RÜZGAR ÇİFTLİKLERİNDE KALAN FAYDALI ÖMÜR TAHMİNİ

Year 2021, Volume: 5 Issue: 2, 145 - 154, 31.08.2021

Barış Bulut

https://doi.org/10.46519/ij3dptdi.922599

Cited By: 1

Abstract

Dünya çapında sayısı artmakta olan rüzgar enerji santrallerinin (RES) işletme ve bakım maliyetleri, üretimin karlılığını belirleyen önemli bir kalemdir. Kestirimci bakım yöntemleriyle rüzgar türbinlerinin güvenli çalışma süreleri uzatılabilmesinin yanında, işletme ve bakım maliyetleri de düşürülebilir. Çalışmamızda, rüzgar türbini ve bileşenlerine dair bir kestirimci bakım yöntemi sunılmaktadır. Kalan faydalı ömür (RUL) yaklaşımıyla, arızaların ne kadar süre sonra gerçekleşeceği ve arızaların hangi alt sistemde meydana geleceği LSTM (Long Short-Term Memory) gibi sıralı derin öğrenme yöntemleriyle tahmin edilebilmektedir. Önerilen çözümde, faaliyetteki bir rüzgar çiftliği bünyesindeki türbinler değerlendirilmektedir.

Keywords

Rüzgar Enerji Santrali (RES), Kalan Kullanışlı Ömür, LSTM

Supporting Institution

TÜBİTAK-TEYDEB

Project Number

9180070

Thanks

Buradaki çalışmaların bir kısmı “Siber Fiziksel Sistemlerde Akıllı Kestirimci Bakım (SMART-PDM)” başlıklı ve 17041 numaralı uluslararası EUREKA – ITEA projesi (https://smart-pdm.eu) kapsamında, TEYDEB 1509 - Uluslararası Sanayi Ar-Ge Projeleri programı bünyesinde 9180070 numaralı proje altında TÜBİTAK tarafından desteklenmiştir.

ESTIMATING REMAINING USEFUL LIFE IN WIND POWER PLANTS

Abstract

The operation and maintenance cost of wind power plants (WPP) whose numbers are increasing worldwide, is an important item that determines the energy prices. Safe operating periods of wind turbines can be extended, operation and maintenance costs can be reduced with predictive maintenance methods. In our study, a predictive maintenance method considering the wind turbine and its components is presented. With the remaining useful life (RUL) approach, the time before the next failure for a particular subsystem can be predicted using sequential deep learning methods such as LSTM (Long Short-Term Memory). In the proposed solution, turbines within an operational wind farm are considered.

Keywords

Wind Power Plant (WPP), Remaining Useful Life (RUL), LSTM

Project Number

9180070

References

• 1. J. M. Pinar Pérez, F. P. García Márquez, A. Tobias ve M. Papaelias, «Wind turbine reliability analysis,» Renewable and Sustainable Energy Reviews, Vol. 23, Sayfa 463-472, 2013.
• 2. B. Lu, Y. Li, X. Wu ve Z. Yang, «A review of recent advances in wind turbine condition monitoring and fault diagnosis,» %1 içinde 2009 IEEE Power Electronics and Machines in Wind Applications (PEMWA), Lincoln, NE, USA, 2009.
• 3. W. Qiao ve D. Lu, «A survey on wind turbine condition monitoring and fault diagnosis-part I: Components and subsystems,» IEEE transactions on industrial electronics (1982), Vol. 62, Issue 10, Sayfa 6536-6545, 2015.
• 4. J. Ribrant ve L. M. Bertling, «Survey of failures in wind power systems with focus on Swedish wind power plants during 1997–2005,» IEEE transactions on energy conversion, Vol. 22, Issue 1, Sayfa 167-173, 2007.
• 5. W. Yang ve J. Jiang, «Wind turbine condition monitoring and reliability analysis by SCADA information,» %1 içinde 2011 Second International Conference on Mechanic Automation and Control Engineering, Inner Mongolia, China, 2011. 6. Y. Zhao, D. Li, A. Dong, D. Kang, Q. Lv ve L. Shang, «Fault prediction and diagnosis of wind turbine generators using SCADA data,» Energies, Vol. 10, Issue 8, p. 1210, 2017.
• 7. F. Cheng, L. Qu ve W. Qiao, «Fault prognosis and remaining useful life prediction of wind turbine gearboxes using current signal analysis,» IEEE transactions on sustainable energy, Vol. 9, Issue 1, Sayfa 157-167, 2018.
• 8. M. Nie ve L. Wang, «Review of condition monitoring and fault diagnosis technologies for wind turbine gearbox,» Procedia CIRP, Vol. 11, Pages 287-290, 2013.
• 9. A. Kusiak ve A. Verma, «A data-mining approach to monitoring wind turbines,» IEEE transactions on sustainable energy, Vol. 3, Issue 1, Pages 150-157, 2012.
• 10. X. Liu, S. Lu, Y. Ren ve Z. Wu, «Wind turbine anomaly detection based on SCADA data mining,» Electronics, Vol. 9, Issue 5, Pages 751, 2020.
• 11. M. Dwarampudi ve N. V. S. Reddy, «Effects of padding on LSTMs and CNNs,» 2019.
• 12. S. Hochreiter ve J. Schmidhuber, «Long short-term memory,» Neural computation, Vol. 9, Issue 8, Pages 1735-1780, 1997.
• 13. M. Schuster ve K. K. Paliwal, «Bidirectional recurrent neural networks,» IEEE transactions on signal processing: a publication of the IEEE Signal Processing Society, Vol. 45, Issue 11, Pages 2673-2681, 1997.
• 14. F. Karim, S. Majumdar, H. Darabi ve S. Harford, «Multivariate LSTM-FCNs for time series classification,» Neural networks: the official journal of the International Neural Network Society, Vol. 116, Pages 237-245, 2019.
• 15. I. Alsyouf, «Wind energy system reliability and maintainability, and operation and maintenance strategies,» %1 içinde Wind Energy Systems, Elsevier, 2011, Pages 303-328.
• 16. G. Sullivan, R. Pugh, A. P. Melendez ve W. D. Hunt, «Operations & maintenance best practices - A guide to achieving operational efficiency (release 3),» Office of Scientific and Technical Information (OSTI), 2010.
• 17. Cpm engineering, 70% operate a ‘run to fail’ strategy, https://cpm-uk.com/2015/09/70-operate-a-run-to-fail-strategy/, September 15, 2015.
• 18. M. Jenvald ve M. Hovmöller, «Reducing Delays for Unplanned Maintenance of Service Parts in MRO Workshops: A case study at an aerospace and defence company,» 2020.
• 19. K. Holmberg, A. Adgar, A. Arnaiz, E. Jantunen, J. Mascolo ve S. Mekid, %1 içinde E-maintenance, London, Springer London, 2010.