CFD has it all wrapped up
Wrapped in a tightly meshed grid, the volunteer is enveloped by lots of little cells that tuck snugly around the body, filling all the available space. Scientists from the Fraunhofer Institute for Building Physics IBP want to understand every little detail, and so have every millimeter covered. Because they know their test person’s exact measurements and understand how these change in different positions, the meshed grid fits them like a glove. The grid in question consists of a million or more cells and took the scientists several weeks to knit around their test subject. How fortunate, then, that in this case the volunteers aren’t real people, but characters in a computational fluid dynamics (CFD) simulation. CFD is used whenever scientists and engineers need to understand and predict the flow behavior of fluids such as air, water and oil. It’s a technique that Sebastian Stratbücker, head of the Simulation group, and his team mainly use when investigating issues relating to indoor climate. In most cases, this involves looking at air and the way it behaves in enclosed spaces. "Our simulations help to optimize the indoor climate for people and technical equipment," explains Stratbücker. "Evaluation parameters include thermal comfort, energy efficiency, humidity, CO2 and pollutant concentration."
Say you want to construct a new building, for example. The planning phase is all-important, and requires those involved to make a lot of important decisions before work has even begun, such as which ventilation system to install. Planners need solutions that will help them make such choices in advance; they have to be able to determine whether the system they intend to install is up to the job while also being efficient and ideally cost-effective at the same time. CFD offers just such assistance, and enables planners to measure and evaluate rooms in great detail with the help of the grid described above. This system allows them to call up the indoor climate conditions for any point in the room at any given point in time, highlighting factors ranging from air velocity, temperature and pressure through air exchange rate and the concentrations of specific substances in the air.
Of course, the boundary conditions for the simulation must be clearly defined; these parameters are often determined by measurements taken in laboratories or in field tests. Fraunhofer IBP scientists also use their own specially developed DressMAN 2.0 measuring system for this purpose. In other cases, they rely on their own databases of building physics reference data, types of construction, building services, and usage profiles. They draw additional data from their customers’ plans. All this allows the computer program to evaluate factors such as the geometry of the space in question, including any air inlets and outlets, exchanges of air between the outside and inside environments, the heating and cooling capacity of the chosen system, the periods when the space is in use, the materials used in its construction, and many more features. Even the physical properties of the windows or the type of clothing worn by the people using the building can be taken into account if required. "It goes without saying that CFD is also quite an intensive process, but one that’s worth it compared to the trial-and-error principle, as it means we avoid making mistakes from the outset," Stratbücker explains, adding: "Simulating different scenarios lets us work up a range of proposed solutions and analyze them, so we end up with the optimum system design."
This demonstrates how flow simulation can benefit individual components, such as those in a ventilation system, as well as large systems, such as entire rooms. "We use CFD in any situation where we’re not certain if the system works in the way we think it should," says Stratbücker. "But we also use other methods, depending on the issue we’re addressing." For instance, particle image velocimetry (PIV) can be used to measure and visualize flow field velocities – which means it can help to verify CFD results at critical points.