FlexSci

Computational Fluid Dynamics Analysis

Backgound

Computational Fluid Dynamics (CFD) is a method of calculating the velocity, pressure and temperature fields in a region of space occupied by a fluid. Essentially the space is divided up into a large number of small volumes (cells) defined by a 3 dimensional grid. The non linear partial differential equations governing the conservation of mass, momentum and energy in the fluid are then solved by a numerical method on a computer. Because the region of space is being artificially separated from its surroundings it is necessary to define the conditions the fluid experiences on the boundaries of the region (the boundary conditions) since this will influence the flow of fluid inside. The more accurately the boundary conditions and surrounding geometry can be specified the more accurate will be the calculation inside the region.

The Process

In practical terms the process usually starts with a CAD drawing of the component, building etc. The drawing is usually supplied by the client is a generic CAD format such as iges, parasolid, iges, step, stl, dxf, etc. A computational grid is then generated inside the region with the highest concentration of cells being placed close to solid boundaries and where the gradient of key variables are likely to be large. Boundary conditions are then defined. These could be as simple as an uniform inlet velocity or pressure or more complicated conditions such as an atmospheric boundary layer over rough terrain. The calculation is further defined by any particular models required, eg any porous media characteristics, transient analysis or particular turbulence models. The properties of the fluid itself must also be defined ie. density, viscosity and specific heat capacity. Finally the parameters relating to the numerical method are selected and the calculation is started. Typically the solution will converge in a few hours, after which the quantities of interest can be extracted from the raw results using sophisticated post processing software. Further sensitivity checks are then carried out, by repeating the calculation with different parameters to ensure that for example, the grid is sufficiently refined to capture the gradients in the flow field.

Why use CFD ?

CFD is one of several CAE tools in common use today throughout industry and academia, Finite Element Analysis (FEA) uses very similar techniques for the analysis of stress and vibration in solid materials. Both methodologies give the engineer the opportunity to rapidly assess different design scenarios without having to resort to prototype manufacture and testing. In addition there are circumstances where it may be impractical or even impossible to create a suitable experiment, for example where scaling arguments cannot be applied, or intrusive measurements are not possible. Advanced post processing tools enable the results from a CFD calculation to provide detailed insight into how the fluid is behaving, for example streamlines, particle tracks, velocity vectors and pressure contour plots. In addition integrated quantities such as normal and shear forces or surface averaged values are naturally extracted from the results providing the engineer with information to measure performance and inform how design changes might be made. As computing power has increased and engineers have gained confidence in CFD its use has become more widespread and more integrated into the design cycle. This trend is set to continue.

1D Powertrain Simulation

FlexSci's CFD engineers use an integrated set of CAE tools for the design and analysis of engines, powertrains, and vehicles.  These tools enable detailed engine, powertrain and vehicles to be modeled and operated virtually.  In addition to performance and acoustic predictions and cycle analysis for fuel economy and emissions, we use these tools for engine heat management and cooling system analysis.

1D Network Analysis

Optimisation of the flow rates and temperatures within both the air-circuit and water-circuit can be done more rapidly using computational network analysis. FlexSci has been at the forefront of implementing steady state and transient network modeling using industry leading proprietary software, where necessary augmented with FlexSci developed bespoke coding.

3D CFD Simulation

Few other engineering consultancies have the knowledge and experience that FlexSci has of applying Computational Fluid Dynamics (CFD) to improving the design of powertrain cooling and climate control systems and components. Examples include optimisation of:

• Underbonnet airflows
• Exhaust and air-induction systems
• Catalytic converters
• Bake cooling
• Ducts and registers
• Heat exchagers
• Simulation of windshiel defrosting