Scholarlink

This study explores the application of linear regression to predict the roughness of the 3D-printed models obtained by FDM technology. Utilizing data from experiments provided freely online, the research analyzes the influence of nine printing parameters on roughness of the printed samples. Results show that layer height exhibits the strongest positive correlation with roughness, while bed temperature and fan speed have a negative impact. The linear regression model, developed using Python's Scikit-learn library, demonstrates high predictive accuracy, with an R2 of 0.922, Mean Absolute Error of 24.01, and Root Mean Squared Error of 27.65. The results provide valuable insights for optimizing 3D printing processes to improve surface quality and reduce production errors, highlighting the role of machine learning in additive manufacturing advancements.

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

Predicting Roughness of the 3D Printed Models Using Linear Regression

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mijodrag Milošević,
  • Mića Đurđev,
  • Dejan Lukić,
  • Grigor Stambolov,
  • Remigiusz Łabudzki,
  • Rafał Świercz

摘要

This study explores the application of linear regression to predict the roughness of the 3D-printed models obtained by FDM technology. Utilizing data from experiments provided freely online, the research analyzes the influence of nine printing parameters on roughness of the printed samples. Results show that layer height exhibits the strongest positive correlation with roughness, while bed temperature and fan speed have a negative impact. The linear regression model, developed using Python's Scikit-learn library, demonstrates high predictive accuracy, with an R2 of 0.922, Mean Absolute Error of 24.01, and Root Mean Squared Error of 27.65. The results provide valuable insights for optimizing 3D printing processes to improve surface quality and reduce production errors, highlighting the role of machine learning in additive manufacturing advancements.