Smart Model Advancing Reusability driven from Track record of additive manufacturing to increase Adaptability and predict Product Performance : SMART-APP

SMART-APP will provide a versatile, commercial, predictive material reuse management tool enabling Additive Manufacturing (AM) to be used more widely, cost effectively and resource efficiently.

SMART-APP will: 1) predict quality change of the virgin powder after each process, 2) propose alternative process parameters on used powder to increase its maximum reusability; resulting in a step change in resource efficiency and decision-making of AM users without impacting desired product quality.

Whilst metal AM is growing rapidly, it is still not cost-effective for all industry sectors (powder waste&time). Materials efficiency and resource management is a development gap within the AM industry.

SMART-APP will focus on Laser Powder Bed Fusion (L-PBF) AM. Current methods of powder characterisation are focussed on limited static and dynamic capabilities of virgin powders, providing limited data in relation to powder processability and, in an underdeveloped area, how powder properties age and link to material performance of manufactured parts.

In L-PBF virgin powder is aged after a limited number of (re)uses, with significant proportion(~2/3) of un-used powder excess from de-powdering the build plate and remaining proportion screened from the overflow chamber (mixed with soot). AM users return un-used (but processed) powder back to the production cycle.

This leads to the powder processability being affected, influencing printability and final product properties. In some industrial cases where powder is assessed, it is not currently possible for industry to make validated changes for increasing processability, limiting shelf life of material, potential uptake of AM and leading to unnecessary material wastage.

SMART-APP will feature state-of-the-art materials characterisation, and mechanical testing at its core, investigating two areas:

1\. Shelf life and processability envelope of environmentally affected common metallic feedstock (2 from stainless steel(SS316L), Inconel(In718), Titanium(Ti-6Al-4V)

2\. Investigate methods to reclaim the powders and understand effects on the final product

These outputs will feed into an advanced database linking powder input properties against AM part performance to provide a predictive tool (via Machine learning) for the AM industry. This database will be exploited through commercial software development made available as a paid for service, further supported by life cycle assessment and processability assessment providing a significant and novel output.

The predictive tool output from this project will develop and enable world class production of AM components, with smart designs for resource efficiency (predictive model closes characterisation loop and makes materials smarter by design) and providing longer in use of materials feedstock reducing materials wastage.