Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid

Jingyi Zhang, Chao Lu, Jennie Si, Jie Song, Yinsheng Su

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

2 Citations (Scopus)

Abstract

We propose a novel load shedding (LS) scheme against voltage instability using deep reinforcement learning (DRL). Both spatial and temporal information of a large power grid are used in the DRL control scheme. Specifically, both dynamic state variables and grid topology information are inputs to the learning controller. Within the DRL scheme, a deep learning neural network is designed and implemented to automatically extract translation-invariance information about voltage instability. The DRL load shedding controller interacts with system dynamics through a sequence of observations, actions and rewards to determine load shedding amounts in a manner that maximizes cumulative future reward, accomplishing coordination within the region rapidly to meet the online application requirements. The DRL based distributed LS scheme is performed to control the China Southern Power Grid (CSG) system. Our results show improved voltage recovery performance by load shedding using the proposed scheme under different unknown test scenarios.

Original languageEnglish (US)
Title of host publication2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Volume2018-July
ISBN (Electronic)9781509060146
DOIs
StatePublished - Oct 10 2018
Event2018 International Joint Conference on Neural Networks, IJCNN 2018 - Rio de Janeiro, Brazil
Duration: Jul 8 2018Jul 13 2018

Other

Other2018 International Joint Conference on Neural Networks, IJCNN 2018
CountryBrazil
CityRio de Janeiro
Period7/8/187/13/18

Fingerprint

Reinforcement learning
Dynamic loads
Voltage control
Reinforcement
Electric potential
Controllers
Invariance
Dynamical systems
Topology
Neural networks
Recovery

Keywords

  • deep reinforcement learning
  • spatial-temporal information fusion
  • Under voltage load shedding

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Zhang, J., Lu, C., Si, J., Song, J., & Su, Y. (2018). Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid. In 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings (Vol. 2018-July). [8489041] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2018.8489041

Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid. / Zhang, Jingyi; Lu, Chao; Si, Jennie; Song, Jie; Su, Yinsheng.

2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Vol. 2018-July Institute of Electrical and Electronics Engineers Inc., 2018. 8489041.

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

Zhang, J, Lu, C, Si, J, Song, J & Su, Y 2018, Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid. in 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. vol. 2018-July, 8489041, Institute of Electrical and Electronics Engineers Inc., 2018 International Joint Conference on Neural Networks, IJCNN 2018, Rio de Janeiro, Brazil, 7/8/18. https://doi.org/10.1109/IJCNN.2018.8489041
Zhang J, Lu C, Si J, Song J, Su Y. Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid. In 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Vol. 2018-July. Institute of Electrical and Electronics Engineers Inc. 2018. 8489041 https://doi.org/10.1109/IJCNN.2018.8489041
Zhang, Jingyi ; Lu, Chao ; Si, Jennie ; Song, Jie ; Su, Yinsheng. / Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid. 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Vol. 2018-July Institute of Electrical and Electronics Engineers Inc., 2018.
@inproceedings{43a31fc08a8f4e20960becfa26a8f714,
title = "Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid",
abstract = "We propose a novel load shedding (LS) scheme against voltage instability using deep reinforcement learning (DRL). Both spatial and temporal information of a large power grid are used in the DRL control scheme. Specifically, both dynamic state variables and grid topology information are inputs to the learning controller. Within the DRL scheme, a deep learning neural network is designed and implemented to automatically extract translation-invariance information about voltage instability. The DRL load shedding controller interacts with system dynamics through a sequence of observations, actions and rewards to determine load shedding amounts in a manner that maximizes cumulative future reward, accomplishing coordination within the region rapidly to meet the online application requirements. The DRL based distributed LS scheme is performed to control the China Southern Power Grid (CSG) system. Our results show improved voltage recovery performance by load shedding using the proposed scheme under different unknown test scenarios.",
keywords = "deep reinforcement learning, spatial-temporal information fusion, Under voltage load shedding",
author = "Jingyi Zhang and Chao Lu and Jennie Si and Jie Song and Yinsheng Su",
year = "2018",
month = "10",
day = "10",
doi = "10.1109/IJCNN.2018.8489041",
language = "English (US)",
volume = "2018-July",
booktitle = "2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid

AU - Zhang, Jingyi

AU - Lu, Chao

AU - Si, Jennie

AU - Song, Jie

AU - Su, Yinsheng

PY - 2018/10/10

Y1 - 2018/10/10

N2 - We propose a novel load shedding (LS) scheme against voltage instability using deep reinforcement learning (DRL). Both spatial and temporal information of a large power grid are used in the DRL control scheme. Specifically, both dynamic state variables and grid topology information are inputs to the learning controller. Within the DRL scheme, a deep learning neural network is designed and implemented to automatically extract translation-invariance information about voltage instability. The DRL load shedding controller interacts with system dynamics through a sequence of observations, actions and rewards to determine load shedding amounts in a manner that maximizes cumulative future reward, accomplishing coordination within the region rapidly to meet the online application requirements. The DRL based distributed LS scheme is performed to control the China Southern Power Grid (CSG) system. Our results show improved voltage recovery performance by load shedding using the proposed scheme under different unknown test scenarios.

AB - We propose a novel load shedding (LS) scheme against voltage instability using deep reinforcement learning (DRL). Both spatial and temporal information of a large power grid are used in the DRL control scheme. Specifically, both dynamic state variables and grid topology information are inputs to the learning controller. Within the DRL scheme, a deep learning neural network is designed and implemented to automatically extract translation-invariance information about voltage instability. The DRL load shedding controller interacts with system dynamics through a sequence of observations, actions and rewards to determine load shedding amounts in a manner that maximizes cumulative future reward, accomplishing coordination within the region rapidly to meet the online application requirements. The DRL based distributed LS scheme is performed to control the China Southern Power Grid (CSG) system. Our results show improved voltage recovery performance by load shedding using the proposed scheme under different unknown test scenarios.

KW - deep reinforcement learning

KW - spatial-temporal information fusion

KW - Under voltage load shedding

UR - http://www.scopus.com/inward/record.url?scp=85056528897&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056528897&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2018.8489041

DO - 10.1109/IJCNN.2018.8489041

M3 - Conference contribution

AN - SCOPUS:85056528897

VL - 2018-July

BT - 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -