The importance of vacuum breakdown mechanism during impulse voltage aging

Author： GOZ Electric Time：2024-08-24 09:35:20 Read：15

    Impulse voltage aging technology is an effective means to improve the insulation capacity of vacuum electrode gaps. Rapid and accurate identification of vacuum breakdown mechanisms is of great significance for revealing the physical evolution mechanism of the impulse voltage aging process. This paper proposes an optimization method based on deep learning to improve the accuracy of vacuum breakdown mechanism identification by highlighting the pre-breakdown process. The same impulse voltage aging test was carried out on five pairs of identical oxygen-free copper spherical electrodes, and the complete breakdown voltage and current waveform samples with a time extraction range of 0-400 μs and the breakdown voltage and current waveform samples with a prominent pre-breakdown process of 0-200 μs were obtained. The deep learning model was used to train and test the two types of breakdown voltage and current waveform samples for the three vacuum breakdown mechanisms induced by pulse current, field emission and particles, and the test results were compared with the actual results for analysis and evaluation. The results show that the breakdown mechanism identification accuracy of the breakdown voltage and current waveform samples of the pre-breakdown process with a time extraction range of 0 to 200 μs is above 87.99%, an average increase of 2.55%, and the corresponding precision, recall and F1 scores are better. The research results show that the processing of the breakdown voltage and current waveform of the pre-breakdown process can effectively optimize the effect of vacuum breakdown mechanism identification and has good engineering application prospects.

    At present, SF6 gas circuit breakers occupy a major market share in high-voltage lines of 72.5 kV and above with excellent and reliable insulation and arc extinguishing capabilities. However, SF6 gas has a strong greenhouse effect, and its global warming coefficient is 22200 to 23900 times that of CO2. Large-scale emission of SF6 gas will aggravate global climate change. In addition, the decomposition products of SF6 gas are highly acidic, which can destroy the ozone layer, lead to the formation of ozone holes, and cause serious harm to the ecosystem. Under the goal of "carbon peak and carbon neutrality" in my country, it has become an inevitable trend to develop environmentally friendly high-voltage circuit breakers to replace SF6 gas circuit breakers. Vacuum circuit breakers are power equipment that uses vacuum as an arc extinguishing medium. They have the advantages of being economical, environmentally friendly, compact, reliable, and fast switching. They are currently widely used in medium and low voltage lines. However, the development of vacuum circuit breakers to high voltage levels is still limited by the insulation capacity of the vacuum electrode gap. Impulse voltage aging technology is a common aging technology. By applying an impulse voltage with a certain frequency, amplitude, and pulse width to remove impurities, residual gas, and processing burrs on the electrode surface, the electrode surface condition can be improved, and the insulation capacity of the vacuum electrode gap can be significantly improved. Therefore, studying the impulse voltage aging characteristics is of great significance for improving the voltage level of vacuum circuit breakers.

    Breakdown voltage is an important parameter for studying the impulse voltage aging characteristics. The relationship between the breakdown voltage and the applied voltage can reflect the aging change process. In the aging saturation stage, the breakdown voltage follows the Weibull distribution, and the aging saturation voltage can be characterized by the 50% breakdown voltage U50 [15-16]. Therefore, a method of judging aging saturation by breakdown voltage and breakdown time was proposed. However, the breakdown voltage cannot reflect the vacuum breakdown mechanism at different stages of aging, and factors such as electrode material properties, electrode surface roughness, and applied voltage frequency and amplitude will affect the breakdown characteristics. Therefore, the impulse voltage aging characteristics cannot be fully revealed only by the breakdown voltage and breakdown time.

    The vacuum breakdown mechanism can describe the evolution of the impulse voltage aging process. Rapidly and accurately identifying the vacuum breakdown mechanism is of great significance for studying the impulse voltage aging characteristics and revealing its physical evolution mechanism. Li Shimin et al. conducted an impulse voltage aging test on a pure copper ball electrode and verified that the breakdown mechanism in the impulse voltage aging process includes three vacuum breakdown mechanisms: pulse current induced breakdown (PB), field emission induced breakdown (FEBD), and particle induced breakdown (PBD). Electrodes of different shapes have a consistent breakdown mechanism for their gaps. The vacuum breakdown mechanism can be identified by the voltage and current characteristics of the pre-breakdown process, but this method requires the displacement current to be eliminated through a mathematical compensation algorithm, and includes steps such as the Fowler-Nordheim formula fitting. The calculation process is cumbersome and complicated, and can only be obtained through data calculation and analysis after the experiment.

    At present, with the rapid development of artificial intelligence, the accuracy and speed of image recognition have made a qualitative leap, and deep learning has been applied in research fields such as fault detection, instrument identification, and pattern recognition in electrical disciplines. Li Shimin et al. [31] extracted and identified different pre-breakdown and breakdown characteristics of different vacuum breakdown mechanisms through deep learning technology, and realized the classification of vacuum breakdown mechanisms during impulse voltage aging. However, there are still a lot of research gaps in the application of deep learning technology in the field of vacuum breakdown, which requires further in-depth exploration. Therefore, based on the relevant literature, this paper proposes a deep learning-based optimization method for identifying vacuum breakdown mechanisms during impulse voltage aging by highlighting the pre-breakdown process of the breakdown voltage and current waveforms.