Introduction
Modeling of localized scouring around hydraulically immersed and coastal structures represents a multi-scale challenge. Traditional approaches in pure Computation Fluid Dynamics typically treat the particle phase, whether it is a micro-scale sediment bed or meso-scale layers of protective rock, as continuous porous medium. This simplified approach does not model the discrete mechanics that describes the particles and thus cannot fully model particle-scale phenomena such as interlocking, dilation and entrainment that drive structural failures.
Our approach to this problem is to couple the traditional CFD approach with Aspherix® DEM using CFDEM®coupling. Using this coupled CFD-DEM methodology we explicitly capture the fluid-particle interactions at these micro- and meso-scales. By fully capturing these fluid-particle interactions we can accurately predict the crucial macro-scale phenomena such scouring, hydraulic resistance and structural stability.
In this article three applications of the CFDEM® approach are addressed with regards to: pillar scouring, coastal revetments, and rip-rap scouring at a weir. Our work here is in collaboration with the Canadian National Research Council (CNRC).
Pillar Scouring
Video 1: The CFDEM®coupling approach provided the ability to investigate the scouring phenomena in much finer and more granular detail than possible when using a continuum approximation for the sediment bed.
In our first application we looked at scouring of the sediment around pillars and piers. Scouring is the removal of the sediment such as sand and gravel due to complex fast-flowing fluid structures induced around the flow obstruction (see Video 1, 2). This phenomena occurs at the micro-scale as the fluid flow picks up the individual sediment particles and deposits them downstream. Ultimately this results in deep holes around the pillar can compromise its structural integrity.
The simulation uses a VoF-based solver to track the liquid-air interface to ensure correct water level height above the sediment bed, an important factor in the scour formation. The fluid and particles are bi-directionally coupled with a drag model [1] which accounts for both the porous drag in the particle bed and the drag for particles entrained in the flow . A secondary lift [2] model accounts for the effect of fluid shear on the surface of the sediment bed.
Coarse-graining is applied to the sub-millimeter sized sediment (sand) which makes the numerical simulation feasible over the long timescales that scouring occurs is while preserving the fundamental particle-scale physics. A special treatment of the turbulence model in the particle bed ensures that the fluid boundary layer is correct thus preserving the complex flow structures responsible for the scouring.
Video 2: Top view of the scouring process represented in Video 1.
Coastal Revetment
For our second application we looked at dynamic waves flowing over a coastal revetment. The primary purpose of a revetment is to absorb and disperse the kinetic energy of waves that over time wear down coastlines. Coastal revetments are often built up using multiple layers of particles: with large irregular rocks and stones on the surface, and looser gravel providing support internally (see Video 3).
The periodic impact of waves on the revetment can dislodge and displace these large particles. exposing the inner core to flow that can ultimately erode the structure as a whole. The simulation uses the same VoF-based solver used for the pillar scouring application.
Multiple layers of convex particles of various sized were used to create the revetment. Smaller particles were used to construct the scourable base layer, and larger particles used to form the protective outer layer. The wave model [3] used at the inlet boundary induced waves that would repeatably crash and flow over the revetment. The particles and fluid were bi-directionally coupled with a drag model [1] for the primary force the waves would apply to the revetment. An appropriate voidfraction model [4] ensured an accurate representation of the convex particles in the flow field.
For our final application we looked at water flowing over a weir and it's interaction with downstream rip-rap. The larger stones in the downstream rip-rap provides the same functionality as those found in coastal revetments: energy dissipation and erosion protection. However, unlike the revetment but similar to pillar scouring, the primary driver of erosion is the continuous application of hydraulic force be the flow that accelerates over the weir. Protecting the soil foundation of the weir is crucial to its structural stability.
The simulation is based on a large scale weir experiment conducted by CNRC. The flow over the weir, including upstream and downstream conditions, is modelled using a VoF-solver. Rip-rap particles of various irregular shapes and sizes were placed directly downstream of the weir, and small boulder particles were placed at regular intervals (slightly embedded) on the top of the rip-rap (see Video 4).
Video 4: CFD-DEM simulation of water flowing over a weir and it's interaction with downstream rip-rap.
Video 3: The CFDEM®coupling approach provides the ability to investigate the efficacy of revetment in both how well it absorbs the kinetic energy of the waves and how well it stands up to the hydraulic forces imparted upon it.
Weir Rip-Rap
Conclusions
The three applications presented here demonstrate the breadth and versatility of the CFDEM®coupling approach for hydraulic and coastal engineering problems.
Across pillar scouring, coastal revetment, and weir rip-rap, a common theme emerges: macro-scale structural behavior is governed by particle-scale physics that continuum models cannot fully capture. Interlocking, dilation, and individual particle entrainment are not artefacts of the model, they are the mechanisms that determine whether a structure holds or fails. By resolving these explicitly through coupled CFD-DEM simulations, we obtain predictions of scour depth, energy dissipation, and structural displacement that are inaccessible to porous-medium approximations.
Coarse-graining and appropriate turbulence treatment make these simulations computationally tractable at engineering-relevant scales, while preserving the underlying particle mechanics. The close collaboration with the Canadian National Research Council, and the availability of physical experimental data from the weir study, provides confidence that the modelling framework is capturing real behaviour and not numerical artifact.
“Advanced fluid–particle modelling approaches are transforming river and coastal engineering by enabling more realistic prediction of sediment transport, scour, debris interactions, and infrastructure resilience under complex hydraulic conditions."
— Abolghasem Pilechi, CNRC
Together, these results position CFDEM®coupling as a practical tool for the design and assessment of hydraulic infrastructure, enabling engineers to probe failure mechanisms, optimise protection layers, and reduce reliance on conservative safety factors derived from purely empirical methods.
References
[1]: R. Di Felice, "The voidage function for fluid-particle interaction systems," International Journal of Multiphase Flow, vol. 20, no. 1, pp. 153–159, 1994. https://doi.org/10.1016/0301-9322(94)90011-6
[2]: R. Mei, "An approximate expression for the shear lift force on a spherical particle at finite Reynolds number," International Journal of Multiphase Flow, vol. 18, no. 1, pp. 145–147, 1992. https://doi.org/10.1016/0301-9322(92)90012-6
[3]: P. Higuera, I.J. Losada, and J.L. Lara, "Realistic wave generation and active wave absorption for Navier-Stokes models: Application to OpenFOAM®," Coastal Engineering, vol. 71, pp. 102–118, 2013. https://doi.org/10.1016/j.coastaleng.2012.07.002
[4]: C. Kloss, C. Goniva, A. Hager, S. Amberger, and S. Pirker, "Models, algorithms and validation for opensource DEM and CFD-DEM," Progress in Computational Fluid Dynamics, vol. 12, no. 2/3, pp. 140–152, 2012. https://doi.org/10.1504/PCFD.2012.047457
Do you want to know more?
The Author:
Dr Daniel Louw
Researcher and Consultant at DCS Computing GmbH.