Convert a CIM File to OpenDSS and Excel

[StormsHalted]

Morning Everyone,

My objective is to convert a CIM file into OpenDSS and Excel. My ultimate goal being to navigate into the data sets holded by the CIM and work on them.

So I complete the CIMhub installation process and also I am able to run execute the python base123.py file correctly. I get the correct output script in command line.

Now, in my insignificant and frail efforts, I attempt to piece together the script to actually perform this task, by following the line by line code exegesis (line 69 - 72) given in the tutorial readme.

My script goes like:

`import cimhub.api as cimhub import cimhub.CIMHubConfig as CIMHubConfig import subprocess import stat import shutil import os import sys

if sys.platform == 'win32': shfile_export = 'go.bat' shfile_glm = './glm/checkglm.bat' shfile_run = 'checkglm.bat' cfg_json = 'cimhubjar.json' else: shfile_export = './go.sh' shfile_glm = './glm/checkglm.sh' shfile_run = './checkglm.sh' cfg_json = 'cimhubdocker.json'

cwd = os.getcwd()

cases = [ {'dssname':'ieee123', 'root':'ieee123', 'mRID':'CBE09B55-091B-4BB0-95DA-392237B12640', 'substation':'Fictitious', 'region':'Texas', 'subregion':'Austin', 'glmvsrc': 2401.78, 'bases':[4160.0], 'export_options':' -l=1.0 -p=1.0 -e=carson', 'check_branches':[{'dss_link': 'TRANSFORMER.REG4A', 'dss_bus': '160'}, {'dss_link': 'TRANSFORMER.REG4B', 'dss_bus': '160'}, {'dss_link': 'TRANSFORMER.REG4C', 'dss_bus': '160'}, {'gld_link': 'REG_REG4', 'gld_bus': '160'}, {'dss_link': 'LINE.L115', 'dss_bus': '149', 'gld_link': 'LINE_L115', 'gld_bus': '149'}]}, ]

cimhub.make_blazegraph_script (cases, './', 'dss/', 'glm/', shfile_export) st = os.stat (shfile_export) os.chmod (shfile_export, st.st_mode | stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH) p1 = subprocess.call (shfile_export, shell=True)`

Now I understand that, I have to modify the cases array to input a CIM XML. But someone needs to tell me exactly how to do it.

(I am also guessing, I will have to add lines 41 - 42 to activate blazegraph, as well)

Next I would attempt to convert to excel or learn Blazegraph to explore and work through my data.

[temcdrm]

Can you see if this helps?

https://cimhub.readthedocs.io/en/latest/Tutorial.html#starting-with-a-new-opendss-file

[StormsHalted]

@temcdrm

Afternoon Tom,

I am actually trying to achieve the opposite of it. I have a CIM file that I need to convert to OpenDSS.

This does gives me additional information in modifying the case statement and a few other parameters.

I still can't add a 'dssname' namespace to my case as I have to start with CIM and not DSS. If their is an already written example please point me to that.

[temcdrm]

• Load your CIM XML file into Blazegraph, through the Web browser
• Try executing some of the SPARQL from queries/queries.txt in the repository, starting with DistFeeder
• It may not return anything, because there are so many different usages of CIM versions and profiles. If that's the case, there are some general queries at the top of the queries/queries.txt file that will help you identify class names, attributes, and name spaces used in your CIM XML file.
• In the src_python and GMDM subdirectories, there are CIMadapter.py and some other scripts we used to translate incoming CIM XML files from the GMDM profile to our CIMHub profile. One option is to start from those examples, and adapt your incoming CIM XML to fit CIMHub's profile. This is what PNNL would do, if we wanted to maintain long-term support of the incoming CIM. It's the closest thing we have to a working example of what you wish to do, based on the interoperability tests from last month. The GMDM profile is close to CIMHub's, but not identical.
• A second option is to use your CIM XML as it is, but write your own SPARQL queries. It's easier to do those in the web browser first, then translate them into a copy of queries/q100.xml. Once you have working queries, make a copy of CPYDAR/ePHASORSIM.py, which can read and run SPARQL from a file like queries/q100.xml. Then you write your own Python code to produce the OpenDSS model. This is more of an ad hoc approach. PNNL implemented it mainly so that we can support other simulators with Python and SPARQL, without someone having to write Java code.

[temcdrm]

This is about all we can do. Suggestions have been added to the CIMHub tutorial.

[StormsHalted]

Hi @temcdrm,

Yes, I am able to load a CIM XML to blazegraph and then Run modified versions of some of the queries.

Now it boils down to me learning a completely new SPARQL syntax to modify those queries.

[AAndersn]

Hi @StormsHalted -- I know this thread is now more than a year old, but we now have this capability without the need for you to do any custom SPARQL as part of the new CIMantic Graphs library: https://github.com/PNNL-CIM-Tools/CIM-Graph/tree/develop and https://pypi.org/project/cim-graph/

from cimgraph.models import FeederModel
import cimgraph.utils as utils

feeder_mrid = "49AD8E07-3BF9-A4E2-CB8F-C3722F837B62"
feeder = cim.Feeder(mRID = feeder_mrid)
network = FeederModel(connection=database, container=feeder, distributed=False)
utils.get_all_data(network)
utils.write_csv(network, output_directory)

See https://github.com/PNNL-CIM-Tools/CIM-Graph/blob/develop/examples/feeder_example.ipynb and https://cimantic-graphs.readthedocs.io/en/latest/03_feeder_model.html for a more detailed example and how you can traverse the model