Migration from the traditional to the smart factory in the die-casting industry: Novel process data acquisition and fault detection based on artificial neural network
- Authors
- Lee, Jeongsu; Lee, Young Chul; Kim, Jeong Tae
- Issue Date
- Apr-2021
- Publisher
- ELSEVIER SCIENCE SA
- Keywords
- Die-casting; Fault detection; Smart factory; Industrial data acquisition; Artificial neural network
- Citation
- JOURNAL OF MATERIALS PROCESSING TECHNOLOGY, v.290
- Journal Title
- JOURNAL OF MATERIALS PROCESSING TECHNOLOGY
- Volume
- 290
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/83897
- DOI
- 10.1016/j.jmatprotec.2020.116972
- ISSN
- 0924-0136
- Abstract
- Although die-casting is one of the most popular mass production processes of precise metal parts, the manufacturing environment of the die-casting factory remains at the traditional level. In this study, we developed three core technologies to realize a smart-factory platform for die-casting industry: 1) a novel cost-effective product-tracking technology to obtain high-quality process data providing individual product information, 2) an advanced process data acquisition system that considers process failure, and 3) a fault detection module based on an artificial neural network. Our newly developed systems for the die-casting process were verified using 1500 test production. Based on the pilot production data, we developed a fault detection module with the pre-processing of time series temperature and pressure measurement data. The developed fault detection module shows 96.9 % accuracy for untrained data. The technologies developed in this study are expected to be a promising smart-factory platform to reduce the defect rate and production cost in die-casting industry.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/83897)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.