Estudo sobre Modelos de Aprendizado de Máquina para Detecção de Falhas em Turbinas Eólicas
DOI:
https://doi.org/10.5540/03.2023.010.01.0048Palavras-chave:
Energia Eólica, Aprendizado de MáquinaResumo
A crescente busca por soluções energéticas renováveis tem trazido destaque para soluções como turbinas eólicas, que são as principais responsáveis pela transformação de energia eólica em elétrica. Assim, o monitoramento, diagnóstico e prognóstico de falhas destas turbinas é fundamental para garantir a produção energética de forma contínua. Estas turbinas são monitorados por sensores e os dados oriundos deste monitoramento podem ser utilizados para criar modelos capazes de detectar estágios iniciais de degradação dos componentes que formam as turbinas, garantindo assim que falhas sejam identificadas rapidamente, reduzindo custos em manutenção. No presente trabalho é apresentada uma breve revisão sobre o assunto, além da aplicação de duas técnicas de aprendizado de máquina em uma base de dados real.
Downloads
Referências
Yaser S Abu-Mostafa, Malik Magdon-Ismail e Hsuan-Tien Lin. Learning From Data. Vol. 4. AMLBook, 2012.
James Carroll et al. “Wind turbine gearbox failure and remaining useful life prediction using machine learning techniques”. Em: Wind Energy 22.3 (2019), pp. 360–375.
Daniel Chan e John Mo. “Life Cycle Reliability and Maintenance Analyses of Wind Turbines”. Em: Energy Procedia 110.December 2016 (2017), pp. 328–333.
EDP - Open Data. https://opendata.edp.com/pages/homepage/, last accessed on 15/08/21.
A. Géron. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, 2019.
Jamie Godwin e Peter Matthews. “Classification and Detection of Wind Turbine Pitch Faults Through SCADA Data Analysis”. Em: International Journal of Prognostics and Health Management 4 (jan. de 2013).
GWEC. “Global Wind Report 2021”. Em: Global Wind Energy Council (2021).
Georg Helbing e Matthias Ritter. “Deep Learning for fault detection in wind turbines”. Em: Renewable and Sustainable Energy Reviews 98.September (2018), pp. 189–198.
Gareth James et al. An Introduction to Statistical Learning: with Applications in R. Springer, 2013.
A Joshuva. e V Sugumaran. “A data driven approach for condition monitoring of wind turbine blade using vibration signals through best-first tree algorithm and functional trees algorithm: A comparative study”. Em: ISA Transactions 67 (2017), pp. 160–172.
Xiaoyuan Liu et al. “Wind turbine anomaly detection based on SCADA data mining”. Em: Electronics (Switzerland) 9.5 (2020).
Jorge Maldonado-Correa et al. “Using SCADA data for wind turbine condition monitoring: A systematic literature review”. Em: Energies 13.12 (2020).
Mateus Mendes et al. “Wind Farm and Resource Datasets: A Comprehensive Survey and Overview”. Em: Energies 13 (set. de 2020).
Rafael Orozco, Shawn Sheng e Caleb Phillips. “Diagnostic Models for Wind Turbine Gearbox Components Using SCADA Time Series Data”. Em: jun. de 2018, pp. 1–9.
Taylor Regan, Christopher Beale e Murat Inalpolat. “Wind Turbine Blade Damage Detection Using Supervised Machine Learning Algorithms”. Em: Journal of Vibration and Acoustics 139.6 (ago. de 2017).
Bernhard Schölkopf et al. “Estimating the support of a high-dimensional distribution”. Em: Neural Computation 13.7 (2001), pp. 1443–1471.
Adrian Stetco et al. “Machine learning methods for wind turbine condition monitoring: A review”. Em: Renewable Energy 133 (2019), pp. 620–635.
Bernhard Strack et al. “Ontology for maintenance of onshore wind turbines”. Em: Forschung im Ingenieurwesen/Engineering Research 85.2 (2021), pp. 265–272.
Peng Sun et al. “A generalized model for wind turbine anomaly identification based on SCADA data”. Em: Applied Energy 168 (2016), pp. 550–567.
Jannis Tautz-Weinert e Simon J. Watson. “Using SCADA data for wind turbine condition monitoring - A review”. Em: IET Renewable Power Generation 11.4 (2017), pp. 382–394.
Pierre Tchakoua et al. “Wind turbine condition monitoring: State-of-the-art review, new trends, and future challenges”. Em: Energies 7.4 (2014), pp. 2595–2630.
Long Wang et al. “Wind Turbine Gearbox Failure Identification With Deep Neural Networks”. Em: IEEE Transactions on Industrial Informatics 13.3 (2017), pp. 1360–1368.
Wenna Zhang e Xiandong Ma. “Simultaneous Fault Detection and Sensor Selection for Condition Monitoring of Wind Turbines”. Em: Energies 9.4 (2016).
Hongshan Zhao et al. “Anomaly detection and fault analysis of wind turbine components based on deep learning network”. Em: Renewable Energy 127 (2018), pp. 825–834.
