Scholarlink

<p>This study investigates the aerodynamic characteristics of crescent-shaped iced conductors through combined wind tunnel experiments and numerical analysis. Wind tunnel tests were conducted to obtain both steady and unsteady aerodynamic coefficients, the latter measured using an oscillating conductor model. The steady coefficients were acquired under various wind velocities ranging from 10 to 18&#xa0;m/s and ice thicknesses of 14, 24 and 33&#xa0;mm, covering trans-critical and super-critical Reynolds number regimes. Analytical expressions relating steady coefficients to the angle of attack were derived via polynomial fitting. A BP neural network model was subsequently developed and trained on this experimental data to predict aerodynamic coefficients for untested conditions. Furthermore, a MATLAB-based quasi-static model was established to convert steady coefficients into unsteady equivalents, enabling a direct comparison with the unsteady oscillation test data to validate the quasi-static assumption. The results confirm that the quasi-static assumption retains reasonable applicability in the studied regime. More importantly, the developed BP neural network demonstrates robust predictive capability, with its outputs showing close agreement with wind tunnel measurements, achieving determination coefficients exceeding 0.95. This work provides enhanced analytical tools and validated data for improving the prediction of conductor galloping—a critical wind-induced vibration—thereby contributing to the safety assessment and structural design of overhead transmission lines.</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

Investigation on the unsteady aerodynamic coefficients of iced conductors and the applicability of quasi-static assumptions

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月21日

查看全文
  • Guanghui Liu,
  • Zhongbin Lv,
  • Bo Zhang,
  • Chuan Wu,
  • Yuhang Nie,
  • Haobo Liang

摘要

This study investigates the aerodynamic characteristics of crescent-shaped iced conductors through combined wind tunnel experiments and numerical analysis. Wind tunnel tests were conducted to obtain both steady and unsteady aerodynamic coefficients, the latter measured using an oscillating conductor model. The steady coefficients were acquired under various wind velocities ranging from 10 to 18 m/s and ice thicknesses of 14, 24 and 33 mm, covering trans-critical and super-critical Reynolds number regimes. Analytical expressions relating steady coefficients to the angle of attack were derived via polynomial fitting. A BP neural network model was subsequently developed and trained on this experimental data to predict aerodynamic coefficients for untested conditions. Furthermore, a MATLAB-based quasi-static model was established to convert steady coefficients into unsteady equivalents, enabling a direct comparison with the unsteady oscillation test data to validate the quasi-static assumption. The results confirm that the quasi-static assumption retains reasonable applicability in the studied regime. More importantly, the developed BP neural network demonstrates robust predictive capability, with its outputs showing close agreement with wind tunnel measurements, achieving determination coefficients exceeding 0.95. This work provides enhanced analytical tools and validated data for improving the prediction of conductor galloping—a critical wind-induced vibration—thereby contributing to the safety assessment and structural design of overhead transmission lines.