journal article·2024-12-16Open access

An alloy-agnostic machine learning framework for process mapping in laser powder bed fusion

Toby Wilkinson, Massimiliano Casata, Daniel Barba

Rapid Prototyping Journal· Volume 30· Issue 11

Abstract

Purpose This study aims to introduce an image-based method to determine the processing window for a given alloy system using laser powder bed fusion equipment based on achieving the desired melting mode across multiple materials for powder-free specimens. The method uses a convolutional neural network trained to classify different track morphologies across different alloy systems to select appropriate printing settings. This method is intended for the development of new alloy systems, where the powder feedstock may be unavailable, or prohibitively expensive to manufacture. Design/methodology/approach A convolutional neural network is designed from scratch to identify the 4 key melting modes that are observed in laser powder bed fusion additive manufacturing across different alloy systems. To increase the prediction accuracy and generalisation accuracy across different materials, the network is trained using a novel hybrid data set that combines fully unsupervised learning with semi-supervised learning. Findings This study demonstrates that our convolutional network with a novel hybrid training approach can be generalised across different materials, and k-fold validation shows that the model retains good accuracy with changing training conditions. The model can predict the processing maps for the different alloys with an accuracy of up to 96% in some cases. It is also shown that powder-free single-track experiments are a useful indicator for predicting the final print quality of a component. Originality/value The “invariant information clustering” (IIC) approach is applied to process optimisation for additive manufacturing, and a novel hybrid data set construction approach that accounts for uncertainty in the ground truth data, enables the trained convolutional model to perform across a range of different materials and most importantly, generalise to materials outside of the training data set. Compared to the traditional cross-sectioning approach, this method considers the whole length of the single track when determining the melting mode.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Additive Manufacturing Materials and Processes · Additive Manufacturing and 3D Printing Technologies · Welding Techniques and Residual Stresses

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Universidad Politécnica de Madrid E.T.S. de Ingeniería Aeronáutica y del Espacio, , Madrid,

References (46)

Show source-provided references
1. Tensorflow: a system for large-scale machine learning
2. Alloys-by-design: a low-modulus titanium alloy for additively manufactured biomedical implants10.1016/j.actamat.2022.117749 ↗
3. Numerical and experimental investigations on laser melting of stainless steel 316l metal powders10.1016/j.jmapro.2014.04.001 ↗
4. A python surrogate modeling framework with derivatives10.1016/j.advengsoft.2019.03.005 ↗
5. Review on volumetric energy density: influence on morphology and mechanical properties of ti6al4v manufactured via laser powder bed fusion10.1007/s40964-022-00328-0 ↗
6. Simulation and analysis of LPBF multi-layer single-track forming process under different particle size distributions10.1007/s00170-021-06987-7 ↗
7. Heat transfer in lattice structures during metal additive manufacturing: numerical exploration of temperature field evolution10.1108/RPJ-11-2018-0288 ↗
8. Investigation of processing windows in additive manufacturing of alsi10mg for faster production utilizing data-driven modeling10.1016/j.addma.2022.102858 ↗
9. Erik · 2013 · 4
10. The effect of powder characteristics on build quality of high-purity tungsten produced via laser powder bed fusion (LPBF)10.1007/s11661-019-05601-6 ↗
11. Overhang constraint for topology optimization of self-supported compliant mechanisms considering additive manufacturing10.1016/j.cad.2018.12.006 ↗
12. Melt Pool characterization for selective laser melting of ti-6al-4v pre-alloyed powder
13. A critical review on effect of process parameters on mechanical and microstructural properties of powder-bed fusion additive manufacturing of ss316l
14. Deep residual learning for image recognition
15. Understanding the laser powder bed fusion of alsi10mg alloy10.1007/s13632-020-00659-w ↗
16. Invariant information clustering for unsupervised image classification and segmentation
17. An efficient algorithm for constructing optimal design of computer experiments10.1016/j.jspi.2004.02.014 ↗
18. Observation of keyhole-mode laser melting in laser powder-bed fusion additive manufacturing10.1016/j.jmatprotec.2014.06.005 ↗
19. Imagenet classification with deep convolutional neural networks10.1145/3065386 ↗
20. Convolutional networks and applications in vision
21. Network in network
22. Machine-learning assisted laser powder bed fusion process optimization for alsi10mg: new microstructure description indices and fracture mechanisms10.1016/j.actamat.2020.10.010 ↗
23. Effect of processing parameters on pore structures, grain features, and mechanical properties in ti-6al-4v by laser powder bed fusion10.1016/j.addma.2022.102915 ↗
24. Direct measurements of laser absorptivity during metal melt Pool formation associated with powder bed fusion additive manufacturing processes10.2351/1.5040636 ↗
25. Support structure constrained topology optimization for additive manufacturing10.1016/j.cad.2016.08.006 ↗
26. Understanding the effect of laser scan strategy on residual stress in selective laser melting through thermo-mechanical simulation10.1016/j.addma.2016.05.014 ↗
27. Towards the next generation of machine learning models in additive manufacturing: a review of process dependent material evolution10.1016/j.pmatsci.2023.101102 ↗
28. Melting modes in laser powder bed fusion10.1016/j.mtla.2020.100591 ↗
29. The influence of beam focus during laser powder bed fusion of a high reflectivity aluminium alloy—alsi10mg10.1016/j.addma.2022.103112 ↗
30. Characterization and qualification of LPBF additively manufactured aisi-316l stainless steel brackets for aerospace application10.1007/s41403-020-00159-x ↗
31. Fiji: an open-source platform for biological-image analysis10.1038/nmeth.2019 ↗
32. A multi-scale convolutional neural network for autonomous anomaly detection and classification in a laser powder bed fusion additive manufacturing process10.1016/j.addma.2018.09.034 ↗
33. Melt Pool geometry and morphology variability for the inconel 718 alloy in a laser powder bed fusion additive manufacturing process10.1016/j.addma.2019.100830 ↗
34. Using machine learning to aid in the parameter optimisation process for metal-based additive manufacturing10.1108/RPJ-08-2019-0213 ↗
35. Very deep convolutional networks for large-scale image recognition
36. The design process of additively manufactured mesoscale lattice structures: a review10.1115/1.4040131 ↗
37. Metallic powder-bed based 3d printing of cellular scaffolds for orthopaedic implants: a state-of-the-art review on manufacturing, topological design, mechanical properties and biocompatibility
38. Prediction of porosity in metal-based additive manufacturing using spatial Gaussian process models10.1016/j.addma.2016.05.009 ↗
39. Influence of keyhole and conduction mode melting for top-hat shaped beam profiles in laser powder bed fusion10.1016/j.jmatprotec.2019.116514 ↗
40. In situ absorptivity measurements of metallic powders during laser powder-bed fusion additive manufacturing10.1016/j.apmt.2017.08.006 ↗
41. Optimizing laser powder bed fusion parameters for in-738lc by response surface method10.3390/ma13214879 ↗
42. Space-time topology optimization for additive manufacturing10.1007/s00158-019-02420-6 ↗
43. In situ tailoring microstructure in additively manufactured ti-6al-4v for superior mechanical performance10.1016/j.actamat.2016.12.027 ↗
44. Density and mechanical properties in selective laser melting of invar 36 and stainless steel 316l10.1016/j.jmatprotec.2018.11.006 ↗
45. Combined effect of powder properties and process parameters on the density of 316l stainless steel obtained by laser powder bed fusion10.1007/s00170-022-09160-w ↗
46. Adam: a method for stochastic optimization