Optimal Power Flow with Convex Restriction

This repository provides implementation code for Convex Restriction of AC Optimal Power Flow problem. Please see references for more details.

Installation Requirements

The script was written in Julia v1.1 using the following packages.

using JuMP, PowerModels, Ipopt, MosekTools, Gurobi, SparseArrays, LinearAlgebra, Plots

Quick Start

## Import PowerFlowCVXRS functions
include("../src/main.jl")

## Read network data using PowerModels.jl
network_data = PowerModels.parse_file("../../cases/case9.m");

## Initiailize the network data by solving OPF problem
network_data=opf_initialization(network_data)

## Run Sequential Convex Restriction with 5 iterations
network_data, result_cvxr = scrs(network_data, 5);

You can try other systems using PGLib test cases

Citation

If you find this content useful for your research, please consider citing:

[1] Convex Restriction of Power Flow Feasibility Set

    @article{lee2019convex,
      author={Lee, Dongchan and Nguyen, Hung D. and Dvijotham, K. and Turitsyn, Konstantin},
      journal={IEEE Transactions on Control of Network Systems},
      title={Convex Restriction of Power Flow Feasibility Sets},
      year={2019}, volume={6}, number={3}, pages={1235-1245}
    }

[2] Feasible Path Identification in Optimal Power Flow with Sequential Convex Restriction

    @article{lee2020feasible,
      author={Lee, Dongchan and Turitsyn, Konstantin and Molzahn, Daniel Kenneth and Roald, Line},
      journal={IEEE Transactions on Power Systems}, 
      title={Feasible Path Identification in Optimal Power Flow With Sequential Convex Restriction}, 
      year={2020},  volume={35},  number={5},  pages={3648-3659}
    }    

[3] Robust Optimal Power Flow with Convex Restriction

    @article{lee2019feasible,
      title={Feasible Path Identification in Optimal Power Flow with Sequential Convex Restriction},
      author={Lee, Dongchan and Turitsyn, Konstantin and Molzahn, Daniel K and Roald, Line A},
      journal={arXiv preprint arXiv:1906.09483},
      year={2019}
    }