Deep anomaly detection in horizontal axis wind turbines using Graph Convolutional Autoencoders for Multivariate Time series
Wind power is one of the fastest-growing renewable energy sectors instrumental in the ongoing decarbonization process. However, wind turbines are subjected to a wide range of dynamic loads which can cause more frequent failures and downtime periods, leading to ever-increasing attention to effective...
Saved in:
| Published in: | Energy and AI Vol. 8; p. 100145 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
01.05.2022
Elsevier |
| Subjects: | |
| ISSN: | 2666-5468, 2666-5468 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Wind power is one of the fastest-growing renewable energy sectors instrumental in the ongoing decarbonization process. However, wind turbines are subjected to a wide range of dynamic loads which can cause more frequent failures and downtime periods, leading to ever-increasing attention to effective Condition Monitoring strategies. In this paper, we propose a novel unsupervised deep anomaly detection framework to detect anomalies in wind turbines based on SCADA data. We introduce a promising neural architecture, namely a Graph Convolutional Autoencoder for Multivariate Time series, to model the sensor network as a dynamical functional graph. This structure improves the unsupervised learning capabilities of Autoencoders by considering individual sensor measurements together with the nonlinear correlations existing among signals. On this basis, we developed a deep anomaly detection framework that was validated on 12 failure events occurred during 20 months of operation of four wind turbines. The results show that the proposed framework successfully detects anomalies and anticipates SCADA alarms by outperforming other two recent neural approaches.
[Display omitted]
•Deep Learning approach for early anomaly detection in wind turbines.•Architecture based on Autoencoders and Graph Convolutional Networks.•Unsupervised and data-driven approach.•Training on healthy operating conditions of wind turbines.•Troubleshooting and faulty component identification through model residuals. |
|---|---|
| AbstractList | Wind power is one of the fastest-growing renewable energy sectors instrumental in the ongoing decarbonization process. However, wind turbines are subjected to a wide range of dynamic loads which can cause more frequent failures and downtime periods, leading to ever-increasing attention to effective Condition Monitoring strategies. In this paper, we propose a novel unsupervised deep anomaly detection framework to detect anomalies in wind turbines based on SCADA data. We introduce a promising neural architecture, namely a Graph Convolutional Autoencoder for Multivariate Time series, to model the sensor network as a dynamical functional graph. This structure improves the unsupervised learning capabilities of Autoencoders by considering individual sensor measurements together with the nonlinear correlations existing among signals. On this basis, we developed a deep anomaly detection framework that was validated on 12 failure events occurred during 20 months of operation of four wind turbines. The results show that the proposed framework successfully detects anomalies and anticipates SCADA alarms by outperforming other two recent neural approaches. Wind power is one of the fastest-growing renewable energy sectors instrumental in the ongoing decarbonization process. However, wind turbines are subjected to a wide range of dynamic loads which can cause more frequent failures and downtime periods, leading to ever-increasing attention to effective Condition Monitoring strategies. In this paper, we propose a novel unsupervised deep anomaly detection framework to detect anomalies in wind turbines based on SCADA data. We introduce a promising neural architecture, namely a Graph Convolutional Autoencoder for Multivariate Time series, to model the sensor network as a dynamical functional graph. This structure improves the unsupervised learning capabilities of Autoencoders by considering individual sensor measurements together with the nonlinear correlations existing among signals. On this basis, we developed a deep anomaly detection framework that was validated on 12 failure events occurred during 20 months of operation of four wind turbines. The results show that the proposed framework successfully detects anomalies and anticipates SCADA alarms by outperforming other two recent neural approaches. [Display omitted] •Deep Learning approach for early anomaly detection in wind turbines.