arfc / predicting-the-past Goto Github PK

Dearest @gyutaepark,
I am wondering why you'd need the recipe cool_spent_uox, since Cyclus can do decay. Considering that you have the option to choose from various burnups, and perhaps for future implementation purposes, wouldn't it make sense for Cyclus to do decay calculations than manually setting what the 'cooled' composition would be?

I'd like to hear what you think.
Thanks

Dear contributors @scopatz @gonuke @katyhuff @eriewski @gyutaepark,

It has been discussed in the ARFC group meeting that this benchmark would be merged with arfc/transition-scenarios for a more organized, centralized repository for Cyclus fuel cycle scenarios. I was wondering if all parties involved (Wisconsin groups) were okay with this change, for we are unsure if CNERG was still working on this.

This issue can be closed when all parties express their opinions on the merge and a conclusion whether to merge or not has been reached.

• Work with others to determine the appropriate range of burnups to represent
• Find (online) or generate (with origen) fresh and spent fuel recipes for a range of burnups

Once #32 is complete, @katyhuff should review the US.ipynb and europe.ipynb files, considering how to archive them as 'living reports' more officially than the current state.

In your virtual machine, install git. Use the linux instructions here, or type:

sudo apt-get install git

Once #21 has been performed and confirmed with Professor Huff,
perform the same analysis for all nuclear power plants around the world.

Fresh fuel recipe used for simulations are not the ones specified in vision recipes.
As discussed with Professor Huff, they should be updated.

Now that the US region has been performed, a same analysis should be done for the EU region.

Go to https://github.com/cyclus/cyclus/tree/v1.4.0-release#installing-dependencies-linux-and-unix and follow the instructions.

US simulation and analysis should be updated using the PRIS reactor text file as discussed with @katyhuff .
The PRIS data file should be more up-to-date and accurate.

This should be completed and updated before #22 .

The Cycamore (Recipe) Reactor has a particular schema.

• Read through this section of the docs and understand the input variables: http://fuelcycle.org/user/cycamoreagents.html#cycamore-reactor
• Compare the Cycamore:Reactor input variables to the Bright lite input variables, in order to understand how your input files will vary from the fleetcomp input files.
• Note that you will need, in particular, input and output recipes representing the reactor fresh and spent fuel batches. Consider how you will construct these. (we will likely need to run origen)
• To demonstrate understanding, view the Bright-lite:ReactorFacility input below and submit an equivalent Cycamore:Reactor xml configuration. (Add it as an xml file in the repository.)

Bright lite input:

 <facility>
    <name>Haddam Neck</name>
    <config>
      <ReactorFacility>
	<reactor_life>348</reactor_life>
        <in_commods>
            <val>UOX</val>       
        </in_commods>
        <out_commod>uf</out_commod>
        <libraries><val>lowLWR</val></libraries>
        <target_burnup>51</target_burnup>
        <nonleakage>0.89</nonleakage>
        <core_mass>96880</core_mass>
        <generated_power>1635.00</generated_power>
        <batches>3</batches>
        <flux_mode>0</flux_mode>        
        <tolerance>0.0001</tolerance>
        <SS_tolerance>0.0001</SS_tolerance>
	<reactor_type>BWR</reactor_type>
        <burnupcalc_timestep>60</burnupcalc_timestep>
	<burnup_time>
           <item><key>-696</key><val>16.4</val></item>
           <item><key>120</key><val>16.4</val></item>
           <item><key>180</key><val>17.03</val></item>
           <item><key>240</key><val>20.5</val></item>
           <item><key>300</key><val>23.44</val></item>
           <item><key>360</key><val>23.44</val></item>
           <item><key>420</key><val>30.3</val></item>
           <item><key>480</key><val>34.58</val></item>
           <item><key>540</key><val>38.76</val></item>
           <item><key>600</key><val>43.02</val></item>
           <item><key>660</key><val>44.72</val></item>
	</burnup_time>
	<avail_time>
           <item><key>-696</key><val>50</val></item>
           <item><key>120</key><val>50</val></item>
           <item><key>180</key><val>52</val></item>
           <item><key>240</key><val>54</val></item>
           <item><key>300</key><val>54</val></item>
           <item><key>360</key><val>63</val></item>
           <item><key>420</key><val>70</val></item>
           <item><key>480</key><val>78</val></item>
           <item><key>540</key><val>87</val></item>
           <item><key>600</key><val>91</val></item>
           <item><key>660</key><val>91</val></item>
	</avail_time>
        <CR_fissile>
            <val>922350</val>
            <val>942380</val>
            <val>942390</val>
            <val>942400</val>
            <val>942410</val>
            <val>942420</val>
            <val>952400</val>
            <val>952420</val>
            <val>952440</val>
        </CR_fissile>
      </ReactorFacility>
    </config>
  </facility>

