Rapid Flow Modeling with FLUENT for CATIA V5 – Part 1

In this series of articles we will discuss some technical highlights of how FLUENT for CATIA V5 was used for a number of industrial applications. This first part introduces the software and provides thumbnails of the diverse application areas.

FLUENT for CATIA V5 is flow modeling software that is fully embedded inside CATIA V5. It was created in the context of Dassault Systèmes’ Software Community Program. The user never leaves the familiar CATIA GUI (graphical user interface) – please refer to the screen snapshot in Figure 1.

Figure 1. FLUENT for CATIA V5 screenshot: Velocity vectors indicate the flow of coolant and contours show the temperatures at the surfaces of the chip cooling plate

The philosophy during the development of FLUENT for CATIA V5 was to build a tool that can be used by both designers and analysts during all stages of the design. Under the hood is the state-of-the-art FLUENT solver. Moving the use of CFD (computational fluid dynamics) into the conceptual design stage can help shorten the design cycle and avoid costly flow-related design changes later in the process. Because of this acceleration of the design cycle and because of its ease of use, we refer to this tool as rapid flow modeling software.

From the user’s perspective, the process goes like this: The engineer brings up the model of the solid part and then launches the FLUENT for CATIA V5 workbench. This creates a new analysis file, and engineers who have used other embedded CAE tools (such as the GAS and GPS structural solvers) will feel instantly at home: they simply work their way down the product tree.

At the top of the product tree we find a FLUENTProduct consisting of the original part and a new FLUENTPart. Moving downward in the product tree, the engineer selects “geometry definition” to define the inflow and outflow ends of the original part. The software automatically generates a new part that represents the volume occupied by the gas or liquid – we call that the flow volume and store it as the FLUENTPart.

With the flow volume defined, the engineer proceeds to the meshing. FLUENT for CATIA V5 relies on enhanced versions of Dassault Systèmes’ FMS and FMD meshers, so that many engineers will feel at home here.

With the mesh done, the engineer specifies whether or not the solid portion from the original part will be included in the analysis. For the above example it would be, since we must include the conduction in the solid cooling plate in order to find out the temperature at the seat of the chip. The designer also specifies whether the flow is laminar or turbulent.

The final step for the set-up is to click on the different inlets and outlets of the flow volume and to assign boundary conditions to them. The conditions might include pressures, flow speeds, temperatures, and turbulence values.

When the engineer clicks the compute icon, the FLUENT solver is launched in the background, perhaps even on several separate computers. The solver steers itself automatically. When it is done, it writes out FLUENT results files (case and data files) and the engineer visualizes the solution using FLUENT for CATIA V5. The postprocessing and reporting tools will again look familiar because they are an enhanced version of the tools used by GPS. To be consistent with good PDM practices, the FLUENT results files are written into the same working directory as the analysis file. A great benefit of storing the data in this manner is that the results files are fully compatible with the full FLUENT solver. For example, the analyst might start with the designer’s preliminary studies and add more complex physics to the problem.

The real power of the truly embedded solution emerges when the original part file is fully parameterized. A change in the original part file is immediately communicated down the product tree and prompts the engineer to compute a new solution for the altered part. At the highest level, the engineer can use some of the parameters to serve as design variables during an optimization. The objective function during this optimization loop might consist of a target value for a sensor that has been defined in the product tree.

The ultimate vision is that the optimization problem will include not just constraints and objective functions derived from the technical requirements. Instead, the designer will benefit from constraints that are based on knowledge from similar, existing designs. Here the power of PLM shows itself in the form of re-use of older designs and application of captured knowledge. Similarly, the optimization will include constraints arising from product requirements and captured business knowledge.

Figure 2. Rubric showing the application of FLUENT for CATIA V5 in diverse industries

In upcoming parts of this series, we will highlight examples from specific industries. Figure 2 shows a rubric of thumbnail sketches of some of those examples. The author’s next contribution will be a discussion of how the enhancements of the FMS and FMD meshers have yielded meshes for entire airplanes.

For more information on FLUENT for CATIA V5, please visit: www.fluentforcatia.com. Also visit www.fluent.com for more general information.