Quick Start Guide

Once PowerModels is installed, Ipopt is installed, and a network data file (e.g. "case3.m" or "case3.raw") has been acquired, an AC Optimal Power Flow can be executed with,

using PowerModels
using Ipopt

solve_ac_opf("matpower/case3.m", Ipopt.Optimizer)

Similarly, a DC Optimal Power Flow can be executed with

solve_dc_opf("matpower/case3.m", Ipopt.Optimizer)

PTI .raw files in the PSS(R)E v33 specification can be run similarly, e.g. in the case of an AC Optimal Power Flow

solve_ac_opf("case3.raw", Ipopt.Optimizer)

Getting Results

The run commands in PowerModels return detailed results data in the form of a dictionary. Results dictionaries from either Matpower .m or PTI .raw files will be identical in format. This dictionary can be saved for further processing as follows,

result = solve_ac_opf("matpower/case3.m", Ipopt.Optimizer)

For example, the algorithm's runtime and final objective value can be accessed with,

result["solve_time"]
result["objective"]

The "solution" field contains detailed information about the solution produced by the run method. For example, the following dictionary comprehension can be used to inspect the bus voltage angles in the solution,

Dict(name => data["va"] for (name, data) in result["solution"]["bus"])

The print_summary(result["solution"]) function can be used show an table-like overview of the solution data. For more information about PowerModels result data see the PowerModels Result Data Format section.

Accessing Different Formulations

The function solve_ac_opf and solve_dc_opf are shorthands for a more general formulation-independent OPF execution, solve_opf. For example, solve_ac_opf is equivalent to,

solve_opf("matpower/case3.m", ACPPowerModel, Ipopt.Optimizer)

where "ACPPowerModel" indicates an AC formulation in polar coordinates. This more generic solve_opf() allows one to solve an OPF problem with any power network formulation implemented in PowerModels. For example, an SOC Optimal Power Flow can be run with,

solve_opf("matpower/case3.m", SOCWRPowerModel, Ipopt.Optimizer)

Formulation Details provides a list of available formulations.

Modifying Network Data

The following example demonstrates one way to perform multiple PowerModels solves while modifing the network data in Julia,

network_data = PowerModels.parse_file("matpower/case3.m")

solve_opf(network_data, ACPPowerModel, Ipopt.Optimizer)

network_data["load"]["3"]["pd"] = 0.0
network_data["load"]["3"]["qd"] = 0.0

solve_opf(network_data, ACPPowerModel, Ipopt.Optimizer)

Network data parsed from PTI .raw files supports data extensions, i.e. data fields that are within the PSS(R)E specification, but not used by PowerModels for calculation. This can be achieved by

network_data = PowerModels.parse_file("pti/case3.raw"; import_all=true)

This network data can be modified in the same way as the previous Matpower .m file example. For additional details about the network data, see the PowerModels Network Data Format section.

Inspecting AC and DC branch flow results

The flow AC and DC branch results are written to the result by default. The following can be used to inspect the flow results:

result = solve_opf("matpower/case3_dc.m", ACPPowerModel, Ipopt.Optimizer)
result["solution"]["dcline"]["1"]
result["solution"]["branch"]["2"]

The losses of an AC or DC branch can be derived:

loss_ac =  Dict(name => data["pt"]+data["pf"] for (name, data) in result["solution"]["branch"])
loss_dc =  Dict(name => data["pt"]+data["pf"] for (name, data) in result["solution"]["dcline"])

Building PowerModels from Network Data Dictionaries

The following example demonstrates how to break a solve_opf call into separate model building and solving steps. This allows inspection of the JuMP model created by PowerModels for the AC-OPF problem,

pm = instantiate_model("matpower/case3.m", ACPPowerModel, PowerModels.build_opf)

print(pm.model)

result = optimize_model!(pm, optimizer=Ipopt.Optimizer)

Alternatively, you can further break it up by parsing a file into a network data dictionary, before passing it on to instantiate_model() like so,

network_data = PowerModels.parse_file("matpower/case3.m")

pm = instantiate_model(network_data, ACPPowerModel, PowerModels.build_opf)

print(pm.model)

result = optimize_model!(pm, optimizer=Ipopt.Optimizer)