Monitor Method of Offshore Wind Turbine Based on Memory-Like and Neural Network Expert System

Abstract:

The number of offshore wind farms increases gradually because of the high capability of power generation. However, the costs of manufacturing, logistics, installation and maintenance of offshore wind turbine are higher than those of onshore wind turbine. Thus the introduction of fault diagnosis is considered as a suitable way to improve reliability of wind turbine and reduce costs of repairs and casualties. In this paper, 3 major failures of direct-driven wind turbine according to urgency and system responses are discussed. A "memory-like" model pretreatment method and a fault diagnosis method for the failures are investigated. The simulation results show that total amount of fault data to be processed and stored is reduced, and difficulties of knowledge gaining and fault reasoning are also decreased.

• Periodical: Applied Mechanics and Materials (Volumes 635-637)
• Pages: 687-693
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.635-637.687
• Online since: September 2014
• Authors: Ling Xia Su*, Xia Xia Ma
• Keywords: Directly Driven Generators, Expert System, Fault Diagnosis, Memory-Like Model, Neural Network (NN), Offshore Wind Turbine

References

[1] European Wind Energy Association (EWEA), Offshore Development, Wind Energy - The Facts, Routledge, London, Page 212, (2009).

[2] D. Milborrow, Operation and Maintenance Costs Compared and Revealed, Wind Stats, Vol. 19, No. 3, (2006).

[3] Caithness Windfarm Information Forum, Summary of Wind Turbine Accident Data to 31 December 2011, Jan, (2012).

[4] S. Stubkier and E. Dyrbye, AVN Energy Designs a New Hydraulic Yaw System and Reduces the Fatigue Loads of the Wind Turbine, Modem Energy Review, Vol 3, Issue 2, Page 45, (2011).

[5] R. Kithil, Case Study of Lightning Damage to Wind Turbine Blade, National Lighting Safety Institute (NLSI), June (2008).

[6] R.C. Atkinson, R.M. Shiffrin, Human Memory: A Proposed System and Its Control Processes, (1968).

[7] P. Li, Y.D. Song, et al. Fault Monitoring Scheme for Large-scale Offshore Wind Turbines [J]. Power System Protection and Control. 2012, 40(1): 43-49. In Chinese.

[8] B.C.P. Lau, E.W.M. Ma, and M. Pecht. Review of Offshore Wind Turbine Failures and Fault Prognostic Methods. Prognostics and System Health Management (PHM), 2012 IEEE Conference on. IEEE, (2012).

DOI: 10.1109/phm.2012.6228954

[9] B. Kim, J-Y. Park, et al. Design and Application of Condition Monitoring System for Wind Turbines. Control, Automation and Systems (ICCAS), 2011, 11th International Conference. IEEE, (2011).

[10] A. Kusiak, and A. Verma. A Data-mining Approach to Monitoring Wind Turbines. Sustainable Energy, IEEE Transactions on 3. 1 (2012): 150-157.

DOI: 10.1109/tste.2011.2163177

[11] C. J. Crabtree, and P. J. Tavner. Condition Monitoring Algorithm Suitable for Wind Turbine Use. Renewable Power Generation (RPG 2011), IET Conference. IET, (2011).

DOI: 10.1049/cp.2011.0163

[12] F. Ding. Neural Network Expert System [M]. Beijing: Science Press, 2006. In Chinese.