How to migrate the geometric Network Electric data to Utility network?

1205
20
Jump to solution
02-02-2024 12:14 AM
Angiex__1
Occasional Contributor

I am having geometric network electric data, where i should have to migrate it to the Utility network, where i am stuck is that I don't know how to organize the Geometry network's feature classes to corresponding Asset Groups and Asset types in Utility network. As the words in the GN are different of that of the UN and i am finding what if i group them wrongly. Any suggestions or help that are recommended here?

0 Kudos
1 Solution

Accepted Solutions
RobertKrisher
Esri Regular Contributor

@Angiex__1 I recommend you tackle this problem in two different ways:

The first and most important step to solving this problem, and it's a common problem, is to find yourself a subject matter expert (SME). That is to say that you should find someone in your organization who understands all the terminology in that model and who is familiar with your system (both the GIS and the real-world) and sit down with them and have them help you pick the right names for everything. If you are friends with an engineer or someone who spends a lot of time in the field, they are often an excellent resource for these kinds of exercises. The online data dictionaries also have descriptions for most of their asset groups and asset types, but since every utility company uses their own terminology and slang for their equipment its often difficult to parse that out into some of the industry standard classifications found in that model.

Second, let the model guide you. There are rules defined for everything, so if you look at the rules and look at your data (and how you've classified things) and the vast majority of your features don't match, you may have picked the wrong asset group/type. When you're ready to test your mappings I recommend you do a quick prototype of them (using the approach outlined in this presentation from IMGIS). Just map your asset groups and asset types and see how many errors you get. If you get thousands of errors saying that there aren't any rules for the features you've mapped, then you should go back to your SME and see whether the problem is with your data, the mappings, or maybe you just have some weird configurations in your data that you need to account for with additional rules.

Finally, if you get stumped this community is a great resource to get answers. You'll find lots of examples in this forum where people have asked how others are modelling specific types of equipment (meters in particular), so if you get stuck and don't see your question already on this forum then start your own thread!

View solution in original post

20 Replies
RichardKoch
New Contributor III

Angie,

couple of suggestions. try not to think of it as a "right or wrong". There will be things you end up wishing you had done different but that is the nature of migrations and changes. An example of this from our migration would be the transformers. Ours are two terminal devices as they should be. We named the terminals as HIGH and LOW. I have situations in my model where we back feed the secondary blades on a standard conventional transformer.  I now wish we had named them as LOAD and SOURCE. It is a small thing but in some of our integrations a such and from a terminology standpoint I feel it would be cleaner. 

I would suggest becoming as familiar as possible with the UN model basics. @RobertKrisher from ESRI has a whole series of videos on these things that are targeted at initial migrations. 

Second, I would suggest thinking of it in the light of how you want to USE your data. Examples would be. When we trace, we want to locate item X, or when we query by attributes, we want to know the difference in Y and Z. 

If you want to set up a call to discuss some things message me here. 

Take Care

Richard Koch

0 Kudos
RobertKrisher
Esri Regular Contributor

@Angiex__1 I recommend you tackle this problem in two different ways:

The first and most important step to solving this problem, and it's a common problem, is to find yourself a subject matter expert (SME). That is to say that you should find someone in your organization who understands all the terminology in that model and who is familiar with your system (both the GIS and the real-world) and sit down with them and have them help you pick the right names for everything. If you are friends with an engineer or someone who spends a lot of time in the field, they are often an excellent resource for these kinds of exercises. The online data dictionaries also have descriptions for most of their asset groups and asset types, but since every utility company uses their own terminology and slang for their equipment its often difficult to parse that out into some of the industry standard classifications found in that model.

Second, let the model guide you. There are rules defined for everything, so if you look at the rules and look at your data (and how you've classified things) and the vast majority of your features don't match, you may have picked the wrong asset group/type. When you're ready to test your mappings I recommend you do a quick prototype of them (using the approach outlined in this presentation from IMGIS). Just map your asset groups and asset types and see how many errors you get. If you get thousands of errors saying that there aren't any rules for the features you've mapped, then you should go back to your SME and see whether the problem is with your data, the mappings, or maybe you just have some weird configurations in your data that you need to account for with additional rules.

Finally, if you get stumped this community is a great resource to get answers. You'll find lots of examples in this forum where people have asked how others are modelling specific types of equipment (meters in particular), so if you get stuck and don't see your question already on this forum then start your own thread!

Angiex__1
Occasional Contributor

Hi Robert, Thank you so much that you have taken time to give me a very brief answer regarding the migration of GN to UN, and I do have doubts regarding clearing the dirty areas, if you do have any suggestion that could make my work easier, can you suggest some?

0 Kudos
RobertKrisher
Esri Regular Contributor

Can you be more specific with what you mean by clearing the dirty areas. Are you concerned about your data mappings being wrong? Overall data quality? Something else?

In general:

  1. Create your data mappings
  2. Work with your SME to get them as close as you can (rename and add asset type if you need to)
  3. Review your dirty areas (topology errors) with your SME to determine the best way to resolve the errors.
  4. Resolve the errors through data cleanup, configuration, and automation
  5. Define subnetwork controls to accurately reflect your network sources/sinks so you can perform tracing and analysis

Steps 1&2 are covered in the data migration learning series I linked above, and steps 3&4 are covered in the connectivity and topology learning series, and step 5 is covered in the subnetworks learning series.

0 Kudos
Angiex__1
Occasional Contributor

@RobertKrisher  I have another doubt in subnetwork controller, is the water network necessarily have one subnetwork controller or multiples based on flow? in GN I have two source points and every time i run the trace in the Geometric network, the trace goes to either of the sources.

0 Kudos
RobertKrisher
Esri Regular Contributor

Subnetworks are allowed to have multiple controllers. When you trace the system analyzes all the controllers for the subnetwork to determine the path upstream, downstream, or to isolate an area.

0 Kudos
Angiex__1
Occasional Contributor

@RobertKrisher will it take a lot of time when Update subnetwork tool is running?

0 Kudos
RobertKrisher
Esri Regular Contributor

@Angiex__1 The amount of time it takes to update a subnetwork has more to do with the size of the subnetwork than the size of the subnetwork controllers. That is to say, a subnetwork with a million features will take about the same time to update regardless of whether it has 1 or 100 controllers. Having said that, subnetworks with many controllers are more likely to be larger, but let's not let correlation imply causation 🙂

0 Kudos
Angiex__1
Occasional Contributor

@RobertKrisher I have about 2,00,000 features and i have made only one subnetwork controller and it has been running for more than 15 hours already, should I be concerned?

0 Kudos