A new dynamic security assessment framework based on semi-supervised learning and data editing
•Penetration of renewables increases diversity of power system operation conditions.•Semi-supervised training reduces labeled training data for DSA updating.•Data editing helps to improve performance of a tri-training trained classifier.•Tri-training can help to speed up online DSA classifier updati...
Gespeichert in:
Veröffentlicht in: | Electric power systems research 2019-07, Vol.172, p.221-229 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 229 |
---|---|
container_issue | |
container_start_page | 221 |
container_title | Electric power systems research |
container_volume | 172 |
creator | Liu, Ruidong Verbič, Gregor Ma, Jin |
description | •Penetration of renewables increases diversity of power system operation conditions.•Semi-supervised training reduces labeled training data for DSA updating.•Data editing helps to improve performance of a tri-training trained classifier.•Tri-training can help to speed up online DSA classifier updating.
In this paper, we propose a new online dynamic security assessment (DSA) framework based on semi-supervised learning and data editing. To reduce the number of labeled samples used by supervised learning in conventional DSA, which is required to ensure a high generalization performance of a classifier, we augment the training set with a large number of unlabeled samples that are easily computed. As an alternative to computationally expensive time-domain simulations, the unlabeled samples are labeled by an algorithm called tri-training. To reduce the noise that comes with incorrectly labeled samples, we use data editing, which significantly improves the classification performance. We demonstrate the performance of the proposed framework in a case study using the IEEE 39-bus New England test system with different levels of wind penetration. The results show that the proposed DSA framework reduces the number of labeled samples required to train the neural network used as an online transient stability classifier, which significantly reduces the computational burden associated with the training of the classifier. |
doi_str_mv | 10.1016/j.epsr.2019.03.009 |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2235020514</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0378779619301014</els_id><sourcerecordid>2235020514</sourcerecordid><originalsourceid>FETCH-LOGICAL-c328t-4631d976904314ac2b8732660bebf962adfc688f2bd540377af01616151fba853</originalsourceid><addsrcrecordid>eNp9UMtqwzAQFKWFpml_oCdBz3ZXki3b0EsIfUGgl_ZaIUvrojSWU8lOyN9XIT2XPSwMM7szQ8gtg5wBk_frHLcx5BxYk4PIAZozMmN1JTIOhTwnMxBVnVVVIy_JVYxrAJBNVc7I54J63FN78Lp3hkY0U3DjgeoYMcYe_Ui7oHvcD-GbtjqipYNPtN5lcdpi2LkjtEEdvPNfVHtLrR41RevGBFyTi05vIt787Tn5eHp8X75kq7fn1-VilRnB6zErpGC2qWQDhWCFNrxN1rmU0GLbNZJr2xlZ1x1vbVmkKJXuUuo0JetaXZdiTu5Od7dh-Jkwjmo9TMGnl4pzUQKHkhWJxU8sE4YYA3ZqG1yvw0ExUMce1Vode1THHhUIlXpMooeTCJP_ncOgonHoTUoY0IzKDu4_-S-gSnwx</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2235020514</pqid></control><display><type>article</type><title>A new dynamic security assessment framework based on semi-supervised learning and data editing</title><source>Access via ScienceDirect (Elsevier)</source><creator>Liu, Ruidong ; Verbič, Gregor ; Ma, Jin</creator><creatorcontrib>Liu, Ruidong ; Verbič, Gregor ; Ma, Jin</creatorcontrib><description>•Penetration of renewables increases diversity of power system operation conditions.•Semi-supervised training reduces labeled training data for DSA updating.•Data editing helps to improve performance of a tri-training trained classifier.•Tri-training can help to speed up online DSA classifier updating.
