Research Status of Process Monitoring and Adaptive Control in DED

Directed Energy Deposition (DED), as a key additive manufacturing technology, is widely used for repairing and fabricating large-scale metal components. However it faces challenges in process stability and quality control. This paper reviews the current research on process monitoring and adaptive control in DED. It focuses on sensor technologies for powder-based and wire-based sub-processes, such as visual, acoustic, and arc sensors for real-time monitoring of the melt pool and powder flow. Also, it shows the intelligent control methods such as machine learning and external field assistance. Machine learning models, including convolutional neural networks and regression algorithms, have demonstrated strong capabilities in predicting process outcomes and optimizing parameters. At the same time, external fields like magnetic and ultrasonic assistance have been shown to effectively improve deposition quality and microstructure. Despite progress, issues such as geometric complexity, transient dynamics, and data scarcity remain. Future work should aim at closed-loop systems integrating sensing, modeling, and control, with physics-informed machine learning to enhance adaptability and reliability. This review provides a systematic reference for advancing intelligent and self-adaptive DED systems.