Scholarlink

<p>Preventing overheating of critical components in oil-cooled permanent magnet synchronous motors (PMSMs) has emerged as a significant area of current research. The complex structure inherent to oil-cooled motors presents challenges in accurately simulating peak motor temperatures. Traditional lumped-parameter thermal network (LPTN) methods are often inadequate for these oil-cooled applications. To address this issue, an enhanced lumped-parameter thermal network model is proposed. This model leverages the relationships between thermal resistance, coolant flow rate, contact area, and temperature to establish and validate a formula for calculating thermal resistance, enabling temperature simulation under both steady-state and transient conditions. Furthermore, a multi-objective optimization algorithm is employed to refine the parameters within the thermal resistance calculation formula, thereby improving the accuracy of the model's simulations. The optimized model accurately predicts temperature under steady-state conditions, with an average error of 5.33% for stator temperature and 6.24% for winding temperature. In transient conditions, the model adequately predicted peak temperatures, meeting the needs of the cooling control system. Under world light vehicle test cycle (WLTC) conditions, the average winding temperature error was 10.85%, and the peak temperature error was 1.8 °C.</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

Improved thermal network model for oil-cooled PMSM based on parameter optimization

  • 期刊论文

  • 发表日期：

    2026年2月16日

查看全文
  • Liange He,
  • Song He,
  • Yan Zhang,
  • Boru Jia,
  • Limin Wu,
  • Zijun Ma

摘要

Preventing overheating of critical components in oil-cooled permanent magnet synchronous motors (PMSMs) has emerged as a significant area of current research. The complex structure inherent to oil-cooled motors presents challenges in accurately simulating peak motor temperatures. Traditional lumped-parameter thermal network (LPTN) methods are often inadequate for these oil-cooled applications. To address this issue, an enhanced lumped-parameter thermal network model is proposed. This model leverages the relationships between thermal resistance, coolant flow rate, contact area, and temperature to establish and validate a formula for calculating thermal resistance, enabling temperature simulation under both steady-state and transient conditions. Furthermore, a multi-objective optimization algorithm is employed to refine the parameters within the thermal resistance calculation formula, thereby improving the accuracy of the model's simulations. The optimized model accurately predicts temperature under steady-state conditions, with an average error of 5.33% for stator temperature and 6.24% for winding temperature. In transient conditions, the model adequately predicted peak temperatures, meeting the needs of the cooling control system. Under world light vehicle test cycle (WLTC) conditions, the average winding temperature error was 10.85%, and the peak temperature error was 1.8 °C.