Laser-Based Additive Manufacturing Process Control

Laser-based additive manufacturing encompasses techniques that fuse powdered or wire‐feedstock metals layer by layer using a high‐power laser. Precision control of laser power, scanning speed, feedstock delivery and thermal history is critical to achieve dimensional accuracy, microstructural consistency and mechanical integrity. In‐situ sensing of melt‐pool dynamics, temperature fields and geometry deviations enables closed‐loop feedback whereby process parameters are automatically adjusted to suppress porosity, residual stress and layer‐to‐layer misalignments. Control strategies range from classic proportional–integral–derivative loops and model-based predictive algorithms to digital twins and data-driven approaches. Recent innovations have integrated high-speed coaxial imaging, infrared thermography and pyrometry with real-time signal processing and machine-learning models to predict track geometry, classify defect states and adapt toolpaths on the fly. Such systems have demonstrated marked improvements in interlayer adhesion, surface finish and part repeatability across aerospace, automotive and biomedical applications, underlining the global significance of robust laser-process control for high-value metal component fabrication.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Several studies published in the past two years have advanced real-time control of laser metal deposition and directed energy deposition processes. A deep-learning framework combining coaxial melt-pool images with process and trajectory data has enabled highly accurate prediction of single‐track width and height, demonstrating invariance to network depth and robust performance on Inconel 718 substrates. Another investigation applied contrastive-learning convolutional neural networks to co-axial process-zone videos during titanium deposition, achieving up to 97 % accuracy in classifying porosity and microstructure anomalies and enabling supervised and semi-supervised quality grading. Complementing these image-based approaches, a regression-based study of wire-fed laser deposition established linear models correlating energy per unit length and travel speed with bead height and width for aluminium and stainless-steel wires. This work yielded a material-independent formula for layer increment, facilitating rapid parameter‐setting during process development. Together, these contributions illustrate the convergence of machine-learning, in-situ sensing and parametric modelling to deliver adaptive, robust control in laser-based additive manufacturing.

Laser-Based Additive Manufacturing Process Control publication trend

The graph below shows the total number of articles in laser-based additive manufacturing process control across all publications each year (not limited to Nature Index journals).

Technical terms

Laser-Based Additive Manufacturing: Any layer-wise fabrication process using a laser to melt and fuse metal powder or wire.

Directed Energy Deposition (DED): A technique in which focused energy is used to melt feedstock as it is deposited, enabling repairs and complex shapes.

Melt pool: The localized region of molten material created by the laser, whose geometry and temperature govern solidification and defect formation.

Coaxial monitoring: Imaging or sensing aligned with the laser axis, providing synchronized views of the melt pool and deposition zone.

Process control: The real-time adjustment of manufacturing parameters via feedback or predictive algorithms to maintain target quality and geometry.

Convolutional Neural Network: A class of deep-learning model specialised in extracting spatial features from images, widely used for defect detection and geometry prediction.

References

1. Track geometry prediction for Laser Metal Deposition based on on-line artificial vision and deep neural networks. Robotics and Computer-Integrated Manufacturing (2023).
2. In situ quality monitoring in direct energy deposition process using co-axial process zone imaging and deep contrastive learning. Journal of Manufacturing Processes (2022).
3. Investigation on the Cause-Effect Relationships between the Process Parameters and the Resulting Geometric Properties for Wire-Based Coaxial Laser Metal Deposition. Metals (2022).