• Understand the python script in https://github.com/arfc/fleetcomp that generates the input xml file.
• Then add a template file defining the Cycamore:Reactor inputs to help the python script create Cycamore:Reactor based simulations rather than Brightlite based simulations.

Write a report for Predicting-the-past in the US region.
Report should contain:

• A methodology with explanation of what functions were used and how they work
• A result section with results of the simulation
• A future works section that outlines future goals for this repository
• A reference section for all the sources used
• A section that explains xml inclusion

If possible, obtain the following data for CYCLUS simulation to improve accuracy:

• number of assemblies per core
• number of assemblies per batch

There are a handful of simple input files in Cycamore. Try to understand what each of them does. Summarize your findings as a comment in this issue.

The input files are here:
https://github.com/cyclus/cycamore/tree/develop/input

Reactor core mass and number of assemblies have been specified and updated in import_pris python file.
However, number of batches are still missing for some of the reactors listed below. These should be updated in the future and are assumed to be 3 for all reactors except CANDU.

• ESBWR
• GCR, HWGCR
• HTGR
• VVER
• VVER-1200

Pris database file lists reactors.
Webscrape lists power plants but not reactors.

Thus, a python script to match reactors to their powerplant should be written for the US region.

• Merge europe_test into master to reflect that #21 has been closed.
• Clean up (delete) deprecated branches (europe and europe_test)
• If necessary, open issues to indicate what work is remaining on that effort (see #21)

In the cyclus directory, run "./cyclus_unit_tests"

In the cycamore directory, run "./cycamore_unit_tests"

Does anything fail? Does everything pass? Report back.

If possible, obtain the following data for CYCLUS simulation to improve accuracy:

• Burnup for individual reactors

For a few of the Cycamore input files (in the cycamore input directory... these are present on your computer once you have cloned cycamore, but can also be found mirrored online here https://github.com/cyclus/cycamore/tree/develop/input), try running them.

• decide which execution flags to use when running cyclus (should you use -i to indicate the file? -o to name the output file?) . Report your decision in this issue.
• Execute cyclus accordingly with one of the cycamore example input files as the input to cyclus
• Does the simulation succeed? Does it fail?
• When it completes, there will be an output file (a database). Find it and see what you can learn about the simulation from the database. You may need an hdf5 viewer or an sqlite viewer installed.

@gyutaepark

When following the new README instructions to run a particular simulation, using the command...

python import_pris.py [region] [sim_start_year]

leads to the following error on my end...

riewski@riewski-VirtualBox:~/cyclus/predicting-the-past$ python import_pris.py US 1960
/home/riewski/cyclus/predicting-the-past/import_fleetcomp.py:4: VnVWarning: pyne.nucname is not yet V&V compliant.
from pyne import nucname as nn
Usage: python import_pris.py [region] [sim_start_yr]
Traceback (most recent call last):
File "import_pris.py", line 344, in
obtain_reactors(pris_file, sys.argv[1], reactors_tmpl)
File "import_pris.py", line 257, in obtain_reactors
in_data = idata.import_csv(in_csv, ',')
File "/home/riewski/cyclus/predicting-the-past/import_fleetcomp.py", line 25, in import_csv
with open(in_csv, encoding='utf-8') as source:
TypeError: 'encoding' is an invalid keyword argument for this function

I was wondering if you had any idea what the issue is, and could possibly figure out what is stopping the simulation from being run.

