基于声纹压缩和代价敏感的变压器状态检测评估方法
TM73; 声纹检测技术可以助力巡检人员对变压器状态进行检测和评估.文中提出一种基于声纹压缩和代价敏感的变压器状态检测和评估方法.该方法首先提取变压器音频的声纹特征,然后在频率维度上对声纹特征进行筛选和压缩,最后使用卷积神经网络评估变压器状态,并引入代价敏感损失函数以提高对难检出样本的关注度.以某35 kV变压器为研究对象,通过收集现场音频、模拟实验和样本扩充得到变压器音频数据集.测试结果表明,文中所提方法将声纹维度从1 025维降低到80维,计算量和显存分别降低到1 025维的8.1％和7.7％.同时,所提方法的声纹识别准确率高达83.5％,并将最难检出的短路电流异常状态的召回率从48.2％...
Saved in:
| Published in | 电力工程技术 Vol. 43; no. 3; pp. 209 - 216 |
|---|---|
| Main Authors | , , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
上海交通大学电气工程系,上海 200240%中国能源建设集团广西电力设计研究院有限公司,广西南宁 530007
28.05.2024
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 2096-3203 |
| DOI | 10.12158/j.2096-3203.2024.03.022 |
Cover
| Abstract | TM73; 声纹检测技术可以助力巡检人员对变压器状态进行检测和评估.文中提出一种基于声纹压缩和代价敏感的变压器状态检测和评估方法.该方法首先提取变压器音频的声纹特征,然后在频率维度上对声纹特征进行筛选和压缩,最后使用卷积神经网络评估变压器状态,并引入代价敏感损失函数以提高对难检出样本的关注度.以某35 kV变压器为研究对象,通过收集现场音频、模拟实验和样本扩充得到变压器音频数据集.测试结果表明,文中所提方法将声纹维度从1 025维降低到80维,计算量和显存分别降低到1 025维的8.1％和7.7％.同时,所提方法的声纹识别准确率高达83.5％,并将最难检出的短路电流异常状态的召回率从48.2％提升至63.6％. |
|---|---|
| AbstractList | TM73; 声纹检测技术可以助力巡检人员对变压器状态进行检测和评估.文中提出一种基于声纹压缩和代价敏感的变压器状态检测和评估方法.该方法首先提取变压器音频的声纹特征,然后在频率维度上对声纹特征进行筛选和压缩,最后使用卷积神经网络评估变压器状态,并引入代价敏感损失函数以提高对难检出样本的关注度.以某35 kV变压器为研究对象,通过收集现场音频、模拟实验和样本扩充得到变压器音频数据集.测试结果表明,文中所提方法将声纹维度从1 025维降低到80维,计算量和显存分别降低到1 025维的8.1％和7.7％.同时,所提方法的声纹识别准确率高达83.5％,并将最难检出的短路电流异常状态的召回率从48.2％提升至63.6％. |
| Abstract_FL | Voiceprint detection technology can assist inspectors in assessing the state of transformers.A method for detecting and assessing transformer states based on voiceprint compression and cost-sensitive techniques is proposed.The method first extracts voiceprint features from transformer audio,then filters and compresses these features in the frequency domain.Subsequently,a convolutional neural network is employed to evaluate the transformer's state,incorporating a cost-sensitive loss function to enhance attention towards difficult samples.Using a 35 kV transformer as the experimental subject,transformer audio data is collected through on-site recordings,simulated experiments and sample augmentation.Test results demonstrate that the proposed method reduces the voiceprint dimensionality from 1 025 to 80,decreasing computational complexity and video memory usage to 8.1％and 7.7％of the original 1 025 dimensions,respectively.Simultaneously,the proposed method achieves a voiceprint recognition accuracy of 83.5％and improves the recall rate of the most challenging short-circuit current anomaly from 48.2％to 63.6％. |
| Author | 陈海威 李喆 胡赵宇 陆忻 |
| AuthorAffiliation | 上海交通大学电气工程系,上海 200240%中国能源建设集团广西电力设计研究院有限公司,广西南宁 530007 |
| AuthorAffiliation_xml | – name: 上海交通大学电气工程系,上海 200240%中国能源建设集团广西电力设计研究院有限公司,广西南宁 530007 |
| Author_FL | LI Zhe LU Xin CHEN Haiwei HU Zhaoyu |
