IES-VE User Forum

Hello!

I am trying to compare IES and CFD (Flovent) induced airflow rates of a naturally ventilated building with double skin façade. I am using the steady state condition (no wind ; no solar radiation, outside temperature = 25C) on both models. The heat sources defined are internal gains (10.8 kW) into each floor of the model. For the IES simulation, I modified the weather file, so it has the conditions defined (outside temperature = 25C, no sun and no wind the whole year). On IES the “techniques for modelling flow in facades and flues” was applied.

I observed the same tendency on the decreasing of the NET airflow rate on the windows over the floors on both software. However, the results have showed that the induced airflow rates on IES are underestimated in relation to Flovent (on the windows of the floors and on the cavity openings). The air mass conservation showed a good agreement on both software. I tried several modifications on both software* in order to trying to have similar airflow rate in the cavity and in the floors. However, in all attempts minor airflow variations were observed and the induced airflow rates on IES are still considerable underestimated in relation to CFD.

* Attempted tests:

1. I performed different simulations varying the windows size on both software. I observed similar tendencies of airflow decreasing on the upper floors on both software. It shows that the problem looks not to be the bidirectional flow algorithm.
2. I increased the discharge coefficient (openable area > 100%) on the cavity top opening - on IES. It did not resulted in variation on the airflow on the floors windows, but only on the top of the cavity opening .
3. Modification of the algorithm calculation - on IES  I tried all the available “internal convection model” of IES, but the variation on the results were still very small.
4. Modification of the surfaces roughness - on Flovent - I observed a small decreasing of the airflow rate on CFD by applying surfaces roughness = 0.05m, instead of smooth walls.
5. Modification of the air relative humidity - on IES - I modified the relative humidity of the weather data (50 and 80%).

I found some references describing the limitations on both packages in simulating the ventilated double skin façade behaviour. Although in building energy simulation software (as IES), the Alamdari & Hammond calculations is the most indicated for double skin buildings, it is still not 100% suited to narrow cavities where boundary layers can interact.
I want to improve the consistence of both models, so I can have confidence that the results are as closer to a real condition as possible. In terms of judging what prediction is the best, some authors are more likely to favour the CFD calculation. However, I want to make sure I covered all the possibilities before making the assumption that the modelling of double skin facade on IES presents limitations.

Is there anyone that have any idea how I could solve my problem, either decreasing the airflow rate on Flovent or increasing it on IES?


Best regards,
S.

Sabrina,

You are really carrying out an apple v/s orange analysis.

IES is a bulk airflow analysis model. FloVent is a CFD model. The entire zone in the IES model is assigned a single flow rate, whereas the the zone has completely resolved flow rate in CFD model.

To produce similar results, you would need to induce stratifications within your IES model. Eg. suppose a 3 storey building with a double skin facade. You cannot model the double skin as a single large flat zone. You would really want to model is a stratified zone with maybe 6 or 9 zones on top of another. You would also want to divide it into squares.

Plus MacroFlo is designed to work well with external conditions when wind is included. If you look at the userguide or the calculations method document, most of it is wind based numbers. When you take wind out of MacroFlo's sail, it is not going to work well is it?

Also can you tell us about the CFD boundary conditions?