Scholarlink

<p>The printing sequence of islands and stripes (i.e., printing entities) in Laser Powder Bed Fusion (LPBF) plays a critical role in determining the temperature distribution within each printed layer. This thermal distribution directly governs the development of thermally induced residual stress and the resulting geometric distortion within the layer and, ultimately, the manufactured part. Intelligent in situ optimization of the printing sequence therefore presents a promising approach for mitigating these thermal effects and improving part quality. Building on the NOTCH (Novel Online Thermography and Closed-loop Hybrid-control) framework introduced in our previous work, this paper presents SmartLayer, a physics-based optimization module that introduces two novel scanning strategies for sequence planning in LPBF layers with arbitrary size and geometry. The first strategy employs a greedy optimization approach, in which the next printing entity is selected at each step to maximize thermal uniformity without considering long-term thermal evolution. The second strategy adopts an evolutionary optimization approach to solve the combinatorial optimization problem of printing sequence planning, generating a complete printing sequence that minimizes thermal non-uniformity across the entire layer. The proposed methods are experimentally validated on a printed plate consisting of 100 printing entities, with performance evaluated in terms of thermally induced residual stress and distortion. The results are compared against two widely used commercial scanning strategies: successive scanning and successive chessboard scanning. The findings demonstrate that SmartLayer-optimized sequences significantly improve residual stress uniformity and reduce thermally induced distortion. These results highlight the importance of thermally informed sequence planning as a key factor in improving LPBF part quality and dimensional accuracy. Ongoing work focuses on integrating machine learning techniques with SmartLayer to enable real-time, adaptive optimization of printing sequences.</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

Smart control and optimization of the printing sequences in laser powder bed fusion

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月24日

查看全文
  • Ehsan Malekipour,
  • Hazim El-Mounayri,
  • Ala Qattawi,
  • Majed Ali,
  • Abdalmageed Almotari

摘要

The printing sequence of islands and stripes (i.e., printing entities) in Laser Powder Bed Fusion (LPBF) plays a critical role in determining the temperature distribution within each printed layer. This thermal distribution directly governs the development of thermally induced residual stress and the resulting geometric distortion within the layer and, ultimately, the manufactured part. Intelligent in situ optimization of the printing sequence therefore presents a promising approach for mitigating these thermal effects and improving part quality. Building on the NOTCH (Novel Online Thermography and Closed-loop Hybrid-control) framework introduced in our previous work, this paper presents SmartLayer, a physics-based optimization module that introduces two novel scanning strategies for sequence planning in LPBF layers with arbitrary size and geometry. The first strategy employs a greedy optimization approach, in which the next printing entity is selected at each step to maximize thermal uniformity without considering long-term thermal evolution. The second strategy adopts an evolutionary optimization approach to solve the combinatorial optimization problem of printing sequence planning, generating a complete printing sequence that minimizes thermal non-uniformity across the entire layer. The proposed methods are experimentally validated on a printed plate consisting of 100 printing entities, with performance evaluated in terms of thermally induced residual stress and distortion. The results are compared against two widely used commercial scanning strategies: successive scanning and successive chessboard scanning. The findings demonstrate that SmartLayer-optimized sequences significantly improve residual stress uniformity and reduce thermally induced distortion. These results highlight the importance of thermally informed sequence planning as a key factor in improving LPBF part quality and dimensional accuracy. Ongoing work focuses on integrating machine learning techniques with SmartLayer to enable real-time, adaptive optimization of printing sequences.