•Architecture based on Autoencoders and Graph Convolutional Networks.•Unsupervised and data-driven approach.•Training on healthy operating conditions of wind turbines.•Troubleshooting and faulty component identification through model residuals. |
| ArticleNumber | 100145 |
| Author | Miele, Eric Stefan Bonacina, Fabrizio Corsini, Alessandro |
| Author_xml | – sequence: 1 givenname: Eric Stefan surname: Miele fullname: Miele, Eric Stefan email: ericstefan.miele@uniroma1.it – sequence: 2 givenname: Fabrizio surname: Bonacina fullname: Bonacina, Fabrizio – sequence: 3 givenname: Alessandro surname: Corsini fullname: Corsini, Alessandro |
| BookMark | eNp9kc9uEzEQxleoSJTSJ-DiF0iwvd5_Bw5VgFKpqJdytsbecTrRxo5sb0rg5XEaKqEeehpr5N838833vjrzwWNVfRR8KbhoP22WuD4ALSWXsnS4UM2b6ly2bbtoVNuf_fd-V12mtOGcy0aIoevOqz9fEHcMfNjCdGAjZrSZgmfk2UOI9Dv4DBODX5TYI_mR5Tka8pjYnMiv2XWE3QNbBb8P03wEy-erOQf0NowYE3Mhsh_zlGkPkSAju6ctsoSRMH2o3jqYEl7-qxfVz29f71ffF7d31zerq9uFVULlheJK2Q5qB-iakUtXNoexxq5rW6kaOWA7WNd3g-iVsdwJIxpTzHUSVC96W19UNyfdMcBG7yJtIR50ANJPjRDXGmImO6GWyta9M7UxMCirOlOG9yisE7YRvZFFazhp2RhSiui0pQxH5zkCTVpwfQxFb_RTKPoYij6FUtj6Bfu8y-vU5xOF5UR7wqiTpXJfHCmWsIoHepX_Cwmpqys |
| CitedBy_id | crossref_primary_10_1016_j_engfailanal_2024_108354 crossref_primary_10_3390_s23125695 crossref_primary_10_3390_s25144499 crossref_primary_10_3390_s22093288 crossref_primary_10_1016_j_compind_2024_104097 crossref_primary_10_3390_en16124678 crossref_primary_10_1109_TIM_2025_3565242 crossref_primary_10_1007_s12206_024_0616_9 crossref_primary_10_1016_j_engstruct_2024_118208 crossref_primary_10_1016_j_ress_2023_109846 crossref_primary_10_3390_en16020620 crossref_primary_10_3390_electronics12061313 crossref_primary_10_3390_en18143630 crossref_primary_10_1016_j_ins_2023_119682 crossref_primary_10_3390_s22186955 crossref_primary_10_1007_s10489_022_04337_y crossref_primary_10_1002_eqe_4047 crossref_primary_10_1016_j_egyai_2024_100467 crossref_primary_10_1080_10589759_2025_2541053 crossref_primary_10_3390_machines12120894 crossref_primary_10_1088_1755_1315_1073_1_012005 crossref_primary_10_3390_s23146502 crossref_primary_10_3390_jsan14050089 crossref_primary_10_1016_j_jnca_2025_104116 crossref_primary_10_1109_ACCESS_2024_3350978 crossref_primary_10_1016_j_measurement_2025_118538 crossref_primary_10_3390_en16093719 crossref_primary_10_3390_ijtpp10030014 crossref_primary_10_1016_j_egyr_2024_10_041 crossref_primary_10_1109_TII_2022_3228691 crossref_primary_10_1080_15435075_2022_2117554 crossref_primary_10_1016_j_engappai_2024_108363 crossref_primary_10_1016_j_ress_2023_109258 crossref_primary_10_1016_j_measurement_2025_118664 crossref_primary_10_1088_1361_6501_ad6f33 crossref_primary_10_1016_j_ress_2025_110995 crossref_primary_10_1016_j_ymssp_2024_112175 |
| Cites_doi | 10.1016/S0167-6105(98)00034-8 10.1002/we.2404 10.3390/en13123132 10.1016/j.renene.2021.03.078 10.1016/j.renene.2020.04.148 10.1016/j.renene.2018.10.047 10.1016/j.measurement.2021.109094 10.1109/TIE.2010.2103910 10.1145/3439950 10.1109/ICCV.2019.00936 10.1109/TITS.2019.2910560 10.1016/j.rser.2007.05.008 10.3390/en13184702 10.1371/journal.pone.0087357 10.2514/8.5282 10.1016/j.isatra.2018.11.041 10.1016/j.renene.2019.09.041 10.1016/j.renene.2020.06.154 10.1145/3394486.3403118 10.1021/ac60214a047 10.3390/w9100796 10.1109/TPAS.1981.316586 10.1002/we.2628 10.1016/j.rser.2012.06.006 10.1186/s40649-019-0069-y 10.1016/j.renene.2019.07.033 |
| ContentType | Journal Article |
| Copyright | 2022 |
| Copyright_xml | – notice: 2022 |
| DBID | 6I. AAFTH AAYXX CITATION DOA |
