An innovative and accurate technology for Multi-Physics Digital Twins in High-Performance Computing

Most of the physical phenomena encountered in engineering applications require multiple disciplines and their interaction to be completely described.
Furthermore studying these complex multi-physics phenomena experimentally can be difficult or impossible.
This can be due to the complex multi-physical interactions themselves, operating environments that might be hostile to sensors, or the immense scale and costs of performing such experiments.
These limitations to study complex multi-physics phenomena through experiments can be overcome by utilizing numerical models.

This thesis establishes a highly flexible, multi-component multi-physics simulation environment through a partitioned coupling approach. The proposed coupling approach uses the preCICE coupling library to couple three numerical solvers: XDEM (solves for particle motion and thermodynamics), OpenFOAM (solves the fluid dynamics and thermodynamics), and CalculiX (solves for the solid deformations).
A 6-way CFD-DEM-FEM momentum coupling and a 2-way CFD-DEM heat \& mass transfer is established.
The proposed coupling approach is verified and validated through various numerical experiments and experimental observations.

The highly flexible partitioned coupling is then used to study several large-scale complex multi-physics phenomena such as: the erosion inside abrasive water jet cutting nozzle; the drying of packed wood particle bed; frictional behavior of gravel in the presence of melting ice; the raceway region of a blast furnace; iron reduction in midrex furnace.
The thesis delves deep into the analysis of results thus giving unprecedented comprehension of such complex multi-physics phenomena.