Skip to main content
Go to the home page of the European Commission (opens in new window)
English en
CORDIS - EU research results
CORDIS

Metal flow in laser additive manufacturing using x-ray imaging and machine learning

Project description

Revealing keyhole dynamics in metal 3D printing

Laser metal deposition (LMD) is a 3D printing process that offers tremendous geometric freedom when repairing damaged metal parts. The process uses a laser beam to melt metal powder in a continuous stream to build up a part. The promise of LMD to revolutionise metal processing is constrained by a widespread problem: the creation of tiny gas pockets – keyhole voids – in the final stage that can lead to reduced mechanical properties. The EU-funded MFILAMUXIAML project will combine X-ray imaging techniques with machine learning to better examine the dynamics of the keyhole phenomenon. Research results could serve as a guide to optimising the process parameters and improve the quality of the 3D printed parts.

Objective

Additive Manufacturing (AM) has been a hot topic for many years. A fundamental understanding of the metal flows in the molten pool is critical to improving the quality of the sample produced by AM. Laser metal deposition (LMD) is one of the most widely used AM methods, which has a high production efficiency, and a special application in repairing the damaged parts with large size and high price. During the LMD process, a powdery filler is injected from the nozzle onto the surface of the base metal, and a laser beam is used to melt the powders and surface of a specimen. Dynamics of the keyhole and molten pool will determine the temperature distribution and the profile of the molten pool, thus affecting the microstructure of the printed bead. It is difficult to observe the dynamics of the molten pool in AM directly with a camera because the molten pool is surrounded by the solid metal. The objectives of this proposal are to reveal the dynamic characteristics of the keyhole and molten pool in LMD and provide a guide for choosing the proper production parameters for defect-free AM. To achieve the objectives, the X-ray imaging system in the host institute will be used to observe the dynamics of the keyhole and molten pool. With the high-melting-point tungsten particles as the tracers mixed with the metal powders, the flow of the liquid metal in the molten pool could be observed. The machine learning technique is then used to track the flow of the tungsten particles, which makes the quick determination of the flow routes and velocities of the liquid metal possible. The novelty of this proposal is that the x-ray imaging method combined with the machine learning technique is first used to visualize the dynamics of the keyhole and molten pool in LMD. This research could provide a guide for optimizing the process parameters and improving the AM quality, and the achievement of this research could contribute to the development of LMD in the industry.

Keywords

Programme(s)

Topic(s)

Funding Scheme

MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)

See all projects funded under this funding scheme

Call for proposal

(opens in new window) H2020-MSCA-IF-2019

See all projects funded under this call

Coordinator

UNIVERSITY COLLEGE LONDON
Net EU contribution
€ 212 933,76
Address
GOWER STREET
WC1E 6BT London
United Kingdom
Region
London Inner London — West Camden and City of London
Activity type
Higher or Secondary Education Establishments
Links
Total cost
€ 212 933,76