| DOI | 10.1016/j.egyai.2022.100145 |
| DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef DOAJ Directory of Open Access Journals |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 2666-5468 |
| ExternalDocumentID | oai_doaj_org_article_24c38fb3bba94c47b4048e1cf1c518b2 10_1016_j_egyai_2022_100145 S2666546822000076 |
| GroupedDBID | 6I. AAEDW AAFTH AAXUO ALMA_UNASSIGNED_HOLDINGS AMRAJ EBS FDB GROUPED_DOAJ M41 M~E NCXOZ OK1 ROL 0R~ AALRI AAYWO AAYXX ACVFH ADCNI ADVLN AEUPX AFJKZ AFPUW AIGII AITUG AKBMS AKYEP APXCP CITATION |
| ID | FETCH-LOGICAL-c414t-4044c7a3faef5d02f197ad3e776624529e69cf879184bc0f1b15b97772a4818c3 |
| IEDL.DBID | DOA |
| ISICitedReferencesCount | 63 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001106796500003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 2666-5468 |
| IngestDate | Fri Oct 03 12:53:20 EDT 2025 Tue Nov 18 21:50:02 EST 2025 Thu Nov 20 00:50:32 EST 2025 Tue Jul 25 20:58:13 EDT 2023 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Condition monitoring Graph Convolutional Autoencoder Deep anomaly detection SCADA data Early fault detection Wind turbine Multivariate Time series |
| Language | English |
| License | This is an open access article under the CC BY-NC-ND license. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c414t-4044c7a3faef5d02f197ad3e776624529e69cf879184bc0f1b15b97772a4818c3 |
| OpenAccessLink | https://doaj.org/article/24c38fb3bba94c47b4048e1cf1c518b2 |
| ParticipantIDs | doaj_primary_oai_doaj_org_article_24c38fb3bba94c47b4048e1cf1c518b2 crossref_citationtrail_10_1016_j_egyai_2022_100145 crossref_primary_10_1016_j_egyai_2022_100145 elsevier_sciencedirect_doi_10_1016_j_egyai_2022_100145 |
| PublicationCentury | 2000 |
| PublicationDate | May 2022 2022-05-00 2022-05-01 |
| PublicationDateYYYYMMDD | 2022-05-01 |
| PublicationDate_xml | – month: 05 year: 2022 text: May 2022 |
| PublicationDecade | 2020 |
| PublicationTitle | Energy and AI |
| PublicationYear | 2022 |
| Publisher | Elsevier Ltd Elsevier |
| Publisher_xml | – name: Elsevier Ltd – name: Elsevier |
| References | Fung, Scheffler, Stolpe (b2) 1981; PAS-100 Chen, Li, Chen, Wang, Jiang (b22) 2020; 147 Zhang, Tong, Xu, Maciejewski (b28) 2019; 6 Xiang, Wang, Yang, Hu, Su (b18) 2021; 175 Wilkinson M, Harman K, Hendriks B, Spinato F, van Delft T, Garrad G, et al. Measuring wind turbine reliability-results of the reliawind project. In: EWEA conference. 2011, p. 1–8. Prechelt (b39) 1998 Cui, Bangalore, Bertling Tjernberg (b14) 2021 Lepot, Aubin, Clemens (b37) 2017; 9 Stetco, Dinmohammadi, Zhao, Robu, Flynn, Barnes (b8) 2019; 133 Menezes, Mendes, Almeida, Farinha (b11) 2020; 13 Savitzky, Golay (b38) 1964; 36 Maldonado-Correa, Martín-Martínez, Artigao, Gómez-Lázaro (b13) 2020; 13 Kong, Tang, Deng, Liu, Han (b17) 2020; 146 Dao, Kazemtabrizi, Crabtree (b7) 2019; 22 Renström, Bangalore, Highcock (b21) 2020; 157 Chen, Liu, Chu, Liu, Xue (b23) 2021; 172 Yu, Gu (b25) 2019; 20 Yan, Ai, Yang, Tong (b27) 2020 Sesto, Casale (b3) 1998; 74 Liserre, Cardenas, Molinas, Rodriguez (b1) 2011; 58 EDP (b31) 2016 Madsen, Risø (b40) 2008 Kim, Parthasarathy, Uluyol, Foslien, Sheng, Fleming (b6) 2011; 54686 Lebranchu, Charbonnier, Bérenguer, Prévost (b10) 2019; 87 Ross (b35) 2014; 9 Hu, Qu, Work (b26) 2020 Kipf, Welling (b34) 2017 Wu, Pan, Chen, Long, Zhang, Philip (b30) 2020 Sahnoun M, Bagui F, Messaadia M. Failure analysis of onshore wind farms based on experimental data. In: Mediterranean conference on information & communication technologies’ 2015. 2015. Pang, Shen, Cao, Hengel (b15) 2021; 54 Wee, Yang, Chou, Padilan (b4) 2012; 16 Chalapathy, Chawla (b24) 2019 Hameed, Hong, Cho, Ahn, Song (b5) 2009; 13 Kelley (b33) 1960; 30 Zaher, McArthur, Infield, Patel (b9) 2009; 12 Pang, He, Jiang, Xie (b16) 2020; 161 Ulmer, Jarlskog, Pizza, Manninen, Goren Huber (b12) 2020 Wu Z, Pan S, Long G, Jiang J, Chang X, Zhang C. Connecting the dots: Multivariate time series forecasting with graph neural networks. In: Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining. 