| Author_FL_xml | – sequence: 1 fullname: HU Zhaoyu – sequence: 2 fullname: LI Zhe – sequence: 3 fullname: CHEN Haiwei – sequence: 4 fullname: LU Xin |
| Author_xml | – sequence: 1 fullname: 胡赵宇 – sequence: 2 fullname: 李喆 – sequence: 3 fullname: 陈海威 – sequence: 4 fullname: 陆忻 |
| BookMark | eNo9j0tLw0AUhWdRwVr7H9y5SpxH5pGllPqAghtdl0kzKQZJwSBuK2RlpQWhom66yEZXFkSlaX9OZuLPcIriWdzvcC_cw9kCtWSQKAB2EHQRRlTsxS6GPnMIhsQ67LmWEOMaqP_vN0EzTc8DiAVfi9ZBW8-KshjrfF4VCz0eVatXfX9XLvNy-WWmE5PNqudMTx7tST-9VLefZnhj8qH5GH2_ZeVqbh4W5n26DTYieZGq5h8b4Oygfdo6cjonh8et_Y6TIoipQyPBBAklFaEiyLNDekrgkBHKMBIUSokh5AHxGEeyxymPwoB7vo8QJ4pFpAF2f_9eyySSSb8bD64uE5vYjdMw7vfWtSGxUeQHp51nRA |
| ClassificationCodes | TM73 |
| ContentType | Journal Article |
| Copyright | Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| Copyright_xml | – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| DBID | 2B. 4A8 92I 93N PSX TCJ |
| DOI | 10.12158/j.2096-3203.2024.03.022 |
| DatabaseName | Wanfang Data Journals - Hong Kong WANFANG Data Centre Wanfang Data Journals 万方数据期刊 - 香港版 China Online Journals (COJ) China Online Journals (COJ) |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| DocumentTitle_FL | Transformer state detection and assessment method based on voiceprint compression and cost-sensitive techniques |
| EndPage | 216 |
| ExternalDocumentID | jsdjgc202403025 |
| GrantInformation_xml | – fundername: 国家自然科学基金 funderid: (52077133) |
| GroupedDBID | -0C -SC -S~ 2B. 2RA 4A8 5VR 92I 92M 93N 9D9 9DC AFUIB ALMA_UNASSIGNED_HOLDINGS CAJEC CQIGP GROUPED_DOAJ PB1 PB9 PSX Q-- R-C RT3 T8S TCJ U1F U5C |
| ID | FETCH-LOGICAL-s1025-5f8683da58de314de3a4e82d635621850aa2007b34671ac757fdb74991173e6f3 |
| ISSN | 2096-3203 |
| IngestDate | Thu May 29 03:58:35 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 3 |
| Keywords | acoustic pattern recognition 声纹压缩 卷积神经网络 声纹识别 代价敏感 模式识别 cost sensitivity convolutional neural networks 变压器检测 transformer detection acoustic pattern compression pattern recognition |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-s1025-5f8683da58de314de3a4e82d635621850aa2007b34671ac757fdb74991173e6f3 |
| PageCount | 8 |
| ParticipantIDs | wanfang_journals_jsdjgc202403025 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-05-28 |
| PublicationDateYYYYMMDD | 2024-05-28 |
| PublicationDate_xml | – month: 05 year: 2024 text: 2024-05-28 day: 28 |
| PublicationDecade | 2020 |
| PublicationTitle | 电力工程技术 |
| PublicationTitle_FL | Jiangsu Electrical Engineering |
| PublicationYear | 2024 |
| Publisher | 上海交通大学电气工程系,上海 200240%中国能源建设集团广西电力设计研究院有限公司,广西南宁 530007 |
| Publisher_xml | – name: 上海交通大学电气工程系,上海 200240%中国能源建设集团广西电力设计研究院有限公司,广西南宁 530007 |
| SSID | ssib028777775 ssib051810600 ssib051374505 ssj0002857319 ssib036435454 ssib041261494 |
| Score | 2.4070988 |
