PV Deployments

This section shows how to generate DISCO PV deployments from raw opendss models.

SourceTree1 PV Deployments

The main command going to be used is the one below,

$ disco pv-deployments source-tree-1 --action <action-name> --hierarchy <hierarchy> --placement <placement type> INPUT_PATH

There are several actions here related to PV deployments manipulation, including

  • transform-loads: Transform Loads.dss file before conducting PV deployments.

  • redirect-pvshapes: Redirect PVShape.dss in both substation and feeder Master.dss files.

  • generate-jobs: Help generate create-pv and create-configs jobs in JSON, i.e., jade config.

  • restore-feeders: Before and during PV deployments, Loads.dss and Master.dss files were modified, need to restore after that.

  • create-pv: create PV deployments on feeders based on placement, sample and penetration levels.

  • check-pv: check if there are PV deployments missing at each placement, sample and penetration level.

  • remove-pv: delete PV deployments in case there’s something wrong.

  • create-configs: create PV config files at each sample deployment level.

  • check-configs: check if there are PV config files missing in deployment directories.

  • remove-configs: remove PV config files in case there’s something wrong.

  • list-feeders: list feeder paths given input of region, substation or feeder.

Transform Loads

Loads.dss files need to be transformed before generating PV deployments.

  • Loads may need to be changed to use suitable center-tap schema.

  • Temporarily disconnect load shapes. The generation process compiles the OpenDSS circuit. If load shapes are present, OpenDSS has to load the data and that can take significant amounts of time. Load shapes are not needed by the generation process, and so it disables them in the OpenDSS text files. A later step re-enables them.

The command to run this is,

$ disco pv-deployments source-tree-1 -a transform-loads -h <hierarchy> INPUT_PATH

Redirect PVShapes

This workflow will generate OpenDSS files with varying counts and sizes of PVSystems. It will assign load shape profiles to those PVSystems from a pool of profiles. You must define these profiles in a PVShapes.dss file and copy that files to all substation and/or feeder directories.

All Master.dss need to redirect to PVShapes.dss. We recommend that you add these lines to your files. If you do that, you can skip to the next section.

If your directory structure aligns with the source-tree-1 expectations, the disco CLI command below will add the redirects automatically.

Todo

Make this code handle all cases generically.

Run this command:

$ disco pv-deployments source-tree-1 -a redirect-pvshapes -h <hierarchy> INPUT_PATH

Generate Jobs

DISCO provides a command to help generate JADE jobs config files for PV deployments and PV configs, that is,

$ disco pv-deployments source-tree-1 -a generate-jobs -h <hierarchy> INPUT_PATH

The hierarchy options are:

  • city

  • region

  • substation

  • feeder

We recommand to run this generate-jobs command with --hierarchy=city and generate jobs on all feeders within the city path, if your simulation/analysis jobs run relatively stable, this way can help avoid the repeated job generation work on regions, substations, or feeders. For test or debug purpose,it’s good to specify --hierarchy=feeder for generating config file with one job, or --hierarchy=substation with a few jobs.

This command will generate two JADE config files:

  • create-pv-jobs.json contains jobs for PV deployments.

  • create-config-jobs.json contains jobs for PV configs

And, you can submit the jobs via jade submit-jobs <config_file> command.

Warning

Since PV configs are based on the result of PV deployments, so you will need to wait PV deployment jobs to complete, before to submit PV config jobs.

PV Deployments

Submit Jobs

To generate PV deployments, you will need to submit the jobs via JADE, that is,

$ jade submit-jobs <OPTIONS> create-pv-jobs.json

If the jobs pass, then the PV deployments task is done. If you’d like to explore details about create-pv action based on your hierarchy and according input path, please check the section below.

Details Exploration

Here are some example commands showing how to create, check and remove PV deployments.

  1. Create PV deployments on feeder1 with --placement random.

$ disco pv-deployments source-tree-1 -a create-pv -h feeder -p random --pv-upscale <feeder1_path>

  1. Create PV deployments on substation1 with a few feeders.

$ disco pv-deployments source-tree-1 -a create-pv -h substation -p random --pv-upscale <substation1_path>

  1. Create PV deployments on region1 with many feeders in parallel by using JADE.

As each region has a large number of feeders, it is recommended to use JADE to parallize the jobs.