2020, p. 753–63. Li G, Muller M, Thabet A, Ghanem B. Deepgcns: Can gcns go as deep as cnns?. In: Proceedings of the IEEE/CVF international conference on computer vision. 2019, p. 9267–76. Yu, Yin, Zhu (b19) 2018 Arora, Basu, Mianjy, Mukherjee (b36) 2016 Baldi (b32) 2012 Zhang (10.1016/j.egyai.2022.100145_b28) 2019; 6 Pang (10.1016/j.egyai.2022.100145_b16) 2020; 161 Hameed (10.1016/j.egyai.2022.100145_b5) 2009; 13 Kelley (10.1016/j.egyai.2022.100145_b33) 1960; 30 Kim (10.1016/j.egyai.2022.100145_b6) 2011; 54686 Chalapathy (10.1016/j.egyai.2022.100145_b24) 2019 Zaher (10.1016/j.egyai.2022.100145_b9) 2009; 12 Wee (10.1016/j.egyai.2022.100145_b4) 2012; 16 Yu (10.1016/j.egyai.2022.100145_b19) 2018 Pang (10.1016/j.egyai.2022.100145_b15) 2021; 54 Yan (10.1016/j.egyai.2022.100145_b27) 2020 Dao (10.1016/j.egyai.2022.100145_b7) 2019; 22 Arora (10.1016/j.egyai.2022.100145_b36) 2016 Lepot (10.1016/j.egyai.2022.100145_b37) 2017; 9 Chen (10.1016/j.egyai.2022.100145_b22) 2020; 147 Kipf (10.1016/j.egyai.2022.100145_b34) 2017 Savitzky (10.1016/j.egyai.2022.100145_b38) 1964; 36 Liserre (10.1016/j.egyai.2022.100145_b1) 2011; 58 Chen (10.1016/j.egyai.2022.100145_b23) 2021; 172 Lebranchu (10.1016/j.egyai.2022.100145_b10) 2019; 87 Sesto (10.1016/j.egyai.2022.100145_b3) 1998; 74 Prechelt (10.1016/j.egyai.2022.100145_b39) 1998 Renström (10.1016/j.egyai.2022.100145_b21) 2020; 157 Maldonado-Correa (10.1016/j.egyai.2022.100145_b13) 2020; 13 Baldi (10.1016/j.egyai.2022.100145_b32) 2012 EDP (10.1016/j.egyai.2022.100145_b31) 2016 10.1016/j.egyai.2022.100145_b29 Hu (10.1016/j.egyai.2022.100145_b26) 2020 Madsen (10.1016/j.egyai.2022.100145_b40) 2008 Kong (10.1016/j.egyai.2022.100145_b17) 2020; 146 Yu (10.1016/j.egyai.2022.100145_b25) 2019; 20 Ulmer (10.1016/j.egyai.2022.100145_b12) 2020 Xiang (10.1016/j.egyai.2022.100145_b18) 2021; 175 Fung (10.1016/j.egyai.2022.100145_b2) 1981; PAS-100 Stetco (10.1016/j.egyai.2022.100145_b8) 2019; 133 Wu (10.1016/j.egyai.2022.100145_b30) 2020 Cui (10.1016/j.egyai.2022.100145_b14) 2021 10.1016/j.egyai.2022.100145_b41 Menezes (10.1016/j.egyai.2022.100145_b11) 2020; 13 10.1016/j.egyai.2022.100145_b20 Ross (10.1016/j.egyai.2022.100145_b35) 2014; 9 10.1016/j.egyai.2022.100145_b42 |
| References_xml | – volume: PAS-100 start-page: 1176 year: 1981 end-page: 1182 ident: b2 article-title: Wind energy - a utility perspective publication-title: IEEE Trans Power Appar Syst – reference: Wu Z, Pan S, Long G, Jiang J, Chang X, Zhang C. Connecting the dots: Multivariate time series forecasting with graph neural networks. In: Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining. 2020, p. 753–63. – volume: 133 start-page: 620 year: 2019 end-page: 635 ident: b8 article-title: Machine learning methods for wind turbine condition monitoring: A review publication-title: Renew Energy – year: 2020 ident: b26 article-title: Graph convolutional networks for traffic anomaly – volume: 161 start-page: 510 year: 2020 end-page: 524 ident: b16 article-title: Spatio-temporal fusion neural network for multi-class fault diagnosis of wind turbines based on SCADA data publication-title: Renew Energy – volume: 87 start-page: 272 year: 2019 end-page: 281 ident: b10 article-title: A combined mono-and multi-turbine approach for fault indicator synthesis and wind turbine monitoring using SCADA data publication-title: ISA Trans – volume: 54686 start-page: 2071 year: 2011 end-page: 2079 ident: b6 article-title: Use of SCADA data for failure detection in wind turbines publication-title: Energy sustainability – start-page: 55 year: 1998 end-page: 69 ident: b39 article-title: Early stopping-but when? publication-title: Neural networks: tricks of the trade – volume: 12 start-page: 574 year: 2009 end-page: 593 ident: b9 article-title: Online wind turbine fault detection through automated SCADA data analysis publication-title: Wind Energy Int J Prog Appl Wind Power Convers Technol – volume: 30 start-page: 947 year: 1960 end-page: 954 ident: b33 article-title: Gradient theory of optimal flight paths publication-title: Ars J – volume: 