Scholarlink

To improve the analysis of DC series arc faults, this paper presents a data-driven surrogate model based on 1-D arc decaying model. Firstly, both the stationary and transient arc decaying models are solved by the finite volume method, and then a data-driven relationship is established between arc conductance and arcing time, as the surrogate arc model used in circuit simulation. To describe DC series arc characteristics, three distinct surrogate arc models, namely steady-state, time-independent, and time-dependent arc models, are developed by optimizing parameters such as current ranges, sampling intervals, and iteration counts. The results show that, for scenarios requiring lower computational precision, either the steady-state or time-independent models can be used, whereas for high-precision applications, the time-dependent model is preferred.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data-Driven Surrogate Model for DC Series Arc Fault Simulation

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xinxiang Zhou,
  • Yi Hao,
  • Baichuan Shen,
  • Pengda Li,
  • Linlin Zhong

摘要

To improve the analysis of DC series arc faults, this paper presents a data-driven surrogate model based on 1-D arc decaying model. Firstly, both the stationary and transient arc decaying models are solved by the finite volume method, and then a data-driven relationship is established between arc conductance and arcing time, as the surrogate arc model used in circuit simulation. To describe DC series arc characteristics, three distinct surrogate arc models, namely steady-state, time-independent, and time-dependent arc models, are developed by optimizing parameters such as current ranges, sampling intervals, and iteration counts. The results show that, for scenarios requiring lower computational precision, either the steady-state or time-independent models can be used, whereas for high-precision applications, the time-dependent model is preferred.