Metal additive manufacturing is an advanced manufacturing technology that forms components by depositing metal materials layer by layer.During the manufacturing process,factors such as the physical environment,machine conditions,and processing parameters can lead to various defects in the fabricated parts.By employing sensors to monitor process signals and integrating machine learning algorithms,it is possible to identify defects in the fabricated parts and evaluate their quality and performance.This paper reviews the research progress on the application of acoustic sensors,thermal sensors,visible light cameras,spectroscopic sensors,and multi-sensor fusion in in-situ monitoring,feature extraction,data fusion,and machine learning algorithms for additive manufacturing.Additionally,the challenges associated with the application of machine learning technologies are discussed,with a focus on the current research status of physics-informed machine learning approaches.Finally,key issues and future research directions are summarized and outlined.
关键词
增材制造/机器学习/传感器监测/特征提取/数据融合/物理信息
Key words
additive manufacturing/machine learning/sensor monitoring/feature extraction/data fusion/physical information
维普
万方数据