54 start-page: 1 year: 2021 end-page: 38 ident: b15 article-title: Deep learning for anomaly detection: A review publication-title: ACM Comput Surv – volume: 157 start-page: 647 year: 2020 end-page: 659 ident: b21 article-title: System-wide anomaly detection in wind turbines using deep autoencoders publication-title: Renew Energy – volume: 13 start-page: 3132 year: 2020 ident: b13 article-title: Using SCADA data for wind turbine condition monitoring: A systematic literature review publication-title: Energies – volume: 13 start-page: 1 year: 2009 end-page: 39 ident: b5 article-title: Condition monitoring and fault detection of wind turbines and related algorithms: A review publication-title: Renew Sustain Energy Rev – volume: 58 start-page: 1081 year: 2011 end-page: 1095 ident: b1 article-title: Overview of multi-MW wind turbines and wind parks publication-title: IEEE Trans Ind Electr – year: 2016 ident: b36 article-title: Understanding deep neural networks with rectified linear units – start-page: 37 year: 2012 end-page: 49 ident: b32 article-title: Autoencoders, unsupervised learning, and deep architectures publication-title: Proceedings of ICML workshop on unsupervised and transfer learning – reference: Wilkinson M, Harman K, Hendriks B, Spinato F, van Delft T, Garrad G, et al. Measuring wind turbine reliability-results of the reliawind project. In: EWEA conference. 2011, p. 1–8. – volume: 6 start-page: 1 year: 2019 end-page: 23 ident: b28 article-title: Graph convolutional networks: a comprehensive review publication-title: Comput Soc Netw – volume: 22 start-page: 1848 year: 2019 end-page: 1871 ident: b7 article-title: Wind turbine reliability data review and impacts on levelised cost of energy publication-title: Wind Energy – volume: 20 start-page: 3940 year: 2019 end-page: 3951 ident: b25 article-title: Real-time traffic speed estimation with graph convolutional generative autoencoder publication-title: IEEE Trans Intell Transp Syst – volume: 74 start-page: 375 year: 1998 end-page: 387 ident: b3 article-title: Exploitation of wind as an energy source to meet the world’s electricity demand publication-title: J Wind Eng Ind Aerodyn – start-page: 9 year: 2020 ident: b12 article-title: Early fault detection based on wind turbine scada data using convolutional neural networks publication-title: 5th European conference of the prognostics and health management society, virtual conference, 27-31 July 2020. Vol. 5 – year: 2019 ident: b24 article-title: Deep learning for anomaly detection: A survey – volume: 147 start-page: 1469 year: 2020 end-page: 1480 ident: b22 article-title: Anomaly detection for wind turbines based on the reconstruction of condition parameters using stacked denoising autoencoders publication-title: Renew Energy – volume: 172 start-page: 829 year: 2021 end-page: 840 ident: b23 article-title: Anomaly detection and critical SCADA parameters identification for wind turbines based on LSTM-AE neural network publication-title: Renew Energy – year: 2021 ident: b14 article-title: A fault detection framework using recurrent neural networks for condition monitoring of wind turbines publication-title: Wind Energy – volume: 9 start-page: 796 year: 2017 ident: b37 article-title: Interpolation in time series: An introductive overview of existing methods, their performance criteria and uncertainty assessment publication-title: Water – start-page: 1 year: 2020 end-page: 23 ident: b27 article-title: Graph convolutional autoencoder model for the shape coding and cognition of buildings in maps publication-title: Int J Geogr Inf Sci – reference: Li G, Muller M, Thabet A, Ghanem B. Deepgcns: Can gcns go as deep as cnns?. In: Proceedings of the IEEE/CVF international conference on computer vision. 