Differentially Private Distributed Optimal Power Flow

Vladimir Dvorkin; Pascal Van Hentenryck; Jalal Kazempour; Pierre Pinson

doi:10.1109/CDC42340.2020.9303768

Differentially Private Distributed Optimal Power Flow

Vladimir Dvorkin, Pascal Van Hentenryck, Jalal Kazempour, Pierre Pinson

Georgia Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

166 Downloads (Orbit)

Abstract

Distributed algorithms enable private Optimal Power Flow (OPF) computations by avoiding the need in sharing sensitive information localized in algorithms sub-problems. However, adversaries can still infer this information from the coordination signals exchanged across iterations. This paper seeks formal privacy guarantees for distributed OPF computations and provides differentially private algorithms for OPF computations based on the consensus Alternating Direction Method of Multipliers (ADMM). The proposed algorithms attain differential privacy by introducing static and dynamic random perturbations of OPF sub-problem solutions at each iteration. These perturbations are Laplacian and designed to prevent the inference of sensitive information, as well as to provide theoretical privacy guarantees for ADMM subproblems. Using a standard IEEE 118-node test case, the paper explores the fundamental trade-offs among privacy, algorithmic.
convergence, and optimality losses.

Original language	English
Title of host publication	Proceedings of the 59^th IEEE Conference on Decision and Control
Publisher	IEEE
Publication date	2021
Pages	2092-2097
ISBN (Print)	9781728174471
DOIs	https://doi.org/10.1109/CDC42340.2020.9303768
Publication status	Published - 2021
Event	59^th IEEE Conference on Decision and Control - International Convention Center Jeju, Jeju Island, Korea, Republic of Duration: 14 Dec 2020 → 18 Dec 2020 https://cdc2020.ieeecss.org/

Conference

Conference	59^th IEEE Conference on Decision and Control
Location	International Convention Center Jeju
Country/Territory	Korea, Republic of
City	Jeju Island
Period	14/12/2020 → 18/12/2020
Internet address	https://cdc2020.ieeecss.org/

Access to Document

10.1109/CDC42340.2020.9303768

FulltextAccepted author manuscript, 195 KB

OpenUrl availability

Full text

Cite this

@inproceedings{ffdaf301fdf64c3e9f8303abee97f16b,

title = "Differentially Private Distributed Optimal Power Flow",

abstract = "Distributed algorithms enable private Optimal Power Flow (OPF) computations by avoiding the need in sharing sensitive information localized in algorithms sub-problems. However, adversaries can still infer this information from the coordination signals exchanged across iterations. This paper seeks formal privacy guarantees for distributed OPF computations and provides differentially private algorithms for OPF computations based on the consensus Alternating Direction Method of Multipliers (ADMM). The proposed algorithms attain differential privacy by introducing static and dynamic random perturbations of OPF sub-problem solutions at each iteration. These perturbations are Laplacian and designed to prevent the inference of sensitive information, as well as to provide theoretical privacy guarantees for ADMM subproblems. Using a standard IEEE 118-node test case, the paper explores the fundamental trade-offs among privacy, algorithmic. convergence, and optimality losses.",

author = "Vladimir Dvorkin and \{Van Hentenryck\}, Pascal and Jalal Kazempour and Pierre Pinson",

year = "2021",

doi = "10.1109/CDC42340.2020.9303768",

language = "English",

isbn = "9781728174471",

pages = "2092--2097",

booktitle = "Proceedings of the 59th IEEE Conference on Decision and Control",

publisher = "IEEE",

address = "United States",

note = "59<sup>th</sup> IEEE Conference on Decision and Control, CDC 2020 ; Conference date: 14-12-2020 Through 18-12-2020",

url = "https://cdc2020.ieeecss.org/",

}

TY - GEN

T1 - Differentially Private Distributed Optimal Power Flow

AU - Dvorkin, Vladimir

AU - Van Hentenryck, Pascal

AU - Kazempour, Jalal

AU - Pinson, Pierre

PY - 2021

Y1 - 2021

N2 - Distributed algorithms enable private Optimal Power Flow (OPF) computations by avoiding the need in sharing sensitive information localized in algorithms sub-problems. However, adversaries can still infer this information from the coordination signals exchanged across iterations. This paper seeks formal privacy guarantees for distributed OPF computations and provides differentially private algorithms for OPF computations based on the consensus Alternating Direction Method of Multipliers (ADMM). The proposed algorithms attain differential privacy by introducing static and dynamic random perturbations of OPF sub-problem solutions at each iteration. These perturbations are Laplacian and designed to prevent the inference of sensitive information, as well as to provide theoretical privacy guarantees for ADMM subproblems. Using a standard IEEE 118-node test case, the paper explores the fundamental trade-offs among privacy, algorithmic. convergence, and optimality losses.

AB - Distributed algorithms enable private Optimal Power Flow (OPF) computations by avoiding the need in sharing sensitive information localized in algorithms sub-problems. However, adversaries can still infer this information from the coordination signals exchanged across iterations. This paper seeks formal privacy guarantees for distributed OPF computations and provides differentially private algorithms for OPF computations based on the consensus Alternating Direction Method of Multipliers (ADMM). The proposed algorithms attain differential privacy by introducing static and dynamic random perturbations of OPF sub-problem solutions at each iteration. These perturbations are Laplacian and designed to prevent the inference of sensitive information, as well as to provide theoretical privacy guarantees for ADMM subproblems. Using a standard IEEE 118-node test case, the paper explores the fundamental trade-offs among privacy, algorithmic. convergence, and optimality losses.

U2 - 10.1109/CDC42340.2020.9303768

DO - 10.1109/CDC42340.2020.9303768

M3 - Article in proceedings

SN - 9781728174471

SP - 2092

EP - 2097

BT - Proceedings of the 59th IEEE Conference on Decision and Control

PB - IEEE

T2 - 59<sup>th</sup> IEEE Conference on Decision and Control

Y2 - 14 December 2020 through 18 December 2020

ER -

Differentially Private Distributed Optimal Power Flow

Abstract

Conference

Access to Document

OpenUrl availability

Fingerprint

Cite this