Scholarlink

<p>This study explores the impact of an innovative three-wire fusion nozzle electroslag welding (FNESW) technique on the microstructural evolution and tensile properties of U75V pearlitic steel rail weld joints. An intelligent monitoring system was developed to systematically capture critical welding parameters, including current, voltage, cooling rate, and magnetic field intensity. Furthermore, a Back Propagation (BP) neural network model was designed and trained to predict the microstructural features and mechanical properties of the welded joints. The model exhibited robust predictive performance, effectively establishing the quantitative relationship between welding parameters and joint performance. Experimental validation corroborated the model’s reliability, with relative errors of key predictive indicators maintained below 15%. The findings provide a scientific basis for optimizing welding parameters and designing high-performance steel rail weld joints through the integration of machine learning techniques, offering new insights into the intelligent control of welding processes.</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

Monitoring process parameters and predicting rail steel welded joint microstructure and mechanical property of three-wire fusion nozzle electroslag welding

  • 期刊论文

  • 发表日期：

    2025年2月14日

查看全文
  • Shengfu Yu,
  • Yang Wang,
  • Zhongyi Zhang

摘要

This study explores the impact of an innovative three-wire fusion nozzle electroslag welding (FNESW) technique on the microstructural evolution and tensile properties of U75V pearlitic steel rail weld joints. An intelligent monitoring system was developed to systematically capture critical welding parameters, including current, voltage, cooling rate, and magnetic field intensity. Furthermore, a Back Propagation (BP) neural network model was designed and trained to predict the microstructural features and mechanical properties of the welded joints. The model exhibited robust predictive performance, effectively establishing the quantitative relationship between welding parameters and joint performance. Experimental validation corroborated the model’s reliability, with relative errors of key predictive indicators maintained below 15%. The findings provide a scientific basis for optimizing welding parameters and designing high-performance steel rail weld joints through the integration of machine learning techniques, offering new insights into the intelligent control of welding processes.