GitHub

Quick Start Guide

Running Integrated Transmission-Distribution Optimal Power Flow (OPFITD)

Once PowerModelsITD.jl is installed, Ipopt is installed, and network data files for the transmission system, distribution system(s), and boundary linking (e.g., "case5_withload.m", "case3_unbalanced.dss, "case5_case3_unbal.json" in the package folder under ./test/data) have been acquired, an Integrated Transmission-Distribution (ITD) AC Optimal Power Flow can be executed as follows,

using PowerModelsITD
using Ipopt

pmitd_type = NLPowerModelITD{ACPPowerModel, ACPUPowerModel} # define the formulation type.

result = solve_opfitd("case5_withload.m", "case3_unbalanced.dss", "case5_case3_unbal.json", pmitd_type, Ipopt.Optimizer)

Running Integrated Transmission-Distribution Optimal Power Flow (OPFITD) with Multiple Distribution Systems

The snippet shown below demonstrates how to run an OPFITD with multiple (different) distribution systems connected to different transmission system buses. In summary, the distribution system files need to be passed as a Vector of files, and the corresponding boundaries must be defined in the boundary linking file. For the following snippet, assume we have access to ./test/data to get the respective files.

using PowerModelsITD
using Ipopt

# Files
pm_file = "case5_with2loads.m"
pmd_file1 = "case3_unbalanced.dss"
pmd_file2 = "case3_balanced.dss"
boundary_file = "case5_case3x2_unbal_bal.json"

pmd_files = [pmd_file1, pmd_file2] # create a vector of distribution system files.

pmitd_type = NLPowerModelITD{ACPPowerModel, ACPUPowerModel} # define the formulation type.

result = solve_opfitd(pm_file, pmd_files, boundary_file, pmitd_type, Ipopt.Optimizer)

Parsing files

To parse the respective files into PowerModelsITD.jl, use the parse_files command

pmitd_data = parse_files(pm_file, pmd_file, boundary_file) # single distribution system file.

To parse for a model with multiple distribution system files, just pass the distribution system argument as a vector of files, as seen below.

pmd_files = [pmd_file1, pmd_file2]
pmitd_data = parse_files(pm_file, pmd_files, boundary_file) # vector of multiple distribution system files.

Getting Results

The solve_ commands in PowerModelsITD.jl return detailed results data in the form of a dictionary. This dictionary can be saved for further processing as follows,

result = solve_opfitd(pmitd_data, NLPowerModelITD{ACPPowerModel, ACPUPowerModel}, Ipopt.Optimizer)

Note: the function solve_opfitd(...) does not neccessarily needs for the data to be parsed (i.e., parse_files(...)). The user can pass directly the files of the problem to the solve_opfitd(...) function, as seen previously.

Alternatively, you can use the function solve_model(...) and specify the build method, in this case build_opfitd:

result = solve_model(pmitd_data, NLPowerModelITD{ACPPowerModel, ACPUPowerModel}, Ipopt.Optimizer, build_opfitd)

Running Multinetwork (mn) Integrated Transmission-Distribution Optimal Power Flow (OPFITD) with Multiple Distribution Systems (ms)

The snippet shown below demonstrates how to run a multinetwork (mn) OPFITD with multiple distribution systems (ms) connected to different transmission system buses. As seen previously, the distribution system files need to be passed as a Vector of files, and the corresponding boundaries are defined in the boundary linking file. To run a multinetwork problem, time series values can be defined inside the opendss file(s) that represent the distribution system(s). Some important things to remember are:

  • The time series values can be defined in a .csv file that is called inside the opendss file(s).

  • When the data is parsed, the transmission system will be automatically replicated to the number of time series values available in the opendss file(s).

  • If data in the transmission system need to be modified for different time steps, the data can be modified after the parsing process is done.

  • Remember to run the special multinetwork function or build function (i.e., solve_mn_opfitd(...) or solve_model(..., build_mn_opfitd)).

For the following snippet, assume we have access to ./test/data to get the respective files.

using PowerModelsITD
using Ipopt

# Files
pm_file = "case5_with2loads.m"
pmd_file = "case3_unbalanced_withoutgen_mn.dss"
boundary_file = "case5_case3x2.json"

pmd_files = [pmd_file, pmd_file] # create a vector of distribution system files.

pmitd_type = NLPowerModelITD{ACPPowerModel, ACPUPowerModel} # define the formulation type.

result = solve_mn_opfitd(pm_file, pmd_files, boundary_file, pmitd_type, Ipopt.Optimizer; auto_rename=true)

Important Note: If you examine the file case3_unbalanced_withoutgen_mn.dss in detail, you will notice how time series power values are defined inside the opendss file as New Loadshape.ls1 pmult=(file=load_profile.csv), and then the load models are defined with parameters model=1 daily=ls1. This definition is sufficient for PowerModelsITD.jl to understand that this is a multinetwork problem. If you examine the content of the load_profile.csv file, you will notice that there are 4 time steps. This will cause the transmission, distribution, and boundary data to be replicated x4 (i.e., 4 times) while it is parsed to represent the overall problem.

Note: The auto_rename=true option used in this example will be explained in later documentation.

Accessing Different Formulations

There is a diverse number of formulations that can be used to solve the OPFITD, PFITD, and other problem specifications. These can be found in types.jl. A non-exhaustive list of the supported formulations is presented below.

Examples

More examples of working with the engineering data model can be found in the /examples folder of the PowerModelsITD.jl repository. These are Pluto Notebooks; instructions for running them can be found in the Pluto documentation.