Project

Machine Learning integration in Discrete Element solvers for modeling of industrial powder handling processes compromising sustainability, safety and costs

Code
EXT/ONZ/000242
Duration
13 May 2022 → 01 December 2023
Research disciplines
  • Natural sciences
    • Machine learning and decision making
  • Engineering and technology
    • Powder and particle technology
    • Computational materials science
Keywords
Machine Learning powder handling processes Discrete Element Method (DEM)
 
Project description

The research consists in exploring ways to incorporate ML techniques on DEM, using our in-house developed software, ScaleDEM. These techniques range from the use of Decision Trees to Artificial Neural Networks (ANN), in order to replace or improve part of the current ScaleDEM implementation. This would necessarily imply the incorporation of a mixed approach to the software. Thus, the software will rely on both physics and data to provide robust, reliable and accurate results.