| Snippet | TM73; 声纹检测技术可以助力巡检人员对变压器状态进行检测和评估.文中提出一种基于声纹压缩和代价敏感的变压器状态检测和评估方法.该方法首先提取变压器音频的声纹特征,然... |
| SourceID | wanfang |
| SourceType | Aggregation Database |
| StartPage | 209 |
| Title | 基于声纹压缩和代价敏感的变压器状态检测评估方法 |
| URI | https://d.wanfangdata.com.cn/periodical/jsdjgc202403025 |
| Volume | 43 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals issn: 2096-3203 databaseCode: DOA dateStart: 20170101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.doaj.org/ omitProxy: true ssIdentifier: ssj0002857319 providerName: Directory of Open Access Journals |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR27btRA0AqhoUEgQLyVgq3QBa_3Xa7vfIqQoEqkdJHP9gWlCBJJmlQBpSIokZCCgCZFGqiIhADl8Tn34DOYWW_OVnKIR7Mazc7Oc88z69tdB8GDAvfOhEI0JBTDDaj_OfzmBIW5TFWmaMaiDA8KP3kqZ-b443kxPzH5qrZraW21M52tjz1X8j9RBRzEFU_J_kNkR0wBATDEF1qIMLR_FWOSCGLaJLYk4djqBDGWkTgkiUJMbBADeB07TJNYhzER0U03Kkb6EogVSSQxQN9GQHNkDqOMRThxeKNrDIGPIVYjjQZZ0o0KoThFANgCDEAsHLEmtu34cFQjdl1GOg2hZaT8CuZpoezkchyL4iwxThxoaAV2gVDkKbGrlGLAtNEuZZSmGbEUAeBRsrGgtapIYEzLuwz00LLqMWiX1l55dItwEulZEumUggjE9bcnEcc__v1pdJzvzmaNulYcXcAsR0642cQ6GZzYUliLWFlzgUR_GT7OBQp9B8GLmuel4H2RUEuRSPg-4IvWQqhb3kUISFTFhI577GaTRl_FiTMz9mYCYCNHY9BkpHG2jwnUaLgDMAqgbmmAwgmIM8U5EBm60GnjMaUaGlzQ9DMO1I6a4-RCb4sYdRpZ-lAwrEWr1BbBwrnBopDV83B5XZd_3rB6Ug1NrT6LyrO551I_1K7a5f4R82kMuLvCuDz6fuZi9aWVfGkxQxpIdJG4EFyMoDIIa69lICdFeF-mqkpsBhU8LDpGOYzTCArcKqcJyhQXVQ4VUEGH0l-BueReZgvF3PeGRnr6DYZowKPfqO-OFC530-XFWvU7eyW47JetU7Z8Bl0NJtafXQuS_t5R72i7v38wPDrsb28NTz73377pHe_3jn8MdncGm3vDj5v9nffQ1f_wafj6-2Dj5WB_Y_Bt6-eXzd7JweDd4eDr7vVgrp3MNmca_rMsjRWKH78WXS01y1Oh84JRDk3KCx3leNMlLBhEmKb4B0iHQQ1G00wJ1c07isNClCpWyC67EUwuP18ubgZTWjLa7cgiDcOMG5rrTl6kgmV5V1LZ4cWtYMqbveAfuysLZ8J2-88kd4JL1U__bjC5-mKtuAdLidXOfRfrX8Puyec |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E5%A3%B0%E7%BA%B9%E5%8E%8B%E7%BC%A9%E5%92%8C%E4%BB%A3%E4%BB%B7%E6%95%8F%E6%84%9F%E7%9A%84%E5%8F%98%E5%8E%8B%E5%99%A8%E7%8A%B6%E6%80%81%E6%A3%80%E6%B5%8B%E8%AF%84%E4%BC%B0%E6%96%B9%E6%B3%95&rft.jtitle=%E7%94%B5%E5%8A%9B%E5%B7%A5%E7%A8%8B%E6%8A%80%E6%9C%AF&rft.au=%E8%83%A1%E8%B5%B5%E5%AE%87&rft.au=%E6%9D%8E%E5%96%86&rft.au=%E9%99%88%E6%B5%B7%E5%A8%81&rft.au=%E9%99%86%E5%BF%BB&rft.date=2024-05-28&rft.pub=%E4%B8%8A%E6%B5%B7%E4%BA%A4%E9%80%9A%E5%A4%A7%E5%AD%A6%E7%94%B5%E6%B0%94%E5%B7%A5%E7%A8%8B%E7%B3%BB%2C%E4%B8%8A%E6%B5%B7+200240%25%E4%B8%AD%E5%9B%BD%E8%83%BD%E6%BA%90%E5%BB%BA%E8%AE%BE%E9%9B%86%E5%9B%A2%E5%B9%BF%E8%A5%BF%E7%94%B5%E5%8A%9B%E8%AE%BE%E8%AE%A1%E7%A0%94%E7%A9%B6%E9%99%A2%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8%2C%E5%B9%BF%E8%A5%BF%E5%8D%97%E5%AE%81+530007&rft.issn=2096-3203&rft.volume=43&rft.issue=3&rft.spage=209&rft.epage=216&rft_id=info:doi/10.12158%2Fj.2096-3203.2024.03.022&rft.externalDocID=jsdjgc202403025 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fjsdjgc%2Fjsdjgc.jpg |