained from two aspects: the expansion of the vacuum arc column and the diffusion of the metal plasma. In successful DC vacuum interruption, Lcritical should be large enough for the formation and expansion of the “C”-shaped arc column, and enough space should be available for the diffusion of the metal plasma.

The objective of this paper is to determine the amplitude-frequency characteristics of the arc voltage of the DC vacuum arc subjected to an external TMF. The vacuum arc tended to be unstable in the presence of the external transverse magnetic field (TMF), where a violent oscillation of the arc voltage was observed. The oscillation can be used to design a new passive resonance vacuum DC circuit breaker. In addition, the parameters of the parallel branch for the passive resonance were preliminarily selected through the simulation. To obtain the amplitude-frequency characteristics of the arc voltage, the experiments were carried out in C-L-R circuit. The arc currents in the experiments were 490 and 800 A, respectively. An external TMF of 24 mT was provided by a pair of Helmhortz coils. The experimental results showed that, in the unstable stage, the frequency with the largest amplitude of the arc voltage was appeared between 100 kHz and 500 kHz and was unfixed. When the frequency was greater than 500 kHz, the amplitude decayed rapidly. According to the simulation, the parameters of the parallel branch was determined. The simulation results show that the characteristic frequency of L-C value around 200 kHz is appropriate, and the magnitude of the amplitude is negatively correlated with the quality factor of the L-C.

The objective of this work intends to propose a novel current interruption technology in vacuum for LVDC power systems, based on the instability of vacuum arc under composite transverse magnetic fields (TMFs). The interruption characteristics were experimentally determined in an overdamping C-L-R circuit. The experimental results indicated that the DC current of 800 V/500 A could be successfully interrupted within 5 ms by the proposed technology. Moreover, the interruption process could be divided into two stages, the Stable and Unstable stages. In the Stable stage, the arc voltage linearly increased from 20 to 35 V, where the arc current was limited slightly. In the Unstable stage, the arc voltage would rapidly increase and exceed the supply voltage, where the current would be forced to zero. Finally, the double-break LVDC vacuum switches, no matter in series or parallel, were difficult to improve the interruption capability, because of the asynchronous occurrence of the Unstable stage in series vacuum interrupters and the fast current-commutation between parallel vacuum interrupters. This technology takes advantages of high interruption capability, short arcing time and low complexity, which is a promising solution to expand the application of vacuum switch to the LVDC power systems.

The application of transfer learning in the fields of computer vision and natural language processing has been successful. It turns out that pre-training some parameters in the network with unsupervised tasks related to downstream tasks is very useful for subsequent task model training. This approach can reduce the dependence of model training on labeled data to a certain extent, which is beneficial to the application of deep learning algorithms. This paper mainly proposes a bearing fault diagnosis model with a small number of labeled samples. The model is trained through pre-training and fine-tuning. It can use a large amount of unlabeled data to obtain sufficient data representation. Finally, fine-tuning is achieved on small sample fault sets Accuracy. At the same time, several optimizations are proposed on the dassic AE fault diagnosis method, and the effectiveness of the optimization content is verifiedin the experimental stage.