A Study on internal product placement in heat treatment process using linear regression and object detection†
Hyejung Park1 · Sanggwon Kim2 · Kukhyun Yeo3 · Dongkyu Moon4
1P&S BigData Science Institute
23Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH)
4Department of Artificial Intelligence Convergence, Dankook University
23Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH)
4Department of Artificial Intelligence Convergence, Dankook University
Correspondence to: † This study has been conducted with the support of the Korea Institude of Industrial Technology as “Development if intelligent root technology with add-on modules (KITECH EO-22-00005)”.
1 Director, P&S BigData Science Institute, Daegu 41951, Korea.
2 Chief Researcher, Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH), Gyeonggi 15014, Korea.
3 Senior Researcher, Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH), Gyeonggi 15014, Korea.
4 Corresponding author: Master course, Department of Artificial Intelligence Convergence, Dankook University, Gyeonggi 16890, Korea. E-mail: dkmoon0530@gmail.com
1 Director, P&S BigData Science Institute, Daegu 41951, Korea.
2 Chief Researcher, Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH), Gyeonggi 15014, Korea.
3 Senior Researcher, Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH), Gyeonggi 15014, Korea.
4 Corresponding author: Master course, Department of Artificial Intelligence Convergence, Dankook University, Gyeonggi 16890, Korea. E-mail: dkmoon0530@gmail.com
Received December 7, 2023; Revised December 30, 2023; Accepted January 5, 2024.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
- Abstract
- The heat treatment process in metal casting involves placing metal components on a jig, loading them into a furnace, and heating them uniformly to enhance product quality. However, variations in furnace size and performance lead to uneven internal temperature distribution. Many heat treatment processes rely on operator experience, resulting in uneven heating and defective products. Currently, to account for defects, 10% more products are produced, incurring extra costs. This study aims to improve product quality and reduce costs by using linear regression and object detection models to estimate internal temperature distribution. It includes detecting products on the jig and determining positional accuracy through a 3D virtual space simulation before loading into the furnace. This research suggests a new direction for metal processing companies to optimize production processes and gain economic advantages.
- Keywords : Heat treatment, linear regression, object detection.
Full Text(PDF) Free
