A Learning-to-Infer Method for Real-Time Power Grid Multi-Line Outage Identification

Research output: Contribution to journalArticle

Abstract

Identifying a potentially large number of simultaneous line outages in power transmission networks in real time is a computationally hard problem. This is because the number of hypotheses grows exponentially with the network size. A new 'Learning-to-Infer' method is developed for efficient inference of every line status in the network. Optimizing the line outage detector is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount rapidly and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time multi-line outage identification. The proposed methods are evaluated in the IEEE 30, 118, and 300 bus systems. Excellent performance in identifying multi-line outages in real time is achieved with a reasonably small amount of data.

Original languageEnglish (US)
Article number8747534
Pages (from-to)555-564
Number of pages10
JournalIEEE Transactions on Smart Grid
Volume11
Issue number1
DOIs
StatePublished - Jan 2020

Fingerprint

Outages
Electric power transmission networks
Flow simulation
Power transmission
Classifiers
Detectors
Costs

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Keywords

  • Line outage detection
  • Monte Carlo method
  • machine learning
  • power system monitoring
  • variational inference

Cite this

@article{bac257e272f744bf8a04657754529d23,
title = "A Learning-to-Infer Method for Real-Time Power Grid Multi-Line Outage Identification",
abstract = "Identifying a potentially large number of simultaneous line outages in power transmission networks in real time is a computationally hard problem. This is because the number of hypotheses grows exponentially with the network size. A new 'Learning-to-Infer' method is developed for efficient inference of every line status in the network. Optimizing the line outage detector is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount rapidly and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time multi-line outage identification. The proposed methods are evaluated in the IEEE 30, 118, and 300 bus systems. Excellent performance in identifying multi-line outages in real time is achieved with a reasonably small amount of data.",
keywords = "Line outage detection, Monte Carlo method, machine learning, power system monitoring, variational inference",
author = "Poor, {H. Vincent}",
year = "2020",
month = "1",
doi = "https://doi.org/10.1109/TSG.2019.2925405",
language = "English (US)",
volume = "11",
pages = "555--564",
journal = "IEEE Transactions on Smart Grid",
issn = "1949-3053",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

A Learning-to-Infer Method for Real-Time Power Grid Multi-Line Outage Identification. / Poor, H. Vincent.

In: IEEE Transactions on Smart Grid, Vol. 11, No. 1, 8747534, 01.2020, p. 555-564.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Learning-to-Infer Method for Real-Time Power Grid Multi-Line Outage Identification

AU - Poor, H. Vincent

PY - 2020/1

Y1 - 2020/1

N2 - Identifying a potentially large number of simultaneous line outages in power transmission networks in real time is a computationally hard problem. This is because the number of hypotheses grows exponentially with the network size. A new 'Learning-to-Infer' method is developed for efficient inference of every line status in the network. Optimizing the line outage detector is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount rapidly and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time multi-line outage identification. The proposed methods are evaluated in the IEEE 30, 118, and 300 bus systems. Excellent performance in identifying multi-line outages in real time is achieved with a reasonably small amount of data.

AB - Identifying a potentially large number of simultaneous line outages in power transmission networks in real time is a computationally hard problem. This is because the number of hypotheses grows exponentially with the network size. A new 'Learning-to-Infer' method is developed for efficient inference of every line status in the network. Optimizing the line outage detector is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount rapidly and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time multi-line outage identification. The proposed methods are evaluated in the IEEE 30, 118, and 300 bus systems. Excellent performance in identifying multi-line outages in real time is achieved with a reasonably small amount of data.

KW - Line outage detection

KW - Monte Carlo method

KW - machine learning

KW - power system monitoring

KW - variational inference

UR - http://www.scopus.com/inward/record.url?scp=85068559016&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068559016&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/TSG.2019.2925405

DO - https://doi.org/10.1109/TSG.2019.2925405

M3 - Article

VL - 11

SP - 555

EP - 564

JO - IEEE Transactions on Smart Grid

JF - IEEE Transactions on Smart Grid

SN - 1949-3053

IS - 1

M1 - 8747534

ER -