$ disco pv-deployments source-tree-1 -a list-feeders -h region <region1_path>
# Create a <commands.txt> file which contains create-pv commands on feeders as above.
$ jade config create <commands.txt> -c config1.json
$ jade submit-jobs config1.json

  1. If you like to check which PV deployments are missing due to job failures,

$ disco pv-deployments source-tree-1 -a check-pv -h feeder -p random <feeder1_path>
$ disco pv-deployments source-tree-1 -a check-pv -h substation  -p random <substation1_path>
$ disco pv-deployments source-tree-1 -a check-pv -h region  -p random <region1_path>

It returns the missing samples and penetrations on each feeder. If don’t have --placement specified, the result would include placement missing information on each feeder.

  1. If you found some issues with the PV deployments, and like to delete them, here are example commands,

$ disco pv-deployments source-tree-1 -a remove-pv -h feeder  -p random <feeder1_path>
$ disco pv-deployments source-tree-1 -a remove-pv -h substation  -p random <substation1_path>
$ disco pv-deployments source-tree-1 -a remove-pv -h region  -p random <region1_path>

PV Configs

Submit Jobs

To generate PV configs, you will need to submit the jobs via JADE, that is,

$ jade submit-jobs <OPTIONS> create-config-jobs.json

If the jobs pass, then the PV configs task is done. If you’d like to explore details about create-configs action based on your hierarchy and according input path, please check the section below.

Details Exploration

After creating PV deployments we need to generate PV config files that define a load shape profile for each PV system. The config files get created in sample directories. The examples below show commands for creating, checking or removing PV config files.

  1. Create PV configs on feeder1 based on PV deployments data.

$ disco pv-deployments source-tree-1 -a create-configs -h feeder <feeder1_path>

  1. Create PV configs on substation1 with a few feeders.

$ disco pv-deployments source-tree-1 -a create-configs -h substation <substation1_path>

Warning

The option -p or --placement does not apply to create-configs action, as after all pv configs created in each feeder, a sum group file based on customer types would be created based on the pv configs of the feeder.

  1. Create PV configs on region1 with many feeders in parallel by using JADE.

$ disco pv-deployments source-tree-1 -a list-feeders -h region <region1_path>
# Create a <commands.txt> file which contains create-configs commands on feeders as above.
$ jade config create <commands.txt> -c config2.json
$ jade submit-jobs config2.json

  1. Check if any feeder is missing PV config files.

$ disco pv-deployments source-tree-1 -a check-configs -h feeder -p random <feeder1_path>
$ disco pv-deployments source-tree-1 -a check-configs -h substation -p random <substation1_path>
$ disco pv-deployments source-tree-1 -a check-configs -h region -p random <region1_path>

  1. Remove PV configs if something is wrong.

$ disco pv-deployments source-tree-1 -a remove-configs -h feeder -p random <feeder1_path>
$ disco pv-deployments source-tree-1 -a remove-configs -h substation -p random <substation1_path>
$ disco pv-deployments source-tree-1 -a remove-configs -h region -p random <region1_path>

Restore Feeders

As the Loads.dss in SourceTree1 models needs to be transformed during PV deployments, and the content of Loads.dss was modified. However, we backed up the original Loads.dss before PV deployments, so we can rename back after that. Simply, the steps look like this.

One more thing, to speed up PV deployments, we commented out LoadShapes.dss before PV deployments in master files, we need to revert it back after PV deployments.

  1. Before PV deployments:

  • Rename raw Loads.dss into Original_Loads.dss.

  1. During PV deployments:

  • DISCO PV deployment program transformed Loads.dss in place.

  • and, stripped yearly=<pv-profile> from the load lines.

  1. After PV deployments:

  • Rename transformed Loads.dss file into PV_Loads.dss.

  • Rename Original_Loads.dss back to Loads.dss.

Run the command below to rename Loads.dss file and related,

$ disco pv-deployments source-tree-1 -a restore-feeders -h <hierarchy> INPUT_PATH