Scholarlink

<p>Large-scale engineering structures, such as truss structures, steel-concrete-steel structures, and lattice structures, utilize metal bars as fundamental building units. These structures offer advantages in lightweight design, high strength, energy absorption, and multifunctionality, making them highly promising for applications in aerospace, marine, and construction. Wire and arc additive manufacturing (WAAM) has been proven to be a powerful technique for fabricating such complex spatial bar structures. However, real-time detection of the unsupported bar deposition process and precise adjustment of manufacturing parameters remain challenging due to the lack of low-cost, efficient, and accurate sensing and control methods. Currently, WAAM process of unsupported bars currently relies heavily on engineers’ experience, leading to low fabrication precision, poor geometric consistency, and limited automation. To address this challenge, this study develops an image processing method based on vision sensing technology, enabling real-time and accurate extraction of arc length during the deposition of unsupported bars. Based on the detected arc length, an adaptive fuzzy logic-based arc length control strategy is proposed to enhance deposition stability. Experimental results demonstrate that the proposed vision-based arc length detection system achieves a maximum error of only 0.09&#xa0;mm. By adjusting the arc duty cycle through adaptive fuzzy inference, the control system effectively compensates for deviations between the actual and target arc lengths. During the deposition of 45° and 90° unsupported bars, the mean absolute error of arc length were 0.1&#xa0;mm and 0.07&#xa0;mm, respectively, confirming the effectiveness of the control strategy in improving process stability. Additionally, maintaining a stable arc length significantly enhances the forming accuracy of unsupported bars by reducing cross-sectional fluctuations and minimizing geometric center displacement. In the arc length tracking experiment, when subjected to a step change in the arc length setpoint, the controller successfully corrected the arc length within 10 pulse cycles, achieving a smooth and rapid transition to a new stable state. This validates the controller’s stability and responsiveness. This study provides a promising solution for the real-time detection and control of unsupported bar deposition process in GTAW-based WAAM, facilitating the broader application of WAAM in engineering structures where metal bars serve as fundamental structural elements.</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

Research on arc length control of unsupported bars in GTAW-based wire and arc additive manufacturing via vision sensing and adaptive fuzzy control

  • 期刊论文

  • 发表日期：

    2025年12月4日

查看全文
  • Hao Mao,
  • Fan Zhang,
  • Haoran Fu,
  • Junjin Huang,
  • Hanwen Xu,
  • Zhongtian Luo,
  • Jiawei Xie,
  • Tianqiu Xu,
  • Xue Ling,
  • Yunze Li,
  • Changmeng Liu

摘要

Large-scale engineering structures, such as truss structures, steel-concrete-steel structures, and lattice structures, utilize metal bars as fundamental building units. These structures offer advantages in lightweight design, high strength, energy absorption, and multifunctionality, making them highly promising for applications in aerospace, marine, and construction. Wire and arc additive manufacturing (WAAM) has been proven to be a powerful technique for fabricating such complex spatial bar structures. However, real-time detection of the unsupported bar deposition process and precise adjustment of manufacturing parameters remain challenging due to the lack of low-cost, efficient, and accurate sensing and control methods. Currently, WAAM process of unsupported bars currently relies heavily on engineers’ experience, leading to low fabrication precision, poor geometric consistency, and limited automation. To address this challenge, this study develops an image processing method based on vision sensing technology, enabling real-time and accurate extraction of arc length during the deposition of unsupported bars. Based on the detected arc length, an adaptive fuzzy logic-based arc length control strategy is proposed to enhance deposition stability. Experimental results demonstrate that the proposed vision-based arc length detection system achieves a maximum error of only 0.09 mm. By adjusting the arc duty cycle through adaptive fuzzy inference, the control system effectively compensates for deviations between the actual and target arc lengths. During the deposition of 45° and 90° unsupported bars, the mean absolute error of arc length were 0.1 mm and 0.07 mm, respectively, confirming the effectiveness of the control strategy in improving process stability. Additionally, maintaining a stable arc length significantly enhances the forming accuracy of unsupported bars by reducing cross-sectional fluctuations and minimizing geometric center displacement. In the arc length tracking experiment, when subjected to a step change in the arc length setpoint, the controller successfully corrected the arc length within 10 pulse cycles, achieving a smooth and rapid transition to a new stable state. This validates the controller’s stability and responsiveness. This study provides a promising solution for the real-time detection and control of unsupported bar deposition process in GTAW-based WAAM, facilitating the broader application of WAAM in engineering structures where metal bars serve as fundamental structural elements.