Graphical Models in Meshed Distribution Grids: Topology Estimation, Change Detection Limitations

Deepjyoti Deka; Saurav Talukdar; Michael Chertkov; Murti V. Salapaka

doi:10.1109/TSG.2020.2978541

Graphical Models in Meshed Distribution Grids: Topology Estimation, Change Detection Limitations

Deepjyoti Deka, Saurav Talukdar, Michael Chertkov, Murti V. Salapaka

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

36 Scopus citations

Abstract

Graphical models is a succinct way to represent the structure of a probability distributions. This article analyzes the graphical model of nodal voltages in non-radial power distribution grids. Using algebraic and structural properties of graphical models, algorithms exactly determining topology and detecting line changes for distribution grids are presented along with their theoretical limitations. We show that if distribution grids have cycles/loops of size greater than three, then nodal voltages are sufficient for efficient topology estimation without additional assumptions on system parameters. In contrast, line failure or change detection using nodal voltages does not require any structural assumption. Under noisy measurements, we provide the first non-trivial bounds on the maximum noise that the system can tolerate for asymptotically correct topology recovery. The performance of the designed algorithms is validated with non-linear AC power flow samples generated by Matpower on test grids, including scenarios with injection correlations and system noise.

Original language	English (US)
Article number	9025204
Pages (from-to)	4299-4310
Number of pages	12
Journal	IEEE Transactions on Smart Grid
Volume	11
Issue number	5
DOIs	https://doi.org/10.1109/TSG.2020.2978541
State	Published - Sep 2020

Bibliographical note

Funding Information:
Manuscript received May 15, 2019; revised October 23, 2019 and January 29, 2020; accepted February 26, 2020. Date of publication March 5, 2020; date of current version August 21, 2020. The work of Deepjyoti Deka and Michael Chertkov was supported in part by the Department of Energy through the Grid Modernization Lab Consortium, and in part by the Center for Non Linear Studies, Los Alamos National Laboratory. The work of Saurav Talukdar and Murti V. Salapaka was supported by ARPA-E under projects titled “A Robust Distributed Framework for Flexible Power Grids” under Grant DE-AR000071, and “Rapidly Viable Sustained Grid” under Grant DE-AR0001016. Paper no. TSG-00681-2019. (Corresponding author: Deepjyoti Deka.) Deepjyoti Deka is with Los Alamos National Laboratory, Los Alamos, NM 87545 USA (e-mail: deepjyoti@lanl.gov).

Publisher Copyright:
© 2010-2012 IEEE.

Keywords

Concentration matrix
conditional independence
distribution grids
graphical lasso
graphical models
line outage
measurement noise
power flows

Access

10.1109/TSG.2020.2978541

OpenUrl availability

Full text

Cite this

@article{2b405074dd174f2ea202c7b15be3e090,

title = "Graphical Models in Meshed Distribution Grids: Topology Estimation, Change Detection Limitations",

abstract = "Graphical models is a succinct way to represent the structure of a probability distributions. This article analyzes the graphical model of nodal voltages in non-radial power distribution grids. Using algebraic and structural properties of graphical models, algorithms exactly determining topology and detecting line changes for distribution grids are presented along with their theoretical limitations. We show that if distribution grids have cycles/loops of size greater than three, then nodal voltages are sufficient for efficient topology estimation without additional assumptions on system parameters. In contrast, line failure or change detection using nodal voltages does not require any structural assumption. Under noisy measurements, we provide the first non-trivial bounds on the maximum noise that the system can tolerate for asymptotically correct topology recovery. The performance of the designed algorithms is validated with non-linear AC power flow samples generated by Matpower on test grids, including scenarios with injection correlations and system noise.",

keywords = "Concentration matrix, conditional independence, distribution grids, graphical lasso, graphical models, line outage, measurement noise, power flows",

author = "Deepjyoti Deka and Saurav Talukdar and Michael Chertkov and Salapaka, {Murti V.}",

