Grasshopper

algorithmic modeling for Rhino

Hi All!

As a quick question for Chris and Mostapha, and anyone else who knows, I am trying to run a parametric iteration study, but it is going to take a long time for them all to run if it can only run one energy and daylight simulation at a time.

Is there a way to run the idfs created as a group run like what energyplus launcher can do with native energyplus?

Views: 3021

Replies are closed for this discussion.

Replies to This Discussion

Hi Elize,

I've looked at your file two questions.

- Is each branch of the data tree coming out of the Set EP Zone construction component meant to be one EnergyPlus simulation?

- Are the names are varied in the concatenate function accordingly?

Hi Anton,

I have set up the file so that it would be able to run one E+ simulation. The parametric input part uses Mostapha's iteration process to change one parameter at a time and then run the new E+ simulation.

- The Set EP Zone construction component is overwriting the constructions set to the surfaces from the building program template for each type (wall, floor, roof, etc). This component is just editing the constructions in other words and continuing the workflow to eventually run a simulation. If you have a look at one of the parametric inputs that I want to vary, you could see that I am changing the wall construction parametrically. The keywords need to be different to pull up the write construction from the library that I have created in the userCustomLibrary.idf

- The concatenate function allows the naming of the files for the parametric input that has been changed. So when you first run a simulation, the name of the file is what the current inputs are set to. If you then animate the sliders, manually or using myslider component, then the file name will change when the parameter changes. This is simply a way to make sure no data is being overwritten.

Hi Anton,

Here's a simple model, I have put some question inside that I was not able to figure out. Right now the schedules are simple as I just want to know how to run everything parallel.

Thank you for the help!

Vinu

Attachments:

Hi Vinu,

Really sorry for the delay in my reply I have been very sick this last week. Since Mostapha has made a component that will run IDFs in parallel I haven't attached my own component here as it is under construction. However I've added your GH file with notes about the questions you asked.

Forgot to attach the file.

Attachments:

Hi Anton,

Thank you for the comments! Would like to hear what Mostapha and Chris has to say about it! 

Vinu

Hi Elzine,

I live in Vancouver, its good to know that you are working on daylight simulation, I am working on human comfort, currently focusing more on thermal comfort and later might consider other comfort parameters. 

Vinu

Hi Vinu,

I am working on both, focusing on mainly how they could be coupled together to provide feedback on how sensitive the parameters which we use to test the performance are using human comfort critieria. Honeybee obviously is amazing at being able to assess both, but I may do some tests to check the simplification of models and their use for accurate daylight simulations...

Elzine

Looks like this discussion is still open. I attach a file that runs a list of batch files in parallel in background. Honeybee already generates a batch file inside the folder once you generate the geometries. If you connect the list of batch files to this component it should run them in parallel.

to find the batch files you can use os.listdir() and filter the files. Check Honeybee_Lookup folder for an example.

Attachments:

Hi Mostapha,

Thank you for this, however the batch files for each idf are only generated once you run an E+ simulation - at least in my case anyway. If you just ask Honeybee to write the idf but not run them, then only an idf is generated.

I want to be able to create a bunch of idfs and then simulate them all in parallel...

I noticed that the writeBatchFile coding is under the RunIdf definitions within the component, can this be moved to happen when you write the idf instead? (in the process of learning python myself...)

Hi Mostapha, 

Thank you! The batch files are created after the energy plus simulation is done (same as Elzine mentioned about). I need to run parallel idf's. Instead of the batch files, can we do it for multiple idf's? And also when I tried running the parallel batch  files the energy plus was getting aborted, but it was running the simulation for single files.

Hi Elzine and Vinu,

My bad! I assumed that .bat files get created with setting write to True. I have modified the re-run Idf component and used it next to the component that I shared with you and was able to run 60 simulations with no problem on my laptop!

Vinue, my guess is that your simulation failed since you're running the files in the same folder. EnergyPlus generates a number of files before running the study that will have the same name regardless of your file name. That can be the reason for the issue. The attached component creates a separate folder for each file and copies idf file and weather file inside the folder and writes the batch file.

Mostapha 

Attachments:

RSS

About

Translate

Search

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service