Scholarlink

<p>This article provides an analytical review of contemporary research on the application of artificial intelligence (AI) for monitoring, analyzing, and automated control of hybrid laser-plasma welding (HLPW). The nonlinear interaction of laser radiation and plasma arc forms a synergistic energy coupling, which intensifies the requirements for process control and constrains the use of traditional single-channel monitoring methods. The key parameters critical to the stability of HLPW and weld seam quality have been systematically formulated: weld pool geometry, keyhole behavior, temperature field, arc current and voltage, gas-dynamic conditions, acoustic emission, and parameters of laser radiation. Sensor channels applicable for monitoring HLPW have been summarized, including high-speed color (RGB) and infrared (IR) cameras, pyrometers, acoustic sensors, video spectroscopic instruments, as well as sensors for current, voltage, gas flow rate, and gas pressure. Their applications are examined with regard to plasma radiation, spatter, optical glare, and the requirement for signal synchronization. Neural network approaches are analyzed according to data type and task: convolutional models for visual data, recurrent models for temporal sequences, and physics-informed as well as hybrid models for integrating experimental data with process physics. This work examines implementations of closed-loop real-time control and the ongoing advancement of multimodal AI monitoring systems.</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

Analysis of modern solutions for the application of artificial intelligence in the development of automated control systems for hybrid laser-plasma welding processes

  • 期刊论文

  • 发表日期：

    2026年7月10日

查看全文
  • Volodymyr Korzhyk,
  • Andrii Kvasnytskyi,
  • Yunqiang Zhao,
  • Oleksandr Voitenko,
  • Zhe Liu,
  • Vladyslav Khaskin,
  • Oksana Konoreva,
  • Oleksandr Bushma,
  • Oleksandr Bozhok

摘要

This article provides an analytical review of contemporary research on the application of artificial intelligence (AI) for monitoring, analyzing, and automated control of hybrid laser-plasma welding (HLPW). The nonlinear interaction of laser radiation and plasma arc forms a synergistic energy coupling, which intensifies the requirements for process control and constrains the use of traditional single-channel monitoring methods. The key parameters critical to the stability of HLPW and weld seam quality have been systematically formulated: weld pool geometry, keyhole behavior, temperature field, arc current and voltage, gas-dynamic conditions, acoustic emission, and parameters of laser radiation. Sensor channels applicable for monitoring HLPW have been summarized, including high-speed color (RGB) and infrared (IR) cameras, pyrometers, acoustic sensors, video spectroscopic instruments, as well as sensors for current, voltage, gas flow rate, and gas pressure. Their applications are examined with regard to plasma radiation, spatter, optical glare, and the requirement for signal synchronization. Neural network approaches are analyzed according to data type and task: convolutional models for visual data, recurrent models for temporal sequences, and physics-informed as well as hybrid models for integrating experimental data with process physics. This work examines implementations of closed-loop real-time control and the ongoing advancement of multimodal AI monitoring systems.