.png){kind=icon}
返回
Machine learning-driven 3D printing: A review
.png){kind=icon}
PRE
AI
DOI：10.1016/j.apmt.2024.102306.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In recent years, Addit. Manuf. (AM, i.e. 3D printing) has been developing rapidly as a much-needed strategic technology in the field of high-end intelligent manufacturing. However, it is constrained by the cross-influence of materials, processes, and algorithms, which make the print quality more varied, and fortunately machine learning has brought new optimization strategies for 3D printing. This paper provides an overview of machine learning-driven 3D printing technology, with a particular focus on its process, monitoring, and motion planning. In addition, the introduction of a 6-DOF robotic arm allows for more versatile printing paths. We found that supervised learning has a high rate of application in process optimization and surface quality inspection through the review. While reinforcement learning and deep learning have a clear advantage in path planning for multidegree of freedom printing platforms, which sets the stage for pioneering applications of AI in Addit. Manuf..
Keyword:
Addit. Manuf.
Machine learning
Process parameters
Robotic arms
Path planning
期刊
IF:
6.9
论文数:
3.4K
被引数:
1.6W
机构
暂无机构信息
引用论文
Coverage path planning for kiwifruit picking robots based on deep reinforcement learning基于深度强化学习的猕猴桃采摘机器人覆盖路径规划
PREAI
分享收藏Preparation of high-density green body based on binder jetting 3D printing using spheroidized SiC powder基于球形化SiC粉末的粘合剂喷射3D打印制备高密度生坯
PREAI
分享收藏Porosity prediction: Supervised-learning of thermal history for direct laser deposition孔隙度预测: 直接激光沉积热历史的监督学习
OAAI
OAAI分享收藏分享收藏A multiobjective optimization parameters applied to additive manufacturing: DOE-based approach to 3D printing.png)
STRUCTURESOAAI
STRUCTURESIF4.3
OAAI分享收藏
查看更多内容