On the use of acoustic emission methods for in-situ monitoring in metal additive manufacturing: A review study

Acoustic emission (AE) sensing is emerging as a powerful, non-intrusive tool for in-situ monitoring and in-process defect detection in metal additive manufacturing (AM). Unlike other methods (e.g., optical or thermal), AE enables the real-time detection of mechanical transients directly related to dynamic events such as crack initiation, layer delamination, pore formation, etc. This review provides a systematic overview of AE-based approaches applied to the main classes of AM processes for metals and other materials. For each process, the paper discusses (i) the sensing principles and typical AE sensor configurations; (ii) methodologies for feature extraction and signal interpretation; (iii) the types of defects and anomalies that can be detected; and (iv) the machine learning and artificial intelligence techniques employed for data fusion, classification, and anomaly detection. Attention is also given to how AE data are integrated with other sensing modalities within multimodal monitoring frameworks. The review concludes by identifying open challenges, including calibration and validation issues, data synchronization, model generalization, and deployment in real industrial environments.