可靠性(半导体)
人工智能
计算机科学
机床
过程(计算)
算法
钥匙(锁)
制造工程
机器学习
材料科学
机械工程
工程类
量子力学
计算机安全
操作系统
物理
功率(物理)
作者
Yanzhou Fu,Austin Downey,Lang Yuan,Tianyu Zhang,Avery Pratt,Yunusa Balogun
标识
DOI:10.1016/j.jmapro.2021.12.061
摘要
Laser-based additive manufacturing (LBAM), a series of additive manufacturing technologies, has unrivaled advantages due to its design freedom to manufacture complex parts with a wide range of applications. Although advancements in LBAM processes and materials have led to increased manufacturing capabilities, the printing process's repeatability, durability, and reliability still face significant challenges. Therefore, a defect detection system for the LBAM processes is essential, as it promises to guarantee product quality and increase the efficiency of the printing process. As a practical and widely applied technology, machine learning methods have been providing novel insights into the manufacturing process, which has proven advantages for defect detection in LBAM. This paper summarizes the machine learning algorithms for defect detection in the metal LBAM processes. To have a comprehensive and systematic summary, machine learning algorithm, material type, defect type, dataset type, and algorithm accuracy for various LBAM technologies are described.
科研通智能强力驱动
Strongly Powered by AbleSci AI