note = "Funding Information: Manuscript received May 15, 2019; revised October 23, 2019 and January 29, 2020; accepted February 26, 2020. Date of publication March 5, 2020; date of current version August 21, 2020. The work of Deepjyoti Deka and Michael Chertkov was supported in part by the Department of Energy through the Grid Modernization Lab Consortium, and in part by the Center for Non Linear Studies, Los Alamos National Laboratory. The work of Saurav Talukdar and Murti V. Salapaka was supported by ARPA-E under projects titled “A Robust Distributed Framework for Flexible Power Grids” under Grant DE-AR000071, and “Rapidly Viable Sustained Grid” under Grant DE-AR0001016. Paper no. TSG-00681-2019. (Corresponding author: Deepjyoti Deka.) Deepjyoti Deka is with Los Alamos National Laboratory, Los Alamos, NM 87545 USA (e-mail: deepjyoti@lanl.gov). Publisher Copyright: {\textcopyright} 2010-2012 IEEE.",

year = "2020",

month = sep,

doi = "10.1109/TSG.2020.2978541",

language = "English (US)",

volume = "11",

pages = "4299--4310",

journal = "IEEE Transactions on Smart Grid",

issn = "1949-3053",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "5",

}

TY - JOUR

T1 - Graphical Models in Meshed Distribution Grids

T2 - Topology Estimation, Change Detection Limitations

AU - Deka, Deepjyoti

AU - Talukdar, Saurav

AU - Chertkov, Michael

AU - Salapaka, Murti V.

N1 - Funding Information: Manuscript received May 15, 2019; revised October 23, 2019 and January 29, 2020; accepted February 26, 2020. Date of publication March 5, 2020; date of current version August 21, 2020. The work of Deepjyoti Deka and Michael Chertkov was supported in part by the Department of Energy through the Grid Modernization Lab Consortium, and in part by the Center for Non Linear Studies, Los Alamos National Laboratory. The work of Saurav Talukdar and Murti V. Salapaka was supported by ARPA-E under projects titled “A Robust Distributed Framework for Flexible Power Grids” under Grant DE-AR000071, and “Rapidly Viable Sustained Grid” under Grant DE-AR0001016. Paper no. TSG-00681-2019. (Corresponding author: Deepjyoti Deka.) Deepjyoti Deka is with Los Alamos National Laboratory, Los Alamos, NM 87545 USA (e-mail: deepjyoti@lanl.gov). Publisher Copyright: © 2010-2012 IEEE.

PY - 2020/9

Y1 - 2020/9

N2 - Graphical models is a succinct way to represent the structure of a probability distributions. This article analyzes the graphical model of nodal voltages in non-radial power distribution grids. Using algebraic and structural properties of graphical models, algorithms exactly determining topology and detecting line changes for distribution grids are presented along with their theoretical limitations. We show that if distribution grids have cycles/loops of size greater than three, then nodal voltages are sufficient for efficient topology estimation without additional assumptions on system parameters. In contrast, line failure or change detection using nodal voltages does not require any structural assumption. Under noisy measurements, we provide the first non-trivial bounds on the maximum noise that the system can tolerate for asymptotically correct topology recovery. The performance of the designed algorithms is validated with non-linear AC power flow samples generated by Matpower on test grids, including scenarios with injection correlations and system noise.

AB - Graphical models is a succinct way to represent the structure of a probability distributions. This article analyzes the graphical model of nodal voltages in non-radial power distribution grids. Using algebraic and structural properties of graphical models, algorithms exactly determining topology and detecting line changes for distribution grids are presented along with their theoretical limitations. We show that if distribution grids have cycles/loops of size greater than three, then nodal voltages are sufficient for efficient topology estimation without additional assumptions on system parameters. In contrast, line failure or change detection using nodal voltages does not require any structural assumption. Under noisy measurements, we provide the first non-trivial bounds on the maximum noise that the system can tolerate for asymptotically correct topology recovery. The performance of the designed algorithms is validated with non-linear AC power flow samples generated by Matpower on test grids, including scenarios with injection correlations and system noise.

KW - Concentration matrix

KW - conditional independence

KW - distribution grids

KW - graphical lasso

KW - graphical models

KW - line outage

KW - measurement noise

KW - power flows

UR - http://www.scopus.com/inward/record.url?scp=85090161405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090161405&partnerID=8YFLogxK

U2 - 10.1109/TSG.2020.2978541

DO - 10.1109/TSG.2020.2978541

M3 - Article

AN - SCOPUS:85090161405

SN - 1949-3053

VL - 11

SP - 4299

EP - 4310

JO - IEEE Transactions on Smart Grid

JF - IEEE Transactions on Smart Grid

IS - 5

M1 - 9025204

ER -

Graphical Models in Meshed Distribution Grids: Topology Estimation, Change Detection Limitations

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this