Performance of drag force models for shock-accelerated flow in dense particle suspensions

TL;DR

This study evaluates various drag force models for shock-accelerated dense particle suspensions using particle-resolved simulations, highlighting the importance of correction schemes and velocity fluctuation models for accurate predictions.

Contribution

It provides a comprehensive assessment of drag force models and correction schemes, demonstrating their effectiveness and limitations in dense particle flow simulations.

Findings

Correction schemes improve force predictions significantly.

Total impulse remains underpredicted despite corrections.

Velocity fluctuation models are crucial for accurate mean flow predictions.

Abstract

Models for prediction of drag forces within a particle cloud following shock-acceleration are evaluated with the aid of results from particle-resolved simulations in order to quantify how much the disturbances introduced by the proximity of nearby particles affect the drag forces. The drag models evaluated here consist of quasi-steady forces, undisturbed flow forces, inviscid unsteady forces, and viscous unsteady forces. Two dense particle curtain correction schemes to these forces, based on volume fraction and input velocity, are also evaluated. The models are tested in two ways; first they are evaluated based on volume-averaged flow fields from particle-resolved simulations; secondly, they are applied in Eulerian-Lagrangian simulations, and the results are compared to the particle-resolved simulations. The results show that both correction schemes significantly improve the particle force predictions, but the average total impulse on the particles is still underpredicted by both correction schemes in both tests. With the volume averaged flow fields as input, the volume fraction correction gives the best results. However, in the Eulerian-Lagrangian simulations it is demonstrated that the velocity fluctuation model, associated with the velocity correction scheme, is crucial for obtaining accurate predictions of the mean flow fields.