Researchers Use Machine Learning to Detect Flaws in Laser Powder Bed Fusion Additive Manufacturing Process

A team of researchers from the University of Virginia, Carnegie Mellon University, and the University of Wisconsin-Madison has made a breakthrough in the field of additive manufacturing that could have significant implications for industries such as aerospace, ScienceDaily reports.

The team, led by Tao Sun of the University of Virginia, published a peer-reviewed paper in Science Magazine titled "Machine learning aided real-time detection of keyhole pore generation in laser powder bed fusion."

The paper addresses the problem of detecting keyhole pores, a major defect in a common additive manufacturing technique called laser powder bed fusion (LPBF).

What Is LPBF?

Yes, 3D printing with metal is already here and it is not just a futuristic concept anymore. Laser powder bed fusion, or LPBF, is a widely used process for creating intricate metal objects through 3D printing.

How does it work? A laser beam is used to melt metal powder in a bed, layer by layer, until the desired three-dimensional object is formed. It's a highly precise process that can yield complex geometries and designs.

A Major Breakthrough

AM tells us that LPBF helps manufacturers achieve near-net-shape metal parts more quickly than other techniques like forging or casting, but the process comes with a certain issue.

Porosity defects remain a challenge for applications that require strong metal parts, such as aircraft wings. Keyhole pores, which are deep and narrow vapor depressions, can make the material more brittle and prone to cracking under environmental stress.

Sun and his colleagues created a real-time detection system using operando synchrotron x-ray imaging, near-infrared imaging, and machine learning to capture the thermal signature associated with keyhole pore formation accurately. They also improved operando synchrotron x-ray imaging and identified two modes of keyhole oscillation in the process.

Sun explains that the method utilizes operando synchrotron x-ray imaging, near-infrared imaging, and machine learning to capture the unique thermal signature associated with keyhole pore generation with sub-millisecond temporal resolution and a prediction rate of 100 percent.

"Our findings not only advance additive manufacturing research, but they can also practically serve to expand the commercial use of LPBF for metal parts manufacturing," said Anthony Rollett, a materials science and engineering professor from Carnegie Mellon University.

Meanwhile, Sun emphasized that porosity in metal parts continues to be a major barrier to the wider use of LPBF technology in several industries. Because it arises stochastically beneath the surface, keyhole porosity is the most difficult defect type to detect in real-time with lab-scale sensors.

Further Studies

Keyhole porosity in laser powder bed fusion (LPBF), a process used in additive manufacturing to make metal parts, has been better-understood thanks to a new study published in Nature Communications.

Using synchrotron x-ray imaging, the researchers were able to observe keyhole and bubble behavior and quantify their formation dynamics. They found that keyhole porosity can initiate not only in unstable keyhole regimes but also in transition keyhole regimes created by high laser power-velocity conditions, leading to fast radial keyhole fluctuations.

They also discovered that transition regime collapse tends to occur partway up the rear wall and that immediately after keyhole collapse, bubbles undergo rapid growth due to pressure equilibration before shrinking due to metal-vapor condensation.

Meanwhile, a separate study has explored the use of acoustic measurements to detect keyhole pores in laser powder bed fusion.

Stay posted here at Tech Times.