Research Article Open Access

Machine Learning Approach for Defects Identification in Dissimilar Friction Stir Welded Aluminium Alloys AA 7075-AA 1100 Joints

Akshansh Mishra1
  • 1 Politecnico Di Milano, Italy

Abstract

Machine learning approaches are now applied in various manufacturing industries. Various machine learning algorithms can be implemented for prediction of the particular mechanical properties like Ultimate Tensile Strength (UTS), Elongation percentage and fracture strength of the given mechanical component and also image processing algorithms can be applied for defects detection in the mechanical components. In our recent work, we have used a novel machine learning approach for the detection of the surface defects in dissimilar Friction Stir Welded joints by using Local Binary Pattern (LBP) algorithm. The results obtained are satisfying and it is concluded that the LBP can be implemented in the detection of surface defects.

Journal of Aircraft and Spacecraft Technology
Volume 4 No. 1, 2020, 88-95

DOI: https://doi.org/10.3844/jastsp.2020.88.95

Submitted On: 30 May 2020 Published On: 1 July 2020

How to Cite: Mishra, A. (2020). Machine Learning Approach for Defects Identification in Dissimilar Friction Stir Welded Aluminium Alloys AA 7075-AA 1100 Joints. Journal of Aircraft and Spacecraft Technology, 4(1), 88-95. https://doi.org/10.3844/jastsp.2020.88.95

  • 1,578 Views
  • 522 Downloads
  • 8 Citations

Download

Keywords

  • Friction Stir Welding
  • Local Binary Pattern
  • Machine Learning
  • Surface Defects