In this paper, we propose a new online dynamic security assessment (DSA) framework based on semi-supervised learning and data editing. To reduce the number of labeled samples used by supervised learning in conventional DSA, which is required to ensure a high generalization performance of a classifier, we augment the training set with a large number of unlabeled samples that are easily computed. As an alternative to computationally expensive time-domain simulations, the unlabeled samples are labeled by an algorithm called tri-training. To reduce the noise that comes with incorrectly labeled samples, we use data editing, which significantly improves the classification performance. We demonstrate the performance of the proposed framework in a case study using the IEEE 39-bus New England test system with different levels of wind penetration. The results show that the proposed DSA framework reduces the number of labeled samples required to train the neural network used as an online transient stability classifier, which significantly reduces the computational burden associated with the training of the classifier.</description><identifier>ISSN: 0378-7796</identifier><identifier>EISSN: 1873-2046</identifier><identifier>DOI: 10.1016/j.epsr.2019.03.009</identifier><language>eng</language><publisher>Amsterdam: Elsevier B.V</publisher><subject>Algorithms ; Classifiers ; Computer simulation ; Data editing ; Data transmission ; Dynamic security assessment ; Editing ; Electric power lines ; Machine learning ; Neural networks ; Noise reduction ; Risk assessment ; Security ; Security systems ; Semi-supervised learning ; Systems stability ; Training ; Transient stability</subject><ispartof>Electric power systems research, 2019-07, Vol.172, p.221-229</ispartof><rights>2019 Elsevier B.V.</rights><rights>Copyright Elsevier Science Ltd. Jul 2019</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c328t-4631d976904314ac2b8732660bebf962adfc688f2bd540377af01616151fba853</citedby><cites>FETCH-LOGICAL-c328t-4631d976904314ac2b8732660bebf962adfc688f2bd540377af01616151fba853</cites><orcidid>0000-0003-4949-768X</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.epsr.2019.03.009$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,780,784,3550,27924,27925,45995</link.rule.ids></links><search><creatorcontrib>Liu, Ruidong</creatorcontrib><creatorcontrib>Verbič, Gregor</creatorcontrib><creatorcontrib>Ma, Jin</creatorcontrib><title>A new dynamic security assessment framework based on semi-supervised learning and data editing</title><title>Electric power systems research</title><description>•Penetration of renewables increases diversity of power system operation conditions.•Semi-supervised training reduces labeled training data for DSA updating.•Data editing helps to improve performance of a tri-training trained classifier.•Tri-training can help to speed up online DSA classifier updating.
In this paper, we propose a new online dynamic security assessment (DSA) framework based on semi-supervised learning and data editing. To reduce the number of labeled samples used by supervised learning in conventional DSA, which is required to ensure a high generalization performance of a classifier, we augment the training set with a large number of unlabeled samples that are easily computed. As an alternative to computationally expensive time-domain simulations, the unlabeled samples are labeled by an algorithm called tri-training. To reduce the noise that comes with incorrectly labeled samples, we use data editing, which significantly improves the classification performance. We demonstrate the performance of the proposed framework in a case study using the IEEE 39-bus New England test system with different levels of wind penetration. The results show that the proposed DSA framework reduces the number of labeled samples required to train the neural network used as an online transient stability classifier, which significantly reduces the computational burden associated with the training of the classifier.</description><subject>Algorithms</subject><subject>Classifiers</subject><subject>Computer simulation</subject><subject>Data editing</subject><subject>Data transmission</subject><subject>Dynamic security assessment</subject><subject>Editing</subject><subject>Electric power lines</subject><subject>Machine learning</subject><subject>Neural networks</subject><subject>Noise reduction</subject><subject>Risk assessment</subject><subject>Security</subject><subject>Security systems</subject><subject>Semi-supervised learning</subject><subject>Systems stability</subject><subject>Training</subject><subject>Transient stability</subject><issn>0378-7796</issn><issn>1873-2046</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNp9UMtqwzAQFKWFpml_oCdBz3ZXki3b0EsIfUGgl_ZaIUvrojSWU8lOyN9XIT2XPSwMM7szQ8gtg5wBk_frHLcx5BxYk4PIAZozMmN1JTIOhTwnMxBVnVVVIy_JVYxrAJBNVc7I54J63FN78Lp3hkY0U3DjgeoYMcYe_Ui7oHvcD-GbtjqipYNPtN5lcdpi2LkjtEEdvPNfVHtLrR41RevGBFyTi05vIt787Tn5eHp8X75kq7fn1-VilRnB6zErpGC2qWQDhWCFNrxN1rmU0GLbNZJr2xlZ1x1vbVmkKJXuUuo0JetaXZdiTu5Od7dh-Jkwjmo9TMGnl4pzUQKHkhWJxU8sE4YYA3ZqG1yvw0ExUMce1Vode1THHhUIlXpMooeTCJP_ncOgonHoTUoY0IzKDu4_-S-gSnwx</recordid><startdate>201907</startdate><enddate>201907</enddate><creator>Liu, Ruidong</creator><creator>Verbič, Gregor</creator><creator>Ma, Jin</creator><general>Elsevier