2019, p. 9267–76. – volume: 175 year: 2021 ident: b18 article-title: Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism publication-title: Measurement – volume: 13 start-page: 4702 year: 2020 ident: b11 article-title: Wind farm and resource datasets: A comprehensive survey and overview publication-title: Energies – year: 2008 ident: b40 article-title: Introduction to the IEC 61400-1 standard – volume: 146 start-page: 760 year: 2020 end-page: 768 ident: b17 article-title: Condition monitoring of wind turbines based on spatio-temporal fusion of SCADA data by convolutional neural networks and gated recurrent units publication-title: Renew Energy – start-page: 3634 year: 2018 end-page: 3640 ident: b19 article-title: Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting publication-title: Proceedings of the 27th international joint conference on artificial intelligence – reference: Sahnoun M, Bagui F, Messaadia M. Failure analysis of onshore wind farms based on experimental data. In: Mediterranean conference on information & communication technologies’ 2015. 2015. – year: 2016 ident: b31 article-title: Wind farm 1 - failures – volume: 36 start-page: 1627 year: 1964 end-page: 1639 ident: b38 article-title: Smoothing and differentiation of data by simplified least squares procedures publication-title: Anal Chem – volume: 16 start-page: 5451 year: 2012 end-page: 5465 ident: b4 article-title: Renewable energy supply chains, performance, application barriers, and strategies for further development publication-title: Renew Sustain Energy Rev – start-page: 1 year: 2017 ident: b34 article-title: Semi-supervised classification with graph convolutional networks publication-title: Proceedings of the 5th international conference on learning representations – volume: 9 year: 2014 ident: b35 article-title: Mutual information between discrete and continuous data sets publication-title: PLoS One – year: 2020 ident: b30 article-title: A comprehensive survey on graph neural networks publication-title: IEEE Trans Neural Netw Learn Syst – volume: 74 start-page: 375 year: 1998 ident: 10.1016/j.egyai.2022.100145_b3 article-title: Exploitation of wind as an energy source to meet the world’s electricity demand publication-title: J Wind Eng Ind Aerodyn doi: 10.1016/S0167-6105(98)00034-8 – volume: 22 start-page: 1848 issue: 12 year: 2019 ident: 10.1016/j.egyai.2022.100145_b7 article-title: Wind turbine reliability data review and impacts on levelised cost of energy publication-title: Wind Energy doi: 10.1002/we.2404 – volume: 12 start-page: 574 issue: 6 year: 2009 ident: 10.1016/j.egyai.2022.100145_b9 article-title: Online wind turbine fault detection through automated SCADA data analysis publication-title: Wind Energy Int J Prog Appl Wind Power Convers Technol – year: 2020 ident: 10.1016/j.egyai.2022.100145_b26 – start-page: 55 year: 1998 ident: 10.1016/j.egyai.2022.100145_b39 article-title: Early stopping-but when? – volume: 13 start-page: 3132 issue: 12 year: 2020 ident: 10.1016/j.egyai.2022.100145_b13 article-title: Using SCADA data for wind turbine condition monitoring: A systematic literature review publication-title: Energies doi: 10.3390/en13123132 – volume: 172 start-page: 829 year: 2021 ident: 10.1016/j.egyai.2022.100145_b23 article-title: Anomaly detection and critical SCADA parameters identification for wind turbines based on LSTM-AE neural network publication-title: Renew Energy doi: 10.1016/j.renene.2021.03.078 – volume: 157 start-page: 647 year: 2020 ident: 10.1016/j.egyai.2022.100145_b21 article-title: System-wide anomaly detection in wind turbines using deep autoencoders publication-title: Renew Energy doi: 10.1016/j.renene.2020.04.148 – volume: 133 start-page: 620 year: 2019 ident: 10.1016/j.egyai.2022.100145_b8 article-title: Machine learning methods for wind turbine condition monitoring: A review publication-title: Renew Energy doi: 10.1016/j.renene.2018.10.047 – year: 2019 ident: 10.1016/j.egyai.2022.100145_b24 – volume: 175 year: 2021 ident: 10.1016/j.egyai.2022.100145_b18 article-title: Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism publication-title: Measurement doi: 10.1016/j.measurement.2021.109094 – volume: 58 start-page: 1081 issue: 4 year: 2011 ident: 10.1016/j.egyai.2022.100145_b1 article-title: Overview of multi-MW wind turbines and wind parks publication-title: