http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
센서 융합형 지능형 부품 제조를 위한 적층 제조 기술 연구
정임두,이민식,우영진,김경태,유지훈,Jung, Im Doo,Lee, Min Sik,Woo, Young Jin,Kim, Kyung Tae,Yu, Ji-Hun 한국분말야금학회 2020 한국분말재료학회지 (KPMI) Vol.27 No.2
The convergence of artificial intelligence with smart factories or smart mechanical systems has been actively studied to maximize the efficiency and safety. Despite the high improvement of artificial neural networks, their application in the manufacturing industry has been difficult due to limitations in obtaining meaningful data from factories or mechanical systems. Accordingly, there have been active studies on manufacturing components with sensor integration allowing them to generate important data from themselves. Additive manufacturing enables the fabrication of a net shaped product with various materials including plastic, metal, or ceramic parts. With the principle of layer-by-layer adhesion of material, there has been active research to utilize this multi-step manufacturing process, such as changing the material at a certain step of adhesion or adding sensor components in the middle of the additive manufacturing process. Particularly for smart parts manufacturing, researchers have attempted to embed sensors or integrated circuit boards within a three-dimensional component during the additive manufacturing process. While most of the sensor embedding additive manufacturing was based on polymer material, there have also been studies on sensor integration within metal or ceramic materials. This study reviews the additive manufacturing technology for sensor integration into plastic, ceramic, and metal materials.
성유진,김영규,정임두,김성호,김시조,김성곤,김학준,박성진 대한금속·재료학회 2017 대한금속·재료학회지 Vol.55 No.11
The material characterization of single crystalline Cu columns was numerically carried out at the submicroscopic level. A molecular dynamics (MD) simulation was employed using the embedded-atom method (EAM) interatomic potential between a pair of Cu atoms to describe the interactions among Cu atoms. First, the relationship between mechanical properties and factors affecting their behavior were numerically investigated using a crystal structure including several defects. The factors were specimen size, strain rate, and temperature. As the specimen size increased the normalized yield stress decreased, which was similar to results obtained at other length-scale. The yield stress tended to lead to exponential strain rate-hardening and a linear temperature-softening. Next, material characterization was conducted based on these results. These computational results can lead to the development of an in silico platform to characterize material properties and MD simulation can lay the groundwork for multi-scale modeling and simulation.
티타늄 적층 제조 표면 조도 향상을 위한 DC GAN 심층신경망 연구
김태경(Tae Kyeong Kim),정임두(Im Doo Jung) 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.6
Direct Energy Deposition (DED) 공정은 금속 분말을 직접 분사하여 레이저로 소결하는 성형 기술이며, 큰 금속 부품을 제조하는 데에 유용한 공정이다. 하지만, 다른 적층 성형 방식에 비해 표면 조도가 낮다는 단점이 있다. 본 연구에서는 티타늄 합금 분말로 DED 공정 시 표면 조도에 영향을 주는 세 가지 공정 변수를 조절하고 그에 따라 제조된 조형체 표면을 2D Scan한 이미지로 합성곱 신경망 기반 기계학습을 진행하였다. 테스트 셋 기준 MAPE 80 %이상의 정확도로 이미지의 공정 조건을 예측해냈다. 또한 DC-GAN으로 생성한 이미지로 조형물 조도 상태 예측을 진행하였다. 이로써 티타늄 합금 분말에 대한 DED 공정시의 표면 조도 개선에 인공 지능이 적극 활용될 것으로 기대한다. The Direct Energy Deposition (DED) process is one of the additive manufacturing (AM) that sprays metal powder directly into a high-power laser and is useful for manufacturing large metal parts. However, there is a bottleneck that surface roughness is relatively low compared to other AM methods. In this work, we controlled three process variables that affect surface illumination during the DED process with titanium powder material and conducted convolutional neural network-based machine learning based on 2D scanned images of the sculptural surface manufactured accordingly. We predicted the process conditions of images with accuracy of more than 80% of MAPE based on test sets. In addition, we visually observed surface roughness with images generated by DC-GAN. It is expected that artificial intelligence will be actively used to improve surface roughness during the DED process for titanium alloy powder materials.