Learning Forum

[luca.vene]

Dear ANSYS Fluent Forum Members,

I am currently exploring the feasibility of conducting a three-phase simulation involving air, water, and sediment using the Eulerian model in ANSYS Fluent. Specifically, I aim to simulate the local scour at the base of a bridge pier subjected to free-surface flow.

The scenario involves the interaction of free-stream water flow with the bridge pier, considering the effects of surface tension between air and water, as well as the drag effect that water exerts on sediment particles.

Simulation Objectives:

    Three-Phase Interaction: Model the simultaneous presence and interaction of air, water, and sediment in the computational domain.

    Surface Tension Effects: Capture the influence of surface tension between air and water on the flow dynamics near the bridge pier.

    Drag Effect of Sediments: Account for the drag exerted by water on sediment particles within the flow, affecting sediment transport and deposition patterns.

Approach and Questions:

- Eulerian Model Suitability: Is the Eulerian multiphase model appropriate for simulating such a scenario? I intend to use this model to handle the distinct phases of air, water, and sediment.

- Recommended Settings: What additional settings or adjustments would you suggest to successfully execute this simulation? Specifically, concerning phase interaction, turbulence modeling, and, more importantly, what settings should I configure within the Multiphase Model window? Any guidance on phase interaction or parameter tuning within this section would be particularly appreciated.

- Potential Limitations: Additionally, I would like to know about any potential limitations associated with this type of simulation, should it be feasible to conduct it in the first place.

I appreciate any insights, recommendations, or past experiences with similar simulations using ANSYS Fluent. Your expertise and guidance would greatly assist me in setting up and executing this complex simulation accurately.

Thank you in advance for your time and assistance.

Best regards,

Luca

[Rob]

Yes, with caveats. Note, I'm also limited to what's in the manuals and that is "public" knowledge. I did do this some years ago to prove a point, and have practical experience with scour pits around ship wrecks. 

The water and free surface part are relatively simple, but mesh resolution may be something you need to check relative to compute & time availability. 

The sediment is a different problem. In the Eulerian model you'd need different phases for each particle size. Additionally the model assumes a "mobile" solid, so saltation and compaction aren't considered. For a river you also have silt/material carried from upstream that may fill in a scour pit. 

[luca.vene]

Thank you for your input. I’d like to clarify the problem I am working on. My focus is on a clear water scenario, meaning that I am not concerned with sediment inflow from upstream. Instead, I aim to determine the maximum potential scour depth. The bed material consists of uniform-sized particles.

In the Multiphase Model setup, along with selecting Inhomogeneous Models: Eulerian, I’m wondering if I should enable the DDPM model or the Multi-fluid VOF model, or possibly both. From my understanding:

   - The DDPM model would allow me to define sand as a discrete phase, while air and water are treated as Eulerian phases. This would be useful if I want to track the individual sand particles' behavior.

   - The Multi-fluid VOF model would be essential for capturing the air-water interface, especially for simulating free surface dynamics.

Since I’m focusing on clear water scour and need to model both the air-water interaction and sand particles as discrete entities, should I enable both the DDPM and Multi-fluid VOF models for this scenario?

At the start of the simulation, I would initialize the domain filled with air, then use the Patch command to specify where the water and sediment are located. I plan to set a velocity inlet for water inflow and a pressure outlet for the outlet and top surface.

I would appreciate any suggestions you may have regarding the simulation setup. Thank you for your time and assistance.

Best regards,

Luca