Real-time power system state estimation via deep unrolled neural networks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Contemporary smart power grids are being challenged by rapid voltage fluctuations, due to large-scale deployment of electric vehicles, demand response programs, and renewable generation. To secure grid operations, it becomes increasingly critical to obtain accurate estimates of power system AC states, namely the complex voltages at all buses, in real time. With the emergent nonconvexity however, past optimization based power system state estimation (PSSE) schemes are either sensitive to initialization, or computationally expensive. To bypass these hurdles, this paper advocates deep neural networks (DNNs) for PSSE. Different from model-agnostic NNs, that are difficult to tune and train, the novel model-specific DNN is obtained by unrolling state-of-the-art physics-based prox-linear PSSE solvers. The proposed prox-linear net requires a minimal tuning effort, and is easy to implement. Simulated tests show improved performance of the proposed prox-linear net relative to 'plain-vanilla' NN-, and the 'workhorse' Gauss-Newton-based PSSE solvers.

Original languageEnglish (US)
Title of host publication2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages907-911
Number of pages5
ISBN (Electronic)9781728112954
DOIs
StatePublished - Feb 20 2019
Event2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Anaheim, United States
Duration: Nov 26 2018Nov 29 2018

Publication series

Name2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

Conference

Conference2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018
CountryUnited States
CityAnaheim
Period11/26/1811/29/18

Fingerprint

State estimation
Smart power grids
Electric potential
Electric vehicles
Physics
Tuning
Deep neural networks

Keywords

  • Deep neural networks
  • Least-absolute-value
  • Power system state estimation
  • Prox-linear algorithm

Cite this

Zhang, L., Wang, G., & Giannakis, G. B. (2019). Real-time power system state estimation via deep unrolled neural networks. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings (pp. 907-911). [8646629] (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2018.8646629

Real-time power system state estimation via deep unrolled neural networks. / Zhang, Liang; Wang, Gang; Giannakis, Georgios B.

2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 907-911 8646629 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, L, Wang, G & Giannakis, GB 2019, Real-time power system state estimation via deep unrolled neural networks. in 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings., 8646629, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 907-911, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018, Anaheim, United States, 11/26/18. https://doi.org/10.1109/GlobalSIP.2018.8646629
Zhang L, Wang G, Giannakis GB. Real-time power system state estimation via deep unrolled neural networks. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 907-911. 8646629. (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). https://doi.org/10.1109/GlobalSIP.2018.8646629
Zhang, Liang ; Wang, Gang ; Giannakis, Georgios B. / Real-time power system state estimation via deep unrolled neural networks. 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 907-911 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).
@inproceedings{ed92684d240944d389c53dadec439765,
title = "Real-time power system state estimation via deep unrolled neural networks",
abstract = "Contemporary smart power grids are being challenged by rapid voltage fluctuations, due to large-scale deployment of electric vehicles, demand response programs, and renewable generation. To secure grid operations, it becomes increasingly critical to obtain accurate estimates of power system AC states, namely the complex voltages at all buses, in real time. With the emergent nonconvexity however, past optimization based power system state estimation (PSSE) schemes are either sensitive to initialization, or computationally expensive. To bypass these hurdles, this paper advocates deep neural networks (DNNs) for PSSE. Different from model-agnostic NNs, that are difficult to tune and train, the novel model-specific DNN is obtained by unrolling state-of-the-art physics-based prox-linear PSSE solvers. The proposed prox-linear net requires a minimal tuning effort, and is easy to implement. Simulated tests show improved performance of the proposed prox-linear net relative to 'plain-vanilla' NN-, and the 'workhorse' Gauss-Newton-based PSSE solvers.",
keywords = "Deep neural networks, Least-absolute-value, Power system state estimation, Prox-linear algorithm",
author = "Liang Zhang and Gang Wang and Giannakis, {Georgios B.}",
year = "2019",
month = "2",
day = "20",
doi = "10.1109/GlobalSIP.2018.8646629",
language = "English (US)",
series = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "907--911",
booktitle = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",

}

TY - GEN

T1 - Real-time power system state estimation via deep unrolled neural networks

AU - Zhang, Liang

AU - Wang, Gang

AU - Giannakis, Georgios B.

PY - 2019/2/20

Y1 - 2019/2/20

N2 - Contemporary smart power grids are being challenged by rapid voltage fluctuations, due to large-scale deployment of electric vehicles, demand response programs, and renewable generation. To secure grid operations, it becomes increasingly critical to obtain accurate estimates of power system AC states, namely the complex voltages at all buses, in real time. With the emergent nonconvexity however, past optimization based power system state estimation (PSSE) schemes are either sensitive to initialization, or computationally expensive. To bypass these hurdles, this paper advocates deep neural networks (DNNs) for PSSE. Different from model-agnostic NNs, that are difficult to tune and train, the novel model-specific DNN is obtained by unrolling state-of-the-art physics-based prox-linear PSSE solvers. The proposed prox-linear net requires a minimal tuning effort, and is easy to implement. Simulated tests show improved performance of the proposed prox-linear net relative to 'plain-vanilla' NN-, and the 'workhorse' Gauss-Newton-based PSSE solvers.

AB - Contemporary smart power grids are being challenged by rapid voltage fluctuations, due to large-scale deployment of electric vehicles, demand response programs, and renewable generation. To secure grid operations, it becomes increasingly critical to obtain accurate estimates of power system AC states, namely the complex voltages at all buses, in real time. With the emergent nonconvexity however, past optimization based power system state estimation (PSSE) schemes are either sensitive to initialization, or computationally expensive. To bypass these hurdles, this paper advocates deep neural networks (DNNs) for PSSE. Different from model-agnostic NNs, that are difficult to tune and train, the novel model-specific DNN is obtained by unrolling state-of-the-art physics-based prox-linear PSSE solvers. The proposed prox-linear net requires a minimal tuning effort, and is easy to implement. Simulated tests show improved performance of the proposed prox-linear net relative to 'plain-vanilla' NN-, and the 'workhorse' Gauss-Newton-based PSSE solvers.

KW - Deep neural networks

KW - Least-absolute-value

KW - Power system state estimation

KW - Prox-linear algorithm

UR - http://www.scopus.com/inward/record.url?scp=85061251608&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061251608&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2018.8646629

DO - 10.1109/GlobalSIP.2018.8646629

M3 - Conference contribution

AN - SCOPUS:85061251608

T3 - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

SP - 907

EP - 911

BT - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -