Chienman's profile - activity

openstudio-server: when is an analysis status = na? In looking through the openstudio-server code, it seems like there a

I have, and neither of those things solved my problem unfortunately. It's quite difficult to diagnose the rubies and ge

What is even MORE odd, is that I have other measures with the same require 'mongo' command and it seems to be loading fi

What is even MORE odd, is that I have other measures with the same require 'mongo' command and it seems to be loading fi

Someone mentioned checking my GEM_PATH. I've tried updating my GEM_PATH on my mac to the local gem folders on the machi

loading an external gem in a measure I'm trying to solve a problem that has occurred when updating to version 2.1 of Ope

It's my understanding that the outputs your specify in the spreadsheet are pushed to the mongo container of the OS-serve

what files to I need to run the OpenStudio-analysis-spreadsheet? Hi there- I've been having fun bootstrapping the analy

I get to answer my own question. The reason I am seeing this happen is because I did not remove the volume before creat

@BrianLBall where can I read more about compatibility? How do I know if the measure is 2.0 compatible?

I am using AWS actually. I had an old AWS instance that was about a month old. Yesterday, it started acting funny. So

I am using AWS actually. I had an old AWS instance that was about a month old. Yesterday, it started acting funny. So

Hi- That is a great question! My dockerfile simply states "latest" as in nrel/openstudio-server:latest You're righ

Hi- That is a great question! My dockerfile simply states "latest" as in nrel/openstudio-server:latest You're righ

Hi- That is a great question! My dockerfile simply states "latest" as in nrel/openstudio-server:latest You're right

AWS EC2 instance of Openstudio-server I've never had an issue launching a new instance of the Openstudio-server. For so

AWS EC2 instance of Openstudio-server I've never had an issue launching a new instance of the Openstudio-server. For so

AWS EC2 instance of Openstudio-server I've never had an issue launching a new instance of the Openstudio-server. For so

Hey Julien, is a command line in the terminal what you mean? I can do that, or try to

Thanks, that would be great! I'll ask to see what the correlation is between the dockerize branch and official AMI rele

Using the openstudio_meta executable getting "cannot load such file -openstudio" Hey everybody. I'm experimenting with

Hey Julian, is a command line in the terminal what you mean? I can do that, or try to

great question @ljbrackney. Not sure I know the answer to that. I provision my OpenStudio-server linux machine using t

Example to use the ReplaceOpenstudioModel measure Hey everyone, I've been running into a brick wall trying to get this m

Using the openstudio_meta executable getting "cannot load such file -openstudio" Hey everybody. I'm experimenting with

I've never tried setting up the server on a local machine, but from what I can see of the instructions, the Docker appro

Both pieces of advice were really helpful. It just turned out that I had gotten confused and tried to use a measure that was not a reporting measure. As soon as I used a reporting measure, my problem went away and everything worked great, so no real magic to report. Duh. Thanks for the help.

BTW-I did not set the measure to a ReportingMeasure, but it still worked just fine.

I have been toying around with the OpenStudio-server and Openstudio-analysis-spreadsheet. I have been able to successfully get all of this working properly thanks to some input from NREL (thanks for that)

Now I am trying to write my own custom reporting measure to work in this environment. When I try to follow the instructions online from here.

I try to use the runner.lastOpenStudioModel call but in the OpenStudio-server environment my measure seems to fail. Does anyone know why it does? Do I need to point the measure to look in a certain place? That is my instinct but not sure how to do this.

I tried following the Dencity measure that comes packaged with the spreadsheet, but couldn't get much insight there either.

Hi Everybody-

Thanks for your feedback. For what it is worth, I couldn't find anything that could quickly meet my needs, but all of your help and comments helped point me in the right direction. Good to see people posting comments!

Anyhow, since I'm really pretty wedded to the .NET environment, and I'm familiar with those code libraries and dependencies, I just decided that it would be easiest to make a library that I wanted with the functionality that meet my current use cases and publish it on Github. This first version lets you grab data from any epw file and returns the hourly data with some statistical analysis of the data. Github explains the rest. I'll be adding to the project over the next couple of days.

https://github.com/chiensiTB/EnergyPl...

Yeah, I'd just scrape it. You can write a recursive routine that just looks for the "btn-group-vertical" class and then iteratively loops through its children, terminating when it sees a "download all" InnerHtml on the anchor tag.

Personally, I like the UX for the weather site for this reason. It is pretty easy to write this iterative routine because of the consistency of the buttons and anchor tags.

Sorry, I don't have any python code for this, but I've done it in c# and javascript on a few different occasions now. JQuery makes this pretty easy.

Just curious if there are any open source projects for grabbing an epw file and turning it into classes with simple commands to extract and make use of the data (in any language).

Hi there.

There is really only one other company that uses shell geometry, and that is IES<ve>. I could take a look at the file for you and see if I could write a script to construct the shell geometry and space boundaries.

You might want to ask this company if they will accept geometry for the shell geometry and space boundary that is identical (zero thickness walls) because everything generated by SketchUp and OS is a very thin planar wall.

I do think that Honeybee is the best way to get more from your Rhino model. It is quite easy to collect your Rhino geometry inside of grasshopper, and use Honeybee components to then prepare the model for analysis. I am biased, I do work on the Honeybee project and contribute to it quite regularly, but there is only one other alternative (Diva).

A few features that are nice about Honeybee. You can have an un-zoned shape (no perimeter/core) and there is a component that, if the shape is not too complicated, will create a perimeter core zoned model for you, as well as break it up into separate floors. Many rhino models come as a single extruded mass. This is actually quite huge. The second thing, is that Honeybee also can automatically determine the difference between interior and exterior walls. This is something that, I believe, Diva cannot do.

