Finding substitutes for sulfur hexafluoride (SF 6 ), a gas with extremely high global warming potential, has been a persistent effort for years in the field of high voltage power equipment, which focuses on the evaluation of the electrical strength and boiling temperature for the practical purpose. Following up the previous proposed linear regression models, this work introduces machine learning algorithms including artificial neural network (ANN) and random forest (RF) as the potential approaches to predict the electrical strength and boiling temperature. Based on a series of descriptors derived from the molecular structure of 74 molecules, the performance of three different methods: multiple linear regression, artificial neural network and random forest are compared and assessed in terms of the sensitivity to the sample size, prediction accuracy and stability, and the interpretability of predictors. Considering the available data are limited, random forest shows superior performance with higher robustness and efficiency. The same approaches were applied to the boiling temperature and random forest produced better results as well. Besides, the variable importance ranked by RF improves understanding of the correlation between the molecular properties and electrical strength. It provides important insights to analyze the properties of the SF 6 substitutes during the design and synthesis of the new eco-friendly gases in power equipment.
5 September 2019 in Journal of Physics D: Applied Physics
This research concentrates on the erosion mechanism of the W 70 Cu 30 anode in DC gaseous arcs. An in situ diagnostic system including the measurement of the anode surface temperature and the fast-imaging of the erosion process was developed. The influence of gas media He, N 2 , H 2 , Ar and current, varying from 50 A to 400 A, was investigated. Two erosion modes were observed, namely an evaporation-dominated mode and a splash-dominated mode. The results indicate that the detachment of the splashing droplet is driven by the explosion of surface bubbles, and it is the surface temperature of the anode that determines the size of the surface bubble as well as the splashing erosion rate. Moreover, the prior vaporization of Cu dispersed in the W 70 Cu 30 anode limits the fusion of the W skeleton but promotes the generation of molten ejections by the explosions of Cu vapor bubbles when the topmost Cu is absent. The large difference between W and Cu in melting and boiling temperatures controls the formation of the molten pool on the W 70 Cu 30 anode and leads to the transition of the two erosion modes. The experimental results contribute to a fundamental understanding of the interaction between thermal plasmas and materials.
7 November 2024 in Journal of Physics D: Applied Physics
C4F7N/N2 and C4F7N/CO2 gas mixtures are considered potential alternative gases to SF6, and it is of particular significance to investigate the plasma decay process in these mixtures for evaluating their circuit breaker breaking performance. To comprehend the decay process of electron density(ne) in arc plasma within C4F7N/N2 and C4F7N/CO2 gas mixtures, an arc-generating circuit and a Thomson scattering experiment platform were established. Through coherent Thomson scattering diagnosis of gas arc plasma under various gas mixture conditions, a series of electron density results over time and space were obtained. The findings suggest that the initial electron density of the arc plasma diminishes with an increase in the proportion of C4F7N, and electron density decay is further accelerated as the proportion of C4F7N increases. Moreover, it was observed that the electron density decay rate is higher in the C4F7N/N2 mixture compared to the C4F7N/CO2 mixture. Notably, the electron decay rate in a 50% C4F7N/N2 mixture closely resembles that observed in pure SF6.
Microparticles are an inevitable component of vacuum arcs. This paper focused on the dynamics of microparticle and erosion behavior in high-current vacuum arcs. The arc was sustained with AC 110 Hz between two butt CuCr alloy contacts with a diameter of 10 mm. An in situ microparticle diagnostic system suitable for vacuum arcs was developed. Three sources of microparticles were observed and the corresponding motion characteristics were obtained by our method. The influence of current and Cr content on erosion characteristics was investigated. The results indicate that the increase of Cr content contributes to inhibiting production of microparticles when the current is small. CuCr30 has the best effect of suppressing microparticle production among the three tested alloy materials under large current. The work of this paper helps with the understanding of microparticle behavior in high-current vacuum arcs and may provide a path for the design of a vacuum electrode.
14 December 2018 in Journal of Physics D: Applied Physics
The arc behaviors of different gas media, including air, N 2 and O 2 , during the current-zero period in the circuit breaker were investigated based on a non-chemically equilibrium (non-CE) model. The arc behaviors during the decay phase by the non-CE model were compared with those by the local thermally equilibrium (LTE) model. In the LTE model, since the thermodynamic and transport properties of air are quite similar to those of N 2 , the results indicate that the arc behaviors in these two gases are quite similar. On the contrary, the results by the non-CE model indicate that the arc behaviors of the air could be quite different from those of N 2 , which is caused by the different reaction processes due to the existence of both the N and O element in the air. It can be concluded that during the current-zero, the reaction process greatly influences the arc decaying behaviors, which cannot be considered in the convectional LTE model. Therefore, the non-CE model is very important to predict the arc behavior during the arc decay phase, and the validity of the non-CE model was further confirmed by the laser Thomson scattering in different gas media.