Scholarlink

In the context of Industry 4.0, new ways of manufacturing, monitoring and data generation related to industrial processes came to light. A new form of parts manufacturing that has been revolutionizing the industry is Additive Manufacturing in many forms, such as the more traditional Fused Deposition Modeling (FDM), or the more innovative ones, such as Laser Metal Deposition (LMD) or Gas Metal Arc Welding (GMAW). New technologies related to monitoring such processes are also emerging, such as Cyber-Physical Systems (CPS) or Digital Twins (DT), which can also be used to enable Artificial Intelligence (AI) powered analysis of generated big data. However, few studies have enabled a comprehensive data analysis based on Digital Twin systems to study the tolerances of manufactured parts using 3D models. With this background in mind, this project uses a Digital Twin-enabled dataflow to constitute a basis for a proposed data analysis pipeline. The pipeline analyzes metal AM-manufactured parts’ tolerances by comparing 3D models obtained by LiDAR and grids of images scans with the positional data of the manufacturing process stored in a cloud database. 3D models of parts conceived and planned for production via CAD/CAM are compared with the real parts’ 3D meshes generated. Stored and analyzed data may be further used to refine the manufacturing of parts, calibration of sensors, and refining of the DT model.

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

Digital Twin-Enabled Quality Assurance Analysis of Metal Manufactured Parts Based on Neural Networks Applied to 3D Meshes

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • João Vítor Arantes Cabral,
  • Antonio Carlos da Cunha Facciolli,
  • Guilherme Caribé de Carvalho,
  • Alberto José Álvares

摘要

In the context of Industry 4.0, new ways of manufacturing, monitoring and data generation related to industrial processes came to light. A new form of parts manufacturing that has been revolutionizing the industry is Additive Manufacturing in many forms, such as the more traditional Fused Deposition Modeling (FDM), or the more innovative ones, such as Laser Metal Deposition (LMD) or Gas Metal Arc Welding (GMAW). New technologies related to monitoring such processes are also emerging, such as Cyber-Physical Systems (CPS) or Digital Twins (DT), which can also be used to enable Artificial Intelligence (AI) powered analysis of generated big data. However, few studies have enabled a comprehensive data analysis based on Digital Twin systems to study the tolerances of manufactured parts using 3D models. With this background in mind, this project uses a Digital Twin-enabled dataflow to constitute a basis for a proposed data analysis pipeline. The pipeline analyzes metal AM-manufactured parts’ tolerances by comparing 3D models obtained by LiDAR and grids of images scans with the positional data of the manufacturing process stored in a cloud database. 3D models of parts conceived and planned for production via CAD/CAM are compared with the real parts’ 3D meshes generated. Stored and analyzed data may be further used to refine the manufacturing of parts, calibration of sensors, and refining of the DT model.