Machine learning for defect detection for PBFAm using high resolution layerwise imaging coupled with post-build CT scans

Jan Petrich, Christian Gobert, Shashi Phoha, Abdalla R. Nassar, Edward William Reutzel

Research output: Contribution to conferencePaper

Abstract

This paper develops a methodology based on machine learning to detect defects during Powder Bed Fusion Additive Manufacturing (PBFAM) processes using data from high resolution images. The methodology is validated experimentally using both a support vector machine (SVM) and a neural network (NN) for binary classification. High resolution images are collected each layer of the build, and the ground truth labels necessary for supervised machine learning are obtained from a 3D computed tomography (CT) scan. CT data is processed using image processing tools-extended to 3D-in order to extract xyz position of voids within the component. Anomaly locations are subsequently transferred from the CT domain into the image domain using an affine transformation. Multi-dimensional features are extracted from the images using data surrounding both anomaly and nominal locations. Using cross-validation strategies for machine learning and testing, accuracies of close to 90% could be achieved when using a neural network for in-situ anomaly detection.

Original languageEnglish (US)
Pages1363-1381
Number of pages19
StatePublished - Jan 1 2020
Event28th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2017 - Austin, United States
Duration: Aug 7 2017Aug 9 2017

Conference

Conference28th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2017
CountryUnited States
CityAustin
Period8/7/178/9/17

All Science Journal Classification (ASJC) codes

  • Surfaces, Coatings and Films
  • Surfaces and Interfaces

Fingerprint Dive into the research topics of 'Machine learning for defect detection for PBFAm using high resolution layerwise imaging coupled with post-build CT scans'. Together they form a unique fingerprint.

  • Cite this

    Petrich, J., Gobert, C., Phoha, S., Nassar, A. R., & Reutzel, E. W. (2020). Machine learning for defect detection for PBFAm using high resolution layerwise imaging coupled with post-build CT scans. 1363-1381. Paper presented at 28th Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference, SFF 2017, Austin, United States.