Rapid Predictions of Particle Deposition Using rCFD

Abstract

Computational Fluid Dynamics (CFD) is widely used within the automotive industry to evaluate designs during early development stages. However, simulations involving turbulent flows and particle deposition are computationally expensive. Recurrence Computational Fluid Dynamics (rCFD) is a method that aims to reduce simulation time by utilizing recurring flow behaviour. The aim of this thesis is to implement and investigate rCFD in Simcenter STAR-CCM+ within an industrially relevant workflow. The method is first applied to a cylinder and then tested on a more complex geometry; a Volvo EX40 mirror. A method for generating a recurrence path and performing particle tracking is developed using STAR-CCM+, Java and Python.

Results show that rCFD can be successfully implemented in STAR-CCM+, reducing computational time for particle simulations, while maintaining acceptable accuracy. Furthermore, the accuracy of the method is influenced by Reynolds numbers. In conclusion, rCFD shows promising potential for reducing computational cost, enabling more efficient contamination analysis in automotive applications.