The main purpose of this tutorial is to learn how to use Coarse-Graining to increase the particle size while maintaining the system behavior, but using a reduced number of (larger) particles.

We will also learn how to export Ansys Fluent results as an .f2r file (optional) and further couple this already-processed CFD simulation with our DEM simulation setup in Rocky.

You will learn how to:

And you will use these programs and features:

This ADVANCED tutorial assumes that you are already with the Rocky user interface (UI) and Rocky project workflow.

The geometry in this tutorial is composed of:

In the tutorial directory, a .cas file containing this geometry can be found.

To set up the Fluent Case, do the following:

In this tutorial, we will export an .f2r (Fluent to Rocky) file, which requires prior installation of a special Fluent/Rocky export component.

Ensure that you have this component by doing the following:

For this tutorial, we want to do all of the following:

Let's start by learning about Coarse-Graining.

Coarse-Graining enables you to increase the original particle size to study the system behavior using a feasible number of (larger) particles.

This physics accomplishes this by doing the following:

Rather than creating an impractical case involving, perhaps, hundreds of millions of tiny particles, Coarse-Graining makes it possible to analyze an approximated case with a more manageable particle count all without sacrificing accuracy.

To apply Coarse-Graining in Rocky, you specify a unique CGM Scale Factor for each Particle set.

This factor is multiplied by the original particle size to obtain an easier-to-process scaled-up particle, called a parcel.

The job of a single, scaled-up parcel is to represent many original-sized particles while retaining the same interaction behaviors as the smaller particles it represents.

To achieve this latter goal, the CGM Scale Factor is also used to adjust the contact, adhesion, thermal, drag, and other interaction properties of the scaled-up parcel so that it behaves more like the original-sized particles.

Five separate geometry components are imported.

For this tutorial, we need only the wall geometry.

Next, we will define a surface that will be used as an inlet to release the particles into the domain.

We will also be creating a spherical-shaped particle group with a range of three different sizes.

Note that we will be using default values for both the Materials and Materials Interactions settings.

For the Inlets and Outlets step, we will create a Particle Inlet.

Now, let's see how many particles are estimated if we simulate at this (original) particle size.

To make this simulation more reasonable to process, we will now set a higher CGM Scale Factor to reduce the particle count.

By viewing again the particle estimates in Inlets and Outlets, a significant reduction of particles can be seen.

For the CFD Coupling step, we will be importing some steady-state CFD results that were already processed in Fluent.

Now that the CFD results are within Rocky, you can visualize the nodes of the cell centroids making up the Fluent mesh.

A better practice is to create a thin slice of your mesh and show only the vectors within that slice.

The colored vectors show that the fluid enters the domain through two inlets, (left portion of the figure) and flow through two outlets (right portion of the figure).

Lastly, we need to define the Solver setup.

With a 3D View window opened, your Data panel and Workspace should look similar to the below image.

The Simulation Summary screen appears, then processing begins.

This completes Part A of this tutorial.

Rocky was used to set up and process a Mixing Tee simulation.

During this tutorial, it was possible to:

What's Next? If you completed this tutorial successfully, then you are ready to move on to Part B and post-process this project.

The purpose of this tutorial is to use the results from the Mixing Tee simulation we set up and processed in Part A to analyze which wall regions are most susceptible to wear and discover how the particles are distributed between the outlets.

You will learn how to:

And you will use these features:

This ADVANCED tutorial assumes that you are already with the Rocky user interface (UI) and Rocky project workflow.

In order to perform statistical analysis, it is necessary to select a period of time in which the amount of particles inside the mixing tee remains stable.

We will do this by measuring the Particles Count curve over time.

The resulting time plot shows that after approximately 0.5 seconds, the amount of particles inside the domain remains stable.

The range between 0.5 s and 1.0 s is a period of continuous operation of the equipment, and could be used for a statistical analysis.

We next want to compare the mass flow between the two outlets and understand how the particles separation occurs.

We will therefore define three Cube User Processes; one for the inlet and another two for the outlets.

It is now possible to visualize the cubes in a 3D View.

In order to measure the total mass that passed through each cube once a steady state is reached (from 0.5 s on), we will create a Particles Time Selection process for each Cube.

Next, we will build a Histogram comparing the total mass and particle sizes between each Cube.

The Histogram shows the mass fraction of three ranges of particle sizes.

The dark red bins show that approximately 66% of the Inlet mass fraction is of small particles. The middle-sized and big particles have almost 17% of mass fraction each.

For the Main Outlet (orange bins), we have almost 80% of small particles mass fraction.

The blue bins (Downwards Outlet) show that the bigger the particle, the higher the mass fraction. In other words, bigger particles tend to flow downwards.

These results make sense as bigger particles (with greater mass) would require more energy to accelerate and escape the downwards outlet as the gravity force acts.

In Rocky, you are able to visualize the effects of wear using the surface wear modification feature.

The wear law used in that feature takes into account the shear work on the geometry surface to predict how it will wear down over time.

However, for the case we are studying in this tutorial, the surface modification of the geometry would not be noticeable due to the short simulation time.

Instead, we can compute the total work done by the shear forces on each triangle surface and then verify which regions of the geometry have the highest shear work.

You might recall that in Part A, we enabled the collection of Intensities data for Boundary Collision Statistics. We can use that data to perform the shear work analysis.

A good approximation of it would be:

(17–1)

Where:

Let's now create a new custom Property that defines this.

First, we are going to calculate the Cumulative Intensity : Shear, which is the factor of the equation.

The new Cumulative Intensity : Shear (Custom) property will appear on Properties tab under Transient.

Now, we are going to calculate the actual Cumulative Shear Work.

The new Cumulative Shear Work (Custom) property will appear on the Properties tab under Transient.

To highlight the surface triangles that are most affected by wear, we will create a Filter User Process to filter them out.

Now we will manually define the limits of the color scale and see how the surface wear changes over time.

To get a better context of how the particles are interacting with the geometry, we can create a Particles Trajectory process:

The results are shown below.

As shown in the images below, most of the shear work happens on the bottom portion of the main pipe.

Of the bottom portion of the main pipe, the regions of highest shear work are:

As the shear work is more severe in these regions, they are the most susceptible to wear.

This completes Part B of this tutorial.

Rocky was used to post-process a simulation of a Mixing Tee.

During this tutorial, it was possible to:

What's Next? If you completed this tutorial successfully, then you are ready to move on to next tutorial.