Scholarlink

<p>Accurate identification of current transformer (CT) error states in distribution networks is critical for grid monitoring and ensuring fair power trade. This paper proposes an imbalance weight-based approach for CT error state identification to address the limitations of the multivariate linear regression coefficient method in handling overlapping error characteristics caused by irregular current fluctuations under dynamic loads. This approach constructs a complex multivariate linear regression equation constrained by Kirchhoff's current law, leveraging physical correlations within the group and secondary output values to solve for characteristic quantities representing individual CT error states. Building upon this, an imbalance weight feature quantifying CT error drift is constructed using a traversal strategy and Manhattan distance. By analyzing the fluctuation patterns of imbalance weight features, the proposed method accurately discriminates CT error states. Tested on a 10&#xa0;kV distribution system, the results show that both the proposed method and the traditional multivariate linear regression approach achieve above 97% identification accuracy under conditions with small current fluctuations and a single faulty CT. However, the proposed method maintains stable identification performance even under complex scenarios with significant current fluctuations, demonstrating its greater stability and robustness.</p>

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

Method for identifying error states of distribution network current transformers based on current imbalance weights

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月27日

查看全文
  • Tong Liu,
  • Xiujuan Zeng,
  • Huiqin Xie,
  • Wei Wu,
  • Jihong Xiao

摘要

Accurate identification of current transformer (CT) error states in distribution networks is critical for grid monitoring and ensuring fair power trade. This paper proposes an imbalance weight-based approach for CT error state identification to address the limitations of the multivariate linear regression coefficient method in handling overlapping error characteristics caused by irregular current fluctuations under dynamic loads. This approach constructs a complex multivariate linear regression equation constrained by Kirchhoff's current law, leveraging physical correlations within the group and secondary output values to solve for characteristic quantities representing individual CT error states. Building upon this, an imbalance weight feature quantifying CT error drift is constructed using a traversal strategy and Manhattan distance. By analyzing the fluctuation patterns of imbalance weight features, the proposed method accurately discriminates CT error states. Tested on a 10 kV distribution system, the results show that both the proposed method and the traditional multivariate linear regression approach achieve above 97% identification accuracy under conditions with small current fluctuations and a single faulty CT. However, the proposed method maintains stable identification performance even under complex scenarios with significant current fluctuations, demonstrating its greater stability and robustness.