The Incompressible LBM analysis type is used to run large transient external aerodynamics simulations when the Mach number is < 0.3 and where the geometry usually involves large domains.
It is a GPU-based solver using the lattice Boltzmann method (LBM), developed by Numeric Systems GmbH, Pacefish®\(^1\). It has the ability to run on multiple GPUs in parallel giving highly accurate results for transient simulations at running times reduced from weeks and days to hours and minutes.
Within SimScale, one can effortlessly set up an incompressible LBM simulation with the steps described below.
Creating an Incompressible LBM Analysis
To create an Incompressible (LBM) analysis, first, select the desired geometry and click on ‘Create Simulation’:
Next, a window with a list of several analysis types supported in SimScale will be displayed:
The Incompressible (LBM) analysis type is restricted to specific professional plans. For more details please visit our product & pricing page. In case the user has this options disabled then please reach out to us.
Choose Incompressible (LBM) analysis type and click on ‘Create Simulation’. This will lead to the Workbench with the following simulation tree and the respective settings:
To access the global settings, click on ‘Incompressible (LBM)’ in the simulation tree. Here you can define the turbulence model that needs to be applied for the simulation. A choice between RANS, LES, and DES turbulence models is available.
Detailed information about each of these turbulence models can be found here.
The Geometry section allows you to view and select the CAD model required for the simulation. It is important that the CAD model is well prepared to avoid any meshing or simulation-related errors. Find more details on CAD preparation and upload here.
Did you know?
While uploading your CAD model for LBM or wind comfort simulations it is recommended to toggle on ‘Optimize for LBM / PWC’.
This option allows you to import a *.stl file that is optimized for the incompressible LBM and wind comfort analysis types. It leaves out complex import steps like sewing and cleanup that are not required by the LBM solver and therefore also allows to import big and complex models fast.
External Flow Domain
Here you can define the cuboidal flow domain that acts as a virtual wind tunnel to perform the fluid flow simulation. You need to define the following parameters:
Minimum: The minimum values for x-, y-, and z-coordinates
Maximum: The maximum values for x-, y-, and z-coordinates
Rotation point: The point about which the flow domain can be rotated. This point can also be picked from the viewer.
Rotation angles: The angles about x-, y-, and z-axes. Positive values will rotate the domain counterclockwise about the axes (right hand rule).
Incompressible LBM simulations currently only support air as the fluid medium. However, air and other gases, at different temperature and pressure conditions, can still be replicated by adjusting the material properties.
Boundary conditions help to add a closure to the problem at hand by defining how a system interacts with the environment. Unlike other analysis types, incompressible LBM has a fixed set of boundary conditions assigned to respective faces labeled A through F.
The available boundary conditions and their corresponding faces are listed below:
Velocity inlet (A)
Pressure outlet (B)
The fundamental boundary conditions used besides the velocity and pressure are the wall boundary conditions.
All the remaining faces of the CAD model are treated as no-slip walls, that is the velocity is automatically assigned a zero value. The user is not required to assign them separately.
The Simulation control settings define the general controls over the simulation runtime. The End time and Maximum runtime for the simulation can be defined.
End time: This represents the time for which the transient effects need to be analyzed (e.g. the physical time).
Maximum runtime: This is the time value in seconds that decides the maximum time for which the simulation will run on the cloud computing instance. The simulation run is automatically canceled if this value is exceeded. This value is to be used as a safeguard to not run excessively large simulations that could unexpectedly use a large amount of computing quota.
It is advised to read our knowledge base article on controlling simulation time and time steps in LBM for details.
Under Advanced concepts, you will find additional setup options, such as Surface roughness, Porous objects, and Rotating walls. Rotating walls can be applied to cylindrical surfaces which rotate, such as wheels.
The Result Control section allows users to define additional simulation result outputs. It controls how the results will be written meaning the write frequency, location, statistics of the output data, etc.
Meshing is the process of discretization of the simulation domain. That means we split up a large domain into multiple smaller domains and solve equations for them.
For an incompressible LBM analysis, the meshing is based on the lattice Boltzmann method (LBM) and is quite different from the finite-volume based fluid dynamics analysis types in SimScale. Here a cartesian background mesh is generated, which is composed only of cube elements that are not necessarily aligned with the geometry of the buildings or the terrain.
The mesh characteristics are largely the same as described under the PWC analysis mesh documentation. However, the computation of the region of interest and reference length follows a different logic. Users are strongly advised to read this document before proceeding further.
Depending on whether the mesh settings chosen are automatic or manual the subsequent settings available are as follows:
For Automatic mesh settings:
Fineness: The mesh density can either be varied from very coarse to very fine. This can be specified here. Alternatively, a minimum target size can be specified under the Target size option. This ensures that the minimum cell size in the region of interest is at most as large as the target cell size (it can be lower). The exact value is reported in the run info.
Reference length computation: This is the longest characteristic length scale of your objects of interest in the simulation. For example, this can include the height of a building or the span of a sports stadium. There’s an option to let the algorithm decide it automatically or enter it manually as a Value.
Region of interest:The logic for the region of interest is different from the one for the PWC analysis. For incompressible LBM analysis it can be either be a:
Region defined using a geometry primitive or;
Part(s) where the user assigns face/s or volume/s directly from the geometry.
For Manual mesh settings:
Fineness: Same as above except that the Target size option is no longer available.
Reference length: Same as above.
Reynolds scaling factor: In SimScale, the Reynolds scaling factor (RSF) can be used to apply scaling automatically to a full-scale geometry. This is usually required when tests include scaled versions of buildings or planes in subsonic flows. For more information click here.
Local Mesh Refinements
In case a region between two buildings or a specific open area of interest needs refinement, a region refinement would be best suited. To assign a specific region for mesh refinement, a geometry primitive of type sphere or cartesian box can be locally created using the ‘+’ button and assigned by activating the slider in front of it.
The sizing for the region of interest is managed either automatically by choosing a fineness between very coarse and very fine or by manually setting a target resolution. The parameter Target resolution defines the length scale to which the entire assigned region will be refined.
A surface refinement is best suited for cases where a specific building/volume or a set of its surfaces or solids should be refined.
The fineness can be defined analogously to the global mesh sizing from very coarse to very fine. Also, in the same way, as it is done for the global mesh sizing inside the region of interest, the surface refinement will result in a layer of 4-6 cells ranging from the smallest cell size near the surface and gradually increasing with a larger distance from the surface.
When in manual mode, the user gets to define a minimum number of these buffer layers. Per layer, an additional number of cells in each of the positive x-, y-, z-direction can also be specified.
Strictly Necessary Cookies
Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.
If you disable this cookie, we will not be able to save your preferences. This means that every time you visit this website you will need to enable or disable cookies again.