期刊
COMPUTERS & INDUSTRIAL ENGINEERING
卷 124, 期 -, 页码 322-330出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.cie.2018.07.041
关键词
Machine learning; Additive manufacturing; Process monitoring; Selective laser melting; Powder bed fusion
Metal 3D printing is one of the fastest growing additive manufacturing (AM) technologies in recent years. Despite much improvements in its technical capabilities, reliable metal printing is still not well understood. One of the barriers of industrialization of metal AM is process monitoring and quality assurance of the printed product. These barriers are especially much highlighted in aerospace and medical device manufacturing industries where the high reliable and quality products are needed. Selective Laser Melting (SLM) is one of the main metal 3D printing methods where more than 50 parameters may affect the quality of the print. However, current SLM printing processes only utilize a fraction of the collected data for quality related tasks. This study proposes a process monitoring framework named MLCPM (Multi-Layer Classifier for Process Monitoring) to predict the likelihood of successful printing at critical printing stages based on collective data provided by identical 3D printing machines producing the same part. The proposed framework provides a blueprint for control strategies during a printing process and aims to prevent defects using data-driven techniques. A numerical study using simulated data is provided to demonstrate how the proposed method can be implemented.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据