Microclimate Hardscape vs Softscape material/geometry/simulation issue

hbzones
epmaterial
radiation
microclimate

#1

Hello everyone! :slight_smile:

I am facing a problem with generating an outdoor microclimate map.
In particular, I am trying to see the difference between hardscapes and softscapes, in order to see the benefit of green areas withing the site.

First I was comparing different types of soil, and didnt see much difference in results (see the images bellow).

This is the first thing that made me confused… So I decided to use green roof material from this example file, instead of basic soil types.
And I faced another problem, where OpenStudio just couldn’t handle this simulation. There are few discussions about this topic on this forum, but they did not help me with the outdoor calculation.


It always pops up different errors, and I even tried different versions of Export to OpenStudio component.
I figured that the issue pops up after solving adjacencies. So I tried to unplug the “Solve adjacencies” component from “Export to OS”, but kept it plugged in to the “Indoor View Factor” component, otherwise “Outdoor Comfort Recipe” will think that I have some indoor spaces.
It finally run the simulation, but the results are weird. The meshes that are assigned with “Vegetation” are warmer than the meshes with assigned “Asphalt” materials.

I was trying to reduce Specific Heat of grass material, just to see if it will show me different results, but it didnt.

What can be the issue?
Am I using the right approach to my goal?
Or perhaps there is a different method for outdoor microclimate map with different surface properties?

I attached the GH file bellow.microclimate.gh (727.6 KB)

P.S. There are a bunch of similar topics on this forum, like this and this , but they are quite old and did not help me with my issue :frowning:

Thank you!!


CFD Outdoor Analysis
Outdoor comfort UTCI microclimat map || Vegetation limitations
#2

I suggested you using Envi-MET for this analysis.


#5

Is there a solution to assign different materials for microclimate simulation thru export to open studio? I am doing it by splitting the ground surface on two surfaces, then assigning different materials by “Create HB Srf” and applying 1 layer constructions to those surfaces. Basically using the same logic as with energy simulations. Just the results are weird. :frowning:


#6

Thank you for the reply!
Is there a way to Import the Envi-MET results and meshes back into GH?


#7

I was trying to analyze the problem myself, and I believe the problem is with OS simulation.

Tested different properties of different materials. Soil properties were taken from E+ website
Asphalt vs Concrete; Green Roof vs Soil


The results were still incorrect.

Tested different methods of Microclimatic analysis. With boundary conditions and without. Still - incorrect



Tested geometry. The geometry seems to be alright.


The result creates 3 meshes: 1- Outoodr microclimate, where softscape and hardscape are one mesh; and 2 - Indoor micrcolimate surfaces, where I located my non context buildings.

Also tried to use one surface analysis instead of 3 (roads and ground). The result is the same

The results appear to me that the surfaces with assigned material AC (Asphalt/Concrete) is cooller than the surface withh assigned Soil/Grass materials. Whereas the difference between AC and Grass/Soil should tromendous, based on this study. So even when simulating typical soil vs concrete types (3 vs 4) that can be assigned in “HoneybeeCreateEPGround” doesen’t show the right results neither. (You can see the results in my first reply)

Thereby I assume the problem is with OS simulation… Or my crooked hands :smiley:


#8

I tried to simulate two different zones, where ground is one closed brep, with one material, and the road is another closed brep with another material.
When solving adjacencies, the simulation crushes, since OS understands the road geometry as “Ceiling”

I tried to run the simulation without solving adjacencies. As a result, the surface temperature of the road is, again, cooler than the ground surface.

gh file: testing.gh (766.0 KB)

BTW this time as the base was used this example file.


#9

For exporting/importing to/from ENVImet see this.
-A.


#10

Thank you very much Abraham. I will check it out!


#11

So should I completely give up combing materials for a microclimatic analysis thru Honeybee?


#12

Hi @mrbro,

this is an example of study with lb_envimet within GH.
The stages are:

  • make a model with GH,
  • generate the geometry ENVI_MET model with lb_envimet,
  • generate a simulation setting file with lb_envimet and run the simulation with ENVI_MET.
  • Finally, read results with lb_envimet.

01_
model with GH

03_
exporting to ENVI_MET
02_

I have two scenarios:

  1. one where the soil is made prevalentely by loamy soil
    image
  2. another one where the soil is made prevalentely by asphalt
    05_
    For each scenarios I have done a simulation using the same settings. Finally, I have made a unique map combining the two air temperature maps of both simulations. The unit of this new map is °C.
    Red color means that scenario 2) gives you almost 1 °C of air temperature more than the scenario 1).
    White color almost 0.5. Blue is netrual.
    06_
    You can calculate and check also other variables, such as relative humidity, mean radiant temperature and so on. Or combine them using thermal comfort indices and do the same with maps.

Regards,
Antonello


#13

Thank you Antonello!
Its very clear!


#14

Hi @mrbro

I found your microclimate experiments interesting and educative as I try to go through the same process. What you finally decided ? Have you finally solved the problem with your honeybee script or you had to go to through LB_envimet process. Envi met is quite expensive…

Any ideas ?

Best,
Charles C.


#15

Hi Charles!

I ended up creating a script that will fake the results, since I needed mostly a graphical representation and some statistics upon how hardscape reduction would improve the site. Also, I had a deadline and it was the only solution I could come up with at that moment.
It basically just empirically increases the temperature of the meshes where i decided to plot asphalt.