IEEE Trans Ind Electr doi: 10.1109/TIE.2010.2103910 – volume: 54 start-page: 1 issue: 2 year: 2021 ident: 10.1016/j.egyai.2022.100145_b15 article-title: Deep learning for anomaly detection: A review publication-title: ACM Comput Surv doi: 10.1145/3439950 – ident: 10.1016/j.egyai.2022.100145_b29 doi: 10.1109/ICCV.2019.00936 – year: 2016 ident: 10.1016/j.egyai.2022.100145_b36 – volume: 20 start-page: 3940 issue: 10 year: 2019 ident: 10.1016/j.egyai.2022.100145_b25 article-title: Real-time traffic speed estimation with graph convolutional generative autoencoder publication-title: IEEE Trans Intell Transp Syst doi: 10.1109/TITS.2019.2910560 – year: 2020 ident: 10.1016/j.egyai.2022.100145_b30 article-title: A comprehensive survey on graph neural networks publication-title: IEEE Trans Neural Netw Learn Syst – volume: 13 start-page: 1 issue: 1 year: 2009 ident: 10.1016/j.egyai.2022.100145_b5 article-title: Condition monitoring and fault detection of wind turbines and related algorithms: A review publication-title: Renew Sustain Energy Rev doi: 10.1016/j.rser.2007.05.008 – volume: 13 start-page: 4702 issue: 18 year: 2020 ident: 10.1016/j.egyai.2022.100145_b11 article-title: Wind farm and resource datasets: A comprehensive survey and overview publication-title: Energies doi: 10.3390/en13184702 – volume: 9 issue: 2 year: 2014 ident: 10.1016/j.egyai.2022.100145_b35 article-title: Mutual information between discrete and continuous data sets publication-title: PLoS One doi: 10.1371/journal.pone.0087357 – volume: 30 start-page: 947 issue: 10 year: 1960 ident: 10.1016/j.egyai.2022.100145_b33 article-title: Gradient theory of optimal flight paths publication-title: Ars J doi: 10.2514/8.5282 – ident: 10.1016/j.egyai.2022.100145_b42 – volume: 87 start-page: 272 year: 2019 ident: 10.1016/j.egyai.2022.100145_b10 article-title: A combined mono-and multi-turbine approach for fault indicator synthesis and wind turbine monitoring using SCADA data publication-title: ISA Trans doi: 10.1016/j.isatra.2018.11.041 – start-page: 3634 year: 2018 ident: 10.1016/j.egyai.2022.100145_b19 article-title: Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting – volume: 147 start-page: 1469 year: 2020 ident: 10.1016/j.egyai.2022.100145_b22 article-title: Anomaly detection for wind turbines based on the reconstruction of condition parameters using stacked denoising autoencoders publication-title: Renew Energy doi: 10.1016/j.renene.2019.09.041 – volume: 161 start-page: 510 year: 2020 ident: 10.1016/j.egyai.2022.100145_b16 article-title: Spatio-temporal fusion neural network for multi-class fault diagnosis of wind turbines based on SCADA data publication-title: Renew Energy doi: 10.1016/j.renene.2020.06.154 – ident: 10.1016/j.egyai.2022.100145_b20 doi: 10.1145/3394486.3403118 – start-page: 1 year: 2017 ident: 10.1016/j.egyai.2022.100145_b34 article-title: Semi-supervised classification with graph convolutional networks – volume: 36 start-page: 1627 issue: 8 year: 1964 ident: 10.1016/j.egyai.2022.100145_b38 article-title: Smoothing and differentiation of data by simplified least squares procedures publication-title: Anal Chem doi: 10.1021/ac60214a047 – start-page: 9 year: 2020 ident: 10.1016/j.egyai.2022.100145_b12 article-title: Early fault detection based on wind turbine scada data using convolutional neural networks – volume: 9 start-page: 796 issue: 10 year: 2017 ident: 10.1016/j.egyai.2022.100145_b37 article-title: Interpolation in time series: An introductive overview of existing methods, their performance criteria and uncertainty assessment publication-title: Water doi: 10.3390/w9100796 – year: 2008 ident: 10.1016/j.egyai.2022.100145_b40 – start-page: 37 year: 2012 ident: 10.1016/j.egyai.2022.100145_b32 article-title: Autoencoders, unsupervised learning, and deep architectures – volume: PAS-100 start-page: 1176 issue: 3 year: 1981 ident: 10.1016/j.egyai.2022.100145_b2 article-title: Wind energy - a utility perspective publication-title: IEEE Trans Power Appar Syst doi: 10.1109/TPAS.1981.316586 – start-page: 