Secondly, it is loaded with other features. You can assign space types, constructions, internal loads, schedules...all inside of the Grasshopper environment. You don't have to, you can do this elsewhere if you have a way that you like to do it. But you can make all of these changes.

Finally, there are two ways to define systems. One way is to use the E+ component, which creates an IdealAirSystem, with or without economizer. I have also worked quite hard on a series of OpenStudio components that automatically create AirLoops and PlantLoops for basic HVAC systems. (packaged units, and vav with chiller and cooling tower). You can also define fan powers and coil efficiencies inside of Grasshopper too, so all of this you can do inside of Rhino/Grasshopper without ever leaving the environment, and run the simulations.

Anyhow, there are a number of videos on this, all on YouTube. I personally think it is worth the investment to learn these tools. it does impress clients a great deal. Let us know (me, Mostapha, or Chris) if you need any help. There is a forum on Grasshopper. If you Google Grasshopper Ladybug Honeybee or something like that, you should see a pretty active user group that is monitored and answered frequently. Just sign up and ask questions. People want to help, and are nice.

YouTube:

• Mostapha's Videos
• Chris' Videos
• My Videos

Also, it is free and completely open source, so if you feel inclined, you can start adding to the source code.

Hi Everyone-

Just a quick question. Is there an example out there anywhere showing how to build up from simple (just a flat charge) to complex (demand, and ratchets) utility rate, leveraging the OpenStudio API?

I'm starting to want to add that feature to an app I'm working on (using C# bindings) and it would be helpful, in Ruby or otherwise.

Andrew, interestingly I thought all I have to do is call the nightCycleControlType()...but my fan did not get a new schedule as a result. It wasn't obvious at the time that this is what I should have done. Now I have got it. thanks for the feedback.

An old hacky way that can be used for estimation is to use two coils in series. The first coil is the waterside economizer coil and the second is a traditional chiller that comes on when the waterside economizer can't reach capacity.

Since you can specify the capacity of any cooling coil in IES, this will limit the off-coil temperature at some point, and the downstream coil will come on. If you want to get super-hacky, then you can also use a generic cooling source with a high COP to estimate the use of fans and pumps on the waterside economizer.

If you are concerned about shutting off this coil when outdoor conditions (wet bulb or dewpoint) are not favorable, then you can use a second controller monitoring outdoor air that can be hooked up to the first coil's controller to shut it off when outdoor conditions would make it impossible to make sufficiently cold water by evaporation alone. This is a quicker and dirtier way if you are stuck.

You might want to consider just using a spreadsheet for a data center, unless you think it really needs to be modeled in energy simulation. I know this is a cop out answer, but when the internal loads begin to dominate, the envelope loads become insignificant. This is what I've done for Facebook datacenters that I've modeled and submitted to LEED.

It is pretty easy to develop a spreadsheet using hourly weather data, fan curves, and internal gains with some logic and equations for economizer hours. Hope this is helpful.

I've been trying to figure this out, and I just can't seem to get past it. I am probably missing something simple because it is late at night. Basically, I'm using the OpenStudio.dll (C# bindings) at the AirLoopHVAC level, I have used OpenStudio's call to nightCycleControlType(), setting the value to "CycleOnAny". The AvailabilityManagerAssignmentList is updated with a n AvailabiltyManager:NightCycle and the NightCycle manager looks like this:

AvailabilityManager:NightCycle, Packaged Rooftop VAV with Reheat NightCycle Manager, !- Name Always On Discrete, !- Applicability Schedule Name Always On Discrete, !- Fan Schedule Name CycleOnAny, !- Control Type 1, !- Thermostat Tolerance {deltaC} 3600;

But, even though there is no zone demand, the fans still run and my main AirLoopHVAC node still has flow. What am I missing or not setting? I thought all I had to do was set this up at the AvailabilityManager level, and everything below it would be overriddedn. Is it something weird, like the applicability schedule name and fan schedule name can't be the same?

Hi-

I've noticed that the EnergyPlus documentation kind of tiptoes around what will and will not be exported to SqLite. The documentation just sort of says something like "there are limited capabilities for specifying SQlite generation."

Does anyone know of a resource that comprehensively explains what are/aren't the available variables and reports in SqlLite, or is there a general rule of thumb? I find myself just doing trial and error to figure out what is possible when Sqlite is enabled as a reporting mechanism

@macumber, this is working great now. I did have to replicate your code that you referenced on github, overwriting the idf produced by OpenStudio vial the C# dlls in Grasshopper.

Everything is working beautifully now in Honeybee.

I have a couple of related questions which I'll post in the AM on the site.

Dan-I took your advice and have been working on an implementation where I modify the idf file after OS has created it. One thing that I've noticed, though I'm not quite done with this exercise, is that in your github example showing source code for energyConsumptionByMonth...the sql query looks like :

"SELECT Value FROM tabulardatawithstrings WHERE \ .... ..."

The thing I notice when I open the sql file is that after we call OS and write to IDF, we have a table "TabularData" but not "TabularDataWithStrings".

I realize that this is not a real table. Wonder if this is somehow contributing

Also wanted to add...your advice on using the Queries was right on. There is not a ReportName WHERE ReportName="Building Energy Performance", so I think that confirms why nothing is returned. I find it interesting though that SQLite doesn't return an error even when that table is not present.

Dan, I think this is really great direction. I think that my only question, is: Is there not a way to add Table Monthly reports directly from the OpenStudio dll's, without having to rely on run manager? As far as I know, I still can't access the Run Manager via C#. We tried to at the hackathon but it failed to work at that time. If that is the case, does this leave me with having to post-edit the idf in my script after OpenStudio has created it?