Many industrial processes involve interaction between both particle and fluid phases. The modelling of these systems can be made realistic by using Computational Fluid Dynamics (CFD) for fluids and Discrete Element Method (DEM) for solids in a coupled simulation.
DEM-CFD co-simulation has proven to be a key method for engineers to investigate a wide variety of particle systems such as particle agglomeration and clumping in fluidized beds; dense phase conveying; filtration; solid-liquid mixing; spray coating; drying, cooling, and heating; and transport of cuttings in oil well drilling.
Astec, Inc., a global leader in the hot mix asphalt plant industry, sees the value of simulation and has been using coupled DEM-CFD methods for many years to model their aggregate drying processes and increase their efficiency. These are complex process and many variables such a drum rotational speed and fill level can affect the performance. Direct observation of the process is impossible due to the harsh environment of the drum therefore simulation provides the best mean to visualize material’s behavior and optimize the flight design.
In a recent project involving Astec, EDEM and ANSYS, the EDEM-Fluent coupling was used to simulate the 2-way heat transfer between the hot gas phase and the aggregate particles in a dryer. This type of simulation contains complex physics and typically involves hundreds of thousands of particles thus is computationally expensive. The objective therefore was to gain key insight into the processes inside the dryer and also implement an innovative way of accelerating complex DEM-CFD simulations by making use of GPU technology.
3D CAD models of various aggregate drum dryers were used for the modelling of the transportation and drying of the hot mix asphalt and different computing conﬁgurations were implemented to investigate the effect of GPU-CPU technology in such a complex simulation. Results showed that the coupled model captured accurately the convective heat transfer from the fluid to the solid phase and demonstrated significant improvements in terms of simulation time.
EDEM-Fluent simulation of an aggregate dryer. With the use of one GPU graphic card (NVIDIA Quadro GP100), the net computational time of the coupled DEM-CFD simulation was reduced by 32% corresponding to a 1.5 speed up. Furthermore, the DEM part of the simulation experienced the greatest reduction in computational time with a net computational time reduction of 83%, making the particle calculations 6.0 times faster.
This project highlights the applicability and effectiveness of GPU technology for complex DEM-CFD simulations. It shows how promising GPU technology is for large-scale industrial applications and for simulating complex applications rapidly and accurately.
This work has been published in the Powder Technology Journal, for detailed information check the publisher paper:
M. Sousani, A.M. Hobbs, A. Anderson, R. Wood (2019) ‘Accelerated heat transfer simulations using coupled DEM and CFD’. Powder Technology. https://doi.org/10.1016/j.powtec.2019.08.095