Scholarlink

<p>Brushless AC excitation machine (BLAC) is widely used in power generation systems within flammable, explosive, and harsh environments. However, its internal temperature distribution is complex, making thermal analysis challenging. Traditional modeling methods suffer from low computational efficiency and insufficient accuracy. To address the challenges of complex end-winding modeling and high computational resource consumption in excitation machine thermal analysis, this paper proposes an end-winding modeling method based on geometric equivalence of arc length. To overcome the issues of excessive mesh count and prolonged solution time in traditional 3D models, the arc length of the end-winding is geometrically equivalenced to a simplified straight segment while preserving actual electrical path length; to accurately calculate electromagnetic losses, a copper loss and iron loss model accounting for the skin effect and harmonic influences was adopted. Ultimately, while maintaining accuracy, the computational cost was significantly reduced, achieving a maximum temperature rise error not exceeding 5&#xa0;°C under rated, 90, and 110% voltage conditions. This method was validated using finite element analysis (FEA) and a 40-kW excitation machine experimental platform, with results demonstrating both high efficiency and accuracy.</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

A simplified end-winding modeling method for thermal analysis of brushless AC excitation machines

  • 期刊论文

  • 发表日期：

    2026年3月10日

查看全文
  • Junxiang Li,
  • Xiaohua Bao,
  • Zixuan Sun,
  • Kang Tian

摘要

Brushless AC excitation machine (BLAC) is widely used in power generation systems within flammable, explosive, and harsh environments. However, its internal temperature distribution is complex, making thermal analysis challenging. Traditional modeling methods suffer from low computational efficiency and insufficient accuracy. To address the challenges of complex end-winding modeling and high computational resource consumption in excitation machine thermal analysis, this paper proposes an end-winding modeling method based on geometric equivalence of arc length. To overcome the issues of excessive mesh count and prolonged solution time in traditional 3D models, the arc length of the end-winding is geometrically equivalenced to a simplified straight segment while preserving actual electrical path length; to accurately calculate electromagnetic losses, a copper loss and iron loss model accounting for the skin effect and harmonic influences was adopted. Ultimately, while maintaining accuracy, the computational cost was significantly reduced, achieving a maximum temperature rise error not exceeding 5 °C under rated, 90, and 110% voltage conditions. This method was validated using finite element analysis (FEA) and a 40-kW excitation machine experimental platform, with results demonstrating both high efficiency and accuracy.