1 year: 2020 ident: 10.1016/j.egyai.2022.100145_b27 article-title: Graph convolutional autoencoder model for the shape coding and cognition of buildings in maps publication-title: Int J Geogr Inf Sci – volume: 54686 start-page: 2071 year: 2011 ident: 10.1016/j.egyai.2022.100145_b6 article-title: Use of SCADA data for failure detection in wind turbines – year: 2021 ident: 10.1016/j.egyai.2022.100145_b14 article-title: A fault detection framework using recurrent neural networks for condition monitoring of wind turbines publication-title: Wind Energy doi: 10.1002/we.2628 – volume: 16 start-page: 5451 issue: 8 year: 2012 ident: 10.1016/j.egyai.2022.100145_b4 article-title: Renewable energy supply chains, performance, application barriers, and strategies for further development publication-title: Renew Sustain Energy Rev doi: 10.1016/j.rser.2012.06.006 – volume: 6 start-page: 1 issue: 1 year: 2019 ident: 10.1016/j.egyai.2022.100145_b28 article-title: Graph convolutional networks: a comprehensive review publication-title: Comput Soc Netw doi: 10.1186/s40649-019-0069-y – volume: 146 start-page: 760 year: 2020 ident: 10.1016/j.egyai.2022.100145_b17 article-title: Condition monitoring of wind turbines based on spatio-temporal fusion of SCADA data by convolutional neural networks and gated recurrent units publication-title: Renew Energy doi: 10.1016/j.renene.2019.07.033 – ident: 10.1016/j.egyai.2022.100145_b41 – year: 2016 ident: 10.1016/j.egyai.2022.100145_b31 |
| SSID | ssj0002511977 |
| Score | 2.443658 |
| Snippet | Wind power is one of the fastest-growing renewable energy sectors instrumental in the ongoing decarbonization process. However, wind turbines are subjected to... |
| SourceID | doaj crossref elsevier |
| SourceType | Open Website Enrichment Source Index Database Publisher |
| StartPage | 100145 |
| SubjectTerms | Condition monitoring Deep anomaly detection Early fault detection Graph Convolutional Autoencoder Multivariate Time series SCADA data Wind turbine |
| Title | Deep anomaly detection in horizontal axis wind turbines using Graph Convolutional Autoencoders for Multivariate Time series |
| URI | https://dx.doi.org/10.1016/j.egyai.2022.100145 https://doaj.org/article/24c38fb3bba94c47b4048e1cf1c518b2 |
| Volume | 8 |
| WOSCitedRecordID | wos001106796500003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2666-5468 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002511977 issn: 2666-5468 databaseCode: DOA dateStart: 20200101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2666-5468 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002511977 issn: 2666-5468 databaseCode: M~E dateStart: 20200101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Lj9MwELZQxQEOK57aLg_5wJGIxPEjOe6WFg5QcQCpt8ivlKyWpGrT7hYkfjszTlrlVC57ycFK7Ghm5PlG_uYzIe-kU3mWCxcJ5-OIc6eiPPcuYomWlnNhmQuNwl_UfJ4tFvm3wVVfyAnr5IE7w31g3KZZaVJjdM4tV4ZDzPnElokVSWbC7guoZ1BM4R7MwvEY9kpDApKR4DI7SA4Fcpdf7nUF1SFjQYQIm5kGaSmo9w-y0yDjzJ6Qsx4q0svuF5-SB75-Rh4PBASfkz8fvV9RXTe_9M2eOt8GYlVNq5r-bNbV7wZ7Ham-qzb0FopvCvnFIM-dIt19ST-hWjWdNPWuD0Bcbts2KG6JBGcKiJaGFt0dlNSASil2jFCMWr95QX7Mpt8nn6P-OoXI8oS3UClybpVOS-1L4WJWgoW0S71SUobzVy9zW2Yqh6LP2LhMTCIMGFExzSGt2_QlGdVN7c8JhSHAGcZkTCMiiME1MlaSy0SKVHA7JuxgzcL2WuN45cVNcSCVXRfBBQW6oOhcMCbvjx-tOqmN069foZuOr6JOdhiA6Cn66Cn-Fz1jIg9OLnrI0UEJmKo6tfrFfaz-ijzCKTv-5Gsyatdb_4Y8tLu22qzfhoiG59e_038NY_sy |
| 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=Deep+anomaly+detection+in+horizontal+axis+wind+turbines+using+Graph+Convolutional+Autoencoders+for+Multivariate+Time+series&rft.jtitle=Energy+and+AI&rft.au=Miele%2C+Eric+Stefan&rft.au=Bonacina%2C+Fabrizio&rft.au=Corsini%2C+Alessandro&rft.date=2022-05-01&rft.issn=2666-5468&rft.eissn=2666-5468&rft.volume=8&rft.spage=100145&rft_id=info:doi/10.1016%2Fj.egyai.2022.100145&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_egyai_2022_100145 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2666-5468&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2666-5468&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2666-5468&client=summon |