Rapid + tct

Towards a Deep Learning-Based Build Processor for LPBF to Achieve Structural Integrity and Repeatability for Heat Exchangers

We propose a new approach to perform slicing and hatching based on deep learning build processor technology. The algorithms are trained leveraging artificial neural network leveraging measured and physics-based data. The technology is leveraging cloud compute capability to achieve higher convergence and minimize error iteratively. The technology is demonstrated on complex heat exchanger technology that requires thin wall repeatability and consistency. We demonstrate stable powder bed fusion process by measuring the quality of the structure.

Learning Objectives:


  • Participant will learn about why achieving repeatability is very complex today especially for heat exchangers
  • Participant will learn about a totally revolutionary approach to generate scan strategies in LPBF
  • Participant will be able to leverage this technology/product and increase chance of success in printing complex geometries.

Katharina Eissing, PhD


Omar Fergani, PhD


We use cookies on our website to understand how you use it. By clicking “Agree”, you consent to their use. Privacy Policy