In situ process monitoring of multi-layer deposition in wire arc additive manufacturing (WAAM) process with acoustic data analysis and machine learning

• Published in: International journal of advanced manufacturing technology Vol. 132; no. 9-10; pp. 5087 - 5101
• Main Authors: Rahman, Md Arifur, Jamal, Suhaima, Cruz, Meenalosini Vimal, Silwal, Bishal, Taheri, Hossein
• Format: Journal Article
• Language: English
• Published: London Springer London 01.06.2024; Springer Nature B.V
• Subjects: Acoustic emission; Acoustics; Additive manufacturing; Algorithms; Arc deposition; CAE) and Design; Clustering; Computer-Aided Engineering (CAD; Data analysis; Engineering; Industrial and Production Engineering; Machine learning; Manufacturing; Mechanical Engineering; Media Management; Monitoring; Monolayers; Multilayers; Nondestructive testing; Original Article; Signatures; Wire; Wire arc additive manufacturing (WAAM); Signal processing; K-mean clustering; In-situ monitoring; Acoustic; Machine learning
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-024-13641-5

Additive manufacturing (AM) of metal components is expanding as a developing technique for fabricating high-value and large-scale metal parts. Among various metal AM procedures, wire arc additive manufacturing (WAAM) received special attention in recent years due to its unique potential for fabricating complex geometries in industrial scale and applications. In this study, a step forward for developing a continuous, multi-layer in-situ monitoring technique based on acoustic signatures recorded by acoustic emission nondestructive method over the deposition process is presented. The major goal of this research is to investigate if previously proven single-layer monitoring procedures based on acoustic signatures can be expanded toward a robust multi-layer and continuous monitoring method. Two different types of materials have been used in a WAAM process equipped with acoustic emission sensors, and recorded signals were analyzed by traditional statistical assessment as well as a K-mean clustering machine learning algorithm. The findings affirm the effectiveness of acoustic signals in monitoring processes during the continuous deposition of material and indicate that acoustic signals can reliably identify distinct process states across all layers. This underscores the reliability of acoustic signals as a multi-layer process monitoring method.