Solar Access Analysis Help

[polatalihan]

Hi,

I am working on a Solar Access Analysis on Ecotect 2011. The analysis is done for an old building with stained glass windows in New York City and the ultimate goal is to learn which windows and parts of the windows are more exposed to heat gain due to sunlight. The 3D model we are working on for the historic building is created through a cloud model with laser scanning. The model is optimised to eliminate unnecessary detail and brought into Ecotect after simplifies in other CAD software. The model is geographically in correct location and we are using a New York / Central Park based climate file. 

The problem we are having is the inconsistent results with in the same windows. The model, by nature is triangulated and we optimised in Ecotect by merging the coincident triangles but still there is a lot of triangulation in the final model on windows (see attached snapshots). We ran a cumulative solar  access analysis from March 21st to September 21st from 9AM to 5PM everyday. The calculation took quite a time like 3 days and the results came in. I am attaching some snapshots from certain parts of the model. My question is, why within the same window, different faces have significantly  different solar radiation levels? Is it because the unit is Wh/m2 and its basically giving an average value based on how big or small that face is? Also is there a way to show more detailed analysis on windows or do they have to be different faces to be represented. In other words, in order to see what part of one window has more solar radiation than the other part of the same window, do we need to create fine grain grid on the windows so rather than seeing one color we can see a gradient through the window surface. 

I really appreciate any help on this matter. Please let me know if you have any questions.

[arqyt]

The model certainly has complexity, becasue the discretization of curves is very thin. In the calculation PreProcessor (shadow masks in the process of interzonal adjacencies) will be very slow and polygons so as tofine the parameter "surface sampling grid point accuracy" should be proportional. You ought to check "surface normals", and that will be very, very complicated issue with these polygons soslender.

I advise simplifiques on all around the windows (just the ones you want analysis) with a coarse of 7-8 polygon vertices, the values ​​are means and effort will be rewarded with a reduced calculation time.
Hope this helps
best regards

[polatalihan]

Hi Jose,

Thank you for your answer. You are right. It is such a complicated model. One problem I face is, whatever I do to reduce the detail on model when I import to Ecotect, the triangulation on the windows always happen. Even if they are a single polygon in other software. I will try to redraft them from scratch and use a less detailed curve to do the tops as you suggest. Thank you!

Best,

a.

[arqyt]

You can redraw it in the original software, or i think directly in ecotect may be much easy. 

Good weekend