Scholarlink

As an important measurement tool in the power system, the accuracy of smart energy meters directly affects the fairness of power measurement and the economic benefits of power enterprises. However, in practical application, the problem of inaccurate assessment of power meter error occurs from time to time, which brings challenges to the stable operation of the power system and fair trading in the power market. Aiming at the current problem of inaccurate power meter error assessment, this paper proposes an improved subtraction average based optimizer and Extreme Learning Machines (ISABO-ELM) power meter error assessment. The method introduces Spatial Pyramid Matching (spm) chaotic mapping and adaptive T-distribution variation strategy, which improves the algorithm’s global search capability and the ability to jump out of local optima. The improved method is used to optimize the ELM algorithm and used to evaluate the energy meter error, and the results demonstrate a significant improvement in evaluation accuracy, achieving a 22.57% increase compared to the Gray Wolf Optimization Algorithm Optimized Extreme Learning Machine (GWO-ELM) and a 15.73% increase compared to the Subtractive Averaging Optimization Algorithm Optimized Extreme Learning Machine (SABO-ELM).

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

Error Compensation Method for Electric Energy Meters Based on Improved Subtraction Average Based Optimizer and Extreme Learning Machines

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Jianli Li,
  • Ling Zhang,
  • Ma Luo,
  • Aoran Pan,
  • Wenpeng Mao

摘要

As an important measurement tool in the power system, the accuracy of smart energy meters directly affects the fairness of power measurement and the economic benefits of power enterprises. However, in practical application, the problem of inaccurate assessment of power meter error occurs from time to time, which brings challenges to the stable operation of the power system and fair trading in the power market. Aiming at the current problem of inaccurate power meter error assessment, this paper proposes an improved subtraction average based optimizer and Extreme Learning Machines (ISABO-ELM) power meter error assessment. The method introduces Spatial Pyramid Matching (spm) chaotic mapping and adaptive T-distribution variation strategy, which improves the algorithm’s global search capability and the ability to jump out of local optima. The improved method is used to optimize the ELM algorithm and used to evaluate the energy meter error, and the results demonstrate a significant improvement in evaluation accuracy, achieving a 22.57% increase compared to the Gray Wolf Optimization Algorithm Optimized Extreme Learning Machine (GWO-ELM) and a 15.73% increase compared to the Subtractive Averaging Optimization Algorithm Optimized Extreme Learning Machine (SABO-ELM).