B.V</general><general>Elsevier Science Ltd</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SP</scope><scope>8FD</scope><scope>FR3</scope><scope>KR7</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0003-4949-768X</orcidid></search><sort><creationdate>201907</creationdate><title>A new dynamic security assessment framework based on semi-supervised learning and data editing</title><author>Liu, Ruidong ; Verbič, Gregor ; Ma, Jin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c328t-4631d976904314ac2b8732660bebf962adfc688f2bd540377af01616151fba853</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Algorithms</topic><topic>Classifiers</topic><topic>Computer simulation</topic><topic>Data editing</topic><topic>Data transmission</topic><topic>Dynamic security assessment</topic><topic>Editing</topic><topic>Electric power lines</topic><topic>Machine learning</topic><topic>Neural networks</topic><topic>Noise reduction</topic><topic>Risk assessment</topic><topic>Security</topic><topic>Security systems</topic><topic>Semi-supervised learning</topic><topic>Systems stability</topic><topic>Training</topic><topic>Transient stability</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Liu, Ruidong</creatorcontrib><creatorcontrib>Verbič, Gregor</creatorcontrib><creatorcontrib>Ma, Jin</creatorcontrib><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>Civil Engineering Abstracts</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>Electric power systems research</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Ruidong</au><au>Verbič, Gregor</au><au>Ma, Jin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A new dynamic security assessment framework based on semi-supervised learning and data editing</atitle><jtitle>Electric power systems research</jtitle><date>2019-07</date><risdate>2019</risdate><volume>172</volume><spage>221</spage><epage>229</epage><pages>221-229</pages><issn>0378-7796</issn><eissn>1873-2046</eissn><abstract>•Penetration of renewables increases diversity of power system operation conditions.•Semi-supervised training reduces labeled training data for DSA updating.•Data editing helps to improve performance of a tri-training trained classifier.•Tri-training can help to speed up online DSA classifier updating.
In this paper, we propose a new online dynamic security assessment (DSA) framework based on semi-supervised learning and data editing. To reduce the number of labeled samples used by supervised learning in conventional DSA, which is required to ensure a high generalization performance of a classifier, we augment the training set with a large number of unlabeled samples that are easily computed. As an alternative to computationally expensive time-domain simulations, the unlabeled samples are labeled by an algorithm called tri-training. To reduce the noise that comes with incorrectly labeled samples, we use data editing, which significantly improves the classification performance. We demonstrate the performance of the proposed framework in a case study using the IEEE 39-bus New England test system with different levels of wind penetration. The results show that the proposed DSA framework reduces the number of labeled samples required to train the neural network used as an online transient stability classifier, which significantly reduces the computational burden associated with the training of the classifier.</abstract><cop>Amsterdam</cop><pub>Elsevier B.V</pub><doi>10.1016/j.epsr.2019.03.009</doi><tpages>9</tpages><orcidid>https://orcid.org/0000-0003-4949-768X</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0378-7796 |
ispartof | Electric power systems research, 2019-07, Vol.172, p.221-229 |
issn | 0378-7796 1873-2046 |
language | eng |
recordid | cdi_proquest_journals_2235020514 |
source | Access via ScienceDirect (Elsevier) |
subjects | Algorithms Classifiers Computer simulation Data editing Data transmission Dynamic security assessment Editing Electric power lines Machine learning Neural networks Noise reduction Risk assessment Security Security systems Semi-supervised learning Systems stability Training Transient stability |
title | A new dynamic security assessment framework based on semi-supervised learning and data editing |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T22%3A34%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20new%20dynamic%20security%20assessment%20framework%20based%20on%20semi-supervised%20learning%20and%20data%20editing&rft.jtitle=Electric%20power%20systems%20research&rft.au=Liu,%20Ruidong&rft.date=2019-07&rft.volume=172&rft.spage=221&rft.epage=229&rft.pages=221-229&rft.issn=0378-7796&rft.eissn=1873-2046&rft_id=info:doi/10.1016/j.epsr.2019.03.009&rft_dat=%3Cproquest_cross%3E2235020514%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2235020514&rft_id=info:pmid/&rft_els_id=S0378779619301014&rfr_iscdi=true |