Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks

Huaxiang Cai; David J. Hill

doi:10.1109/APPEEC48164.2020.9220527

Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks

Huaxiang Cai
, David J. Hill

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

1 Citation (Scopus)

Abstract

For machine learning techniques, it is difficult to transfer knowledge between different domains due to the system discrepancy and limited training data. In this paper, we implement a knowledge transfer system (KTS) for long-term voltage stability assessment between power grids. First, system behaviours are converted to heatmaps with a more general method to avoid negative transfer. Then, a deep domain adaptation network (DDAN) is introduced to learn domain-invariant representations with strong semantic separations by adding a maximum-mean-discrepancy calculator. The KTS based on DDAN is performed to transfer knowledge from IEEE 39-bus system to IEEE 14-bus system. Case studies are also given to show its potential under those scenarios of data shortage.

Original language	English
Title of host publication	2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020)
Editors	Feng Wu, Ming Ni
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1345-1349
Number of pages	5
ISBN (Electronic)	9781728157481
ISBN (Print)	9781728157498
DOIs	https://doi.org/10.1109/APPEEC48164.2020.9220527
Publication status	Published - 2020
Externally published	Yes
Event	IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) 2020 - Nanjing, China Duration: 20 Sept 2020 → 23 Sept 2020 Conference number: 12th https://ieeexplore.ieee.org/xpl/conhome/9212309/proceeding (Proceedings)

Publication series

Name	Asia-Pacific Power and Energy Engineering Conference, APPEEC
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Volume	2020-September
ISSN (Print)	2157-4839
ISSN (Electronic)	2157-4847

Conference

Conference	IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) 2020
Abbreviated title	APPEEC 2020
Country/Territory	China
City	Nanjing
Period	20/09/20 → 23/09/20
Internet address	https://ieeexplore.ieee.org/xpl/conhome/9212309/proceeding (Proceedings)

Keywords

convolutional neural networks
deep domain adaptation
knowledge transfer
voltage stability

Access to Document

10.1109/APPEEC48164.2020.9220527

Cite this

Cai, H., & Hill, D. J. (2020). Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks. In F. Wu, & M. Ni (Eds.), 2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020) (pp. 1345-1349). (Asia-Pacific Power and Energy Engineering Conference, APPEEC; Vol. 2020-September). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/APPEEC48164.2020.9220527

Cai, Huaxiang ; Hill, David J. / Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks. 2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020). editor / Feng Wu ; Ming Ni. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2020. pp. 1345-1349 (Asia-Pacific Power and Energy Engineering Conference, APPEEC).

@inproceedings{87fd7ad33cc44ccf85d8634ab448d987,

title = "Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks",

abstract = "For machine learning techniques, it is difficult to transfer knowledge between different domains due to the system discrepancy and limited training data. In this paper, we implement a knowledge transfer system (KTS) for long-term voltage stability assessment between power grids. First, system behaviours are converted to heatmaps with a more general method to avoid negative transfer. Then, a deep domain adaptation network (DDAN) is introduced to learn domain-invariant representations with strong semantic separations by adding a maximum-mean-discrepancy calculator. The KTS based on DDAN is performed to transfer knowledge from IEEE 39-bus system to IEEE 14-bus system. Case studies are also given to show its potential under those scenarios of data shortage.",

keywords = "convolutional neural networks, deep domain adaptation, knowledge transfer, voltage stability",

author = "Huaxiang Cai and Hill, \{David J.\}",

note = "Funding Information: The work described in this paper was fully supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region under the Theme-based Research Scheme through Project No. T23-701/14-N and General Research Fund through Project No. 17208817. Publisher Copyright: {\textcopyright} 2020 IEEE.; IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) 2020, APPEEC 2020 ; Conference date: 20-09-2020 Through 23-09-2020",

year = "2020",

doi = "10.1109/APPEEC48164.2020.9220527",

language = "English",

isbn = "9781728157498",

series = "Asia-Pacific Power and Energy Engineering Conference, APPEEC",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

pages = "1345--1349",

editor = "Feng Wu and Ming Ni",

booktitle = "2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020)",

address = "United States of America",

url = "https://ieeexplore.ieee.org/xpl/conhome/9212309/proceeding",

}

Cai, H & Hill, DJ 2020, Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks. in F Wu & M Ni (eds), 2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020). Asia-Pacific Power and Energy Engineering Conference, APPEEC, vol. 2020-September, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 1345-1349, IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) 2020, Nanjing, China, 20/09/20. https://doi.org/10.1109/APPEEC48164.2020.9220527

Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks. / Cai, Huaxiang; Hill, David J.
2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020). ed. / Feng Wu; Ming Ni. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers, 2020. p. 1345-1349 (Asia-Pacific Power and Energy Engineering Conference, APPEEC; Vol. 2020-September).

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

TY - GEN

T1 - Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks

AU - Cai, Huaxiang

AU - Hill, David J.

N1 - Conference code: 12th

PY - 2020

Y1 - 2020

N2 - For machine learning techniques, it is difficult to transfer knowledge between different domains due to the system discrepancy and limited training data. In this paper, we implement a knowledge transfer system (KTS) for long-term voltage stability assessment between power grids. First, system behaviours are converted to heatmaps with a more general method to avoid negative transfer. Then, a deep domain adaptation network (DDAN) is introduced to learn domain-invariant representations with strong semantic separations by adding a maximum-mean-discrepancy calculator. The KTS based on DDAN is performed to transfer knowledge from IEEE 39-bus system to IEEE 14-bus system. Case studies are also given to show its potential under those scenarios of data shortage.

AB - For machine learning techniques, it is difficult to transfer knowledge between different domains due to the system discrepancy and limited training data. In this paper, we implement a knowledge transfer system (KTS) for long-term voltage stability assessment between power grids. First, system behaviours are converted to heatmaps with a more general method to avoid negative transfer. Then, a deep domain adaptation network (DDAN) is introduced to learn domain-invariant representations with strong semantic separations by adding a maximum-mean-discrepancy calculator. The KTS based on DDAN is performed to transfer knowledge from IEEE 39-bus system to IEEE 14-bus system. Case studies are also given to show its potential under those scenarios of data shortage.

KW - convolutional neural networks

KW - deep domain adaptation

KW - knowledge transfer

KW - voltage stability

UR - https://www.scopus.com/pages/publications/85094317218

U2 - 10.1109/APPEEC48164.2020.9220527

DO - 10.1109/APPEEC48164.2020.9220527

M3 - Conference Paper

AN - SCOPUS:85094317218

SN - 9781728157498

T3 - Asia-Pacific Power and Energy Engineering Conference, APPEEC

SP - 1345

EP - 1349

BT - 2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020)

A2 - Wu, Feng

A2 - Ni, Ming

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

T2 - IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) 2020

Y2 - 20 September 2020 through 23 September 2020

ER -

Cai H, Hill DJ. Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks. In Wu F, Ni M, editors, 2020 12th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2020). Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2020. p. 1345-1349. (Asia-Pacific Power and Energy Engineering Conference, APPEEC). doi: 10.1109/APPEEC48164.2020.9220527

Knowledge transfer for long-term voltage stability assessment between power grids based on deep domain adaptation networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this