Idealization Process for CFD Simulation

Gain understanding of the CAD geometry idealization process for CFD simulation by walking through the preparation of a building’s corner office space.

Every project has a unique set of objectives, time frame, complexity, and design data.  This makes CAD idealization project specific and impossible to detail for every situation.  Nonetheless, this module includes an overview of an effective process for approaching AEC CAD idealization. For more, watch this CAD Module Video.

By relying on CAD idealization for CFD simulation, you will capture design intent with the minimum amount of necessary complexity. You will start simple and build in complexity as you successfully complete analyses, and explicitly control important design variables as you iterate towards a viable design solution.

CAD Idealization Approach

A simulation project requires geometry.  That geometry is either created specifically for simulation or it already exists as a CAD file.  (Creating new CAD will not be covered here as it should be straightforward and well documented by your CAD platform.)

Existing project files were most likely developed with non-simulation priorities in mind.  This leads to a need for CAD idealization specific to simulation.  Idealizing existing geometry is unique to every project and CAD platform; however, there is commonality that will be presented here as an effective approach to optimizing existing geometry for simulation.

Idealization Flow Chart

Idealization Flow Chart

Creating Geometry for CFD Simulation     

New geometry for simulation specific tasks often needs to be created. And it can be useful to create this new geometry referencing existing entities; the most common AEC entity being an air volume or fluid domain.  Further, portions of the geometry may be unnecessary as in the example discussed here.

Simulation CFD can automatically create volumes (see Help), however, meeting the fully enclosed constraint is sometimes more difficult or time consuming than generating the air volume manually.  This is because idealizing a handful of components (walls, floors, and ceilings) would require more effort than a single extrusion of the volume that is enclosed.  Geometric representations of those components can also be considered unnecessary as they can be accounted for as boundary conditions.

Office space to be simulated (top left) with an idealized wall (top right).
Sketch referencing inner walls (bottom left) is used to create air volume extrusion (bottom right).

Creating new geometry for simulation, by referencing existing geometry, is a great way to develop a clean simulation model that captures design intent.  This strategy if often less time consuming than modifying and troubleshooting existing data.          

Example: Corner office air flow analysis

The air flow and thermal distribution of an office space (top floor corner office) is to be simulated.  


Office space of building to be simulated.

Launching the entire building (as is) into Simulation CFD is unnecessary and not possible to directly simulate.  Launching marginal geometry introduces the opportunity for error.

The simplest way to start would be with a single extrusion, representing the office air space.  The extruded air volume can reference existing geometry to expedite creation.

Extruded air volume (left) representing the corner office air space (right).

With the air space represented, supplies and returns will be added next to move the air.  Again, simple extrusions provide the necessary geometric detail to apply the appropriate boundary conditions (define supply and return air movement). 

NOTE: The extrusion length of outlets should extend enough for fluid flow to fully develop prior to entering or exiting the system.  A good “rule of thumb” is to start with a length equal to two (2) times the hydraulic diameter.  A Knowledge Base article provides further reference.    


Multiple extrusions representing the office air space, supplies (1), and return (2).

TIP: Creating new geometry allows you to explicitly control critical simulation dimensions for optimization, including supply size and location.

Setting up and simulating this simple model will provide insight related to simulation inputs, performance, and results.  For instance, resolving solver errors due to improper boundary conditions on a very simple model will take much less time than a model with numerous volumes and surfaces. 

The results may even offer quick design guidance (based on the assumptions) associated with the office space flow characteristics.  


Air velocity in corner office shows supply air moving directly to the return (arrow).

Complexity can be added as necessary to meet your objective; additional details would consist of new components or the re-purposing of existing ones.  Details such as humans, furniture, structural members, stairs, and more are only modeled as required after simulating simple air volume models.           


Simplified air volume results superimposed over detailed CAD for presentation purposes.