Eric

Using the reactor and recipe definitions you created in previous issues, create a fully working simulation input file to represent the past history of the united states reactor fleet.

The simulation has the following parameters:

• Start: 1965
• End: 2050
• Null Region (contains power institution and FC institution) call it US
• Power Institution:
  • Use the DeployInst to deploy your reactors on the right months/years
    • each reactor should be deployed on the appropriate simulation month (months since Jan. 1965)
    • each reactor will have a lifetime corresponding to the number of months it is in operation
  • Repository: Sink Facility (incommod='casked_spent_uox', start this 2040)
• FC Institution (1 facility each of the following):
• Mine : Source facility type that produces natural uranium (nat_u)
• Mill : Storage facility (incommod = nat_u, time='processing time', outcommod='u308')
• Conversion: Storage facility (incommod = u3o8, time='processing time', outcommod='uf6')
• Enrichment: Enrichment facility (incommod="uf6" product_commod="enr_u", tails_commod="tails", no initial feed, tails assay = 0.001, max inventory = inf 1e+299, max_enrich=0.045)
• Fuel Fabrication : Storage Facility (incommod="enr_u", time="processing time", outcommod='fresh_uox'
• Wet Storage: Storage Facility (incommod='spent_uox', time=5 years minimum, outcommod='cool_spent_uox')
• Dry Storage: Storage Facility (incommod='cool_spent_uox', time=0, outcommod='casked_spent_uox')

This issue can be closed when a complete and running xml file is submitted to this repository.

As identified in #45, the current workflow for producing coordinate merged reactors_pris with webscrape uses a modified version of the original reactors_pris_2016.csv.

This should be updated so that the script that merges the two database, should use reactors_pris_2016.original.csv.

Hi @eriewski , I saw you closed a bunch of these issues. That's great.
I'm not sure if you also installed Cycamore (https://github.com/cyclus/cycamore). If so, you can close this new issue. If not, then that's next up.

Go to : https://github.com/cyclus/cyclus/tree/v1.4.0-release#installing-cyclus-linux-unix-and-mac-osx and follow the instructions for linux.

The reactors_pris_2016.csv file available in arfc BOX does not provide all the information necessary for a successful input file creation for CYCLUS. I've recognized that Teddy has used certain values for the missing data but am not sure where some of the data is from.

Do you happen to know where I may be able to get the following information for the following reactor types?
Information

• number of assemblies
• assembly size (mass)

Reactor Type

• WCR
• GCR
• FBR
• LWGR

I was unable to find any from the picture of the book that Teddy has posted on arfc slack. Any help would be greatly appreciated.

Input and output fuel composition for reactors should be specified.
At this moment, all reactors use a LWR fuel with a burnup of 51 GWd/mthm

@katyhuff @jbae11 @eriewski @gonuke

Hi, this is Gyutae Park. I am the undergraduate student that is working under Professor Huff. I've heard from Teddy (Jin Whan Bae) that Professor Wilson was planning to make Eric work on Predicting-the-past repository.

I've made this issue to note the current status and future goals that this issue has so that Eric can successfully continue (if he desires) from where I have had to stop. I am currently not working on this repository.

At this moment, this repository has finished Predicting-the-past analysis for the US region. This is available in master branch. A report of what I've done, the assumptions made, and the analysis performed is all available in US.ipynb file.

Currently, the Europe simulation lacks specifications for GCRs, Candus, and other reactors. I've used PWR specifications instead. The results and the analysis done is available in the Europe_test branch.

Potential future goals have also been outlined in the Issues.

If you have any questions regarding the repository or the work that I have done, please let me know.

Thank you

The bug we looked at the other day, where there was an order of magnitude error in the power generation graph.... it should be fixed with a (single, atomic) pull request. Use a separate branch, please, so that only this change is present in the PR.

Go to https://www.continuum.io/downloads and follow the linux (or command line) instructions. This will likely have you download a file and execute a script. The download may be one of the commands it tells you to execute (wget, curl). So, don't stress if you don't click something to download it.