Distributed Energy Coordinator Application

This repository contains the distributed energy coordinator application for over-voltage mitigation in power distribution systems. The application solves a fairness-based energy coordination problem that enables customers to share the responsibility of voltage regulation in distribution systems. Since the focus of this application is on solving the over-voltage problem, it is effective only on the test feeder with high penetration of distributed energy resources (DERs) such as photovoltaics (PVs). For the demonstration purpose, a modified IEEE 123-bus test feeder is created by replacing the nominal loads with secondary configuration and populating the system with customer-owned PVs and utility-scale PVs to generate a hypothetical over-voltage scenario. The CIM XML file of the test case is available inside the inputs/feeder directory of this repository. The focus of this README is not on the application itself, but rather on how to configure/run it.

Prerequisites

The proposed application is developed in Python, and it requires the following packages for formulating and solving the optimization problem. Note that all these packages can be installed using pip

1. cvxpy
2. networkx
3. tabulate

Running the distributed energy coordinator application

1. From the command line execute the following commands to clone the repository

   user@user> git clone https://github.com/GRIDAPPSD/distributed-energy-coordinator
   user@user> cd distributed-energy-coordinator

2. Run the gridappsd docker. Use the develop tag to download the containers

   user@user> cd gridappsd-docker
   user@user> ./run.sh -t develop
   
   # You will now be inside the container, the following starts gridappsd
   
   gridappsd@f4ede7dacb7d:/gridappsd$ ./run-gridappsd.sh

3. Once the platform is up and running, start the application using the following commands. "feeder_mrid" is the model ID that is unique to each feeder.

   user@user> cd distributed-energy-coordinator
   user@user/distributed-energy-coordinator> cd dec
   user@user/distributed-energy-coordinator/dec> python3 run_dec.py "feeder_mrid"

   To simulate the over-voltage scenarios in the network, it is assumed that the PV generation is at its peak and nominal loads are scaled by a constant factor; this emulates an operating condition for a particular time of the day. In GridAPPS-D simulation, such operating conditions will be extracted from the simulation output using Simulation API. Once the application is invoked, it will start the optimization, and after the convergence, voltages and curtailment factors will be plotted. A CIM difference message will be created for the PV setpoint, and it will be dumped to a JSON file inside dec/outputs directory. Individual JSON files will be created for each agent that will contain the device information and its corresponding setpoints.

4. The CIM XML file of the modified IEEE 123-bus test feeder described above can be ingested into the database using the following commands.

   user@user> cd distributed-energy-coordinator
   user@user/distributed-energy-coordinator> cd inputs/feeder
   user@user> ./ingest_feeder.sh