Carnegie Mellon University develops machine vision for metal Additive Manufacturing

June 12, 2017

Researchers from Carnegie Mellon University’s Department of Materials Science and Engineering (MSE), Pittsburgh, Pennsylvania, USA, have developed machine vision technology that can autonomously identify and sort metal AM powder types better than humans, the university reports. This system will enable machine users to accurately test and qualify printed metal parts for any number of applications including aerospace and medical devices. The CMU research team expects that this technology will be integrated into the AM field within the next five years.

“In traditional manufacturing, parts are often qualified through destructive testing. A company might produce multiple parts and physically test them to see how they hold up to stress and fatigue. However, that costs a lot of time and money, so it should be avoided in Additive Manufacturing in order to preserve the on-demand nature of 3D printing,” explains Elizabeth Holm, Professor of Materials Science and Engineering, and primary investigator of this research. “We therefore are looking to new qualification concepts like machine learning to guarantee successful 3D printed builds.”

 

 

By training a computer to autonomously identify and sort powders, Holm and her team can easily recognise whether or not a metal powder has the microstructural qualities associated with the production of a part with desired properties, such as strength, fatigue life and toughness. Importantly, the machine vision approach is autonomous, objective, and repeatable. This type of standardisation is necessary to advance quality assurance in the field.

In the study, “Computer vision and machine vision for autonomous characterisation of AM powder feedstocks,” Holm and her team applied computer vision and machine learning methods to eight different commercial feedstock powders. They found that the machine vision system captures more about metal AM powder than is possible with normal manual measurement. It can measure important information such as how big particles are, how particles group together, the surface roughness of particles and the shape of particles. The team also found that the technology can tell metal powder types apart even where humans cannot.

www.cmu.edu

Download the latest issue of Metal AM magazine

Our latest issue is now available to view online or download in PDF format.

As well as an extensive AM industry news section, this 176-page issue includes articles and reports on:

  • Innovation and differentiation: Precious metal Additive Manufacturing in the jewellery sector
  • Unrealised potential: The story and status of Electron Beam Powder Bed Fusion
  • Freemelt AB: Open source technology to explore the potential of Electron Beam Powder Bed Fusion
  • In pursuit of perfection: A case study on how Bugatti and APWORKS leverage the full potential of AM
  • Facing obstacles to profitability in metal AM: An Operational Excellence perspective
  • Mass-production using PBF-LB: How laser beam measurements can help pave the way
  • Link3D’s ‘Additive Manufacturing Maturity Model’: Developing an agile and resilient supply chain
  • Design for Additive Manufacturing (DfAM): Binder Jetting Technology demystified
  • > More information

Latest news

    E-newsletter

    Sign up to our e-newsletter, sent weekly to AM professionals and end-users around the world. We'll also let you know each time a new issue of Metal AM magazine is available.

    Discover our magazine archive…

    The free to access Metal Additive Manufacturing magazine archive offers unparalleled insight into the world of metal Additive Manufacturing from a commercial and technological perspective through:
    • Reports on visits to leading metal AM part manufacturers and industry suppliers
    • Articles on technology and application trends
    • Information on materials developments
    • Reviews of key technical presentations from the international conference circuit
    • International industry news
    All past issues are available to download in PDF format, in either single page format or as double-page “spreads” for viewing on large monitors. All downloads are free of charge. Go to archive...

    Connect with us

    Visit our social media channels and sign up to our e-newsletter

    Copy link
    Powered by Social Snap