Loughborough University
Leicestershire, UK
LE11 3TU
+44 (0)1509 222222
Loughborough University

Centre for Renewable Energy Systems Technology (CREST)

Wind and Water Power

Condition Monitoring

Condition Monitoring of Wind Turbine Drive Trains

Image showing Continuous Wavelet Transform (CWT) contour plot showing

key characteristic vibration frequencies superimposed on inferred vibration

level vs. generator frequency measured using a current sensor for a small

25kW wind turbine.

Unplanned shutdowns reduce wind turbine availability and therefore energy yield adding significantly to the cost of energy, particularly in the offshore environment (van Bussel and Schöntag, 1997). They often also result in costly deployment of equipment, e.g. ships and jack-up barges for maintenance. Some parts of a turbine, for example the gearbox require a lot of time and equipment to replace. Gearbox failures are often caused by bearing wear, in turn caused by shaft misalignment.

Continuous condition monitoring of wind turbines has the potential to catch problems early, enable preventative maintenance and thereby reduce turbine downtime and prevent secondary mechanical damage. Gearbox and bearing failures are responsible for significant percentages of turbine failure and downtime (Tavner et al., 2007). Accelerometers applied to mechanical components of the drive train are traditionally used for condition monitoring but require their own data acquisition system and analysis software. In contrast, the electrical current and voltage data is continuously measured and could also be used for condition monitoring more cheaply. Monitoring of electrical power has been shown to be useful for the detection of generator shaft misalignment leading to bearing failure (Watson et al., 2010).

An experimental data acquisition system has been installed on a small (25kW) onshore turbine in Leicestershire UK to compare 6 electrical power signals with 6 accelerometer signals. Temperature data was also recorded in the same locations. Data have been recorded over several months including periods before and after a gearbox failure and replacement. Data were recorded at a frequency of 4kHz and analysed using both Fourier Transform and Morlet Continuous Wavelet Transform methods. Results show that the electrical data do indeed show the same principal vibration frequencies as the accelerometers and could therefore be used for condition monitoring.

References

Van Bussel, G. J. and C Schöntag C., 1997. ‘Operation and maintenance aspects of large offshore windfarms.’ Proceedings of the European Wind Energy Conference, Dublin, Ireland, October 6-9, 1997, pp 272-275.

Tavner, P.J., Xiang, J. and Spinato F., 2007. ‘Reliability analysis for wind turbines’. Wind Energy 10, pp 1-18.

Watson, S.J., Xiang, B.J., Yang, W., Tavner, P.J. and Crabtree, C.J., 2010. ‘Condition monitoring of the power output of wind turbine generators using wavelets’. IEEE Transactions on Energy Conversion 25, pp 715-721.

 

 

Condition Monitoring of Wind Turbine Blades

Image showing the experimental set-up for the accelerometer based testing of a

wind turbine blade.

Wind turbine blades are structurally composed of different types of materials including: wood and/or steel used for constructing the main-spar and glass reinforced plastic (GRP) for building the downwind and upwind sides of the blade. During manufacture and over the operation lifetime of wind turbine blades, they are subject to defects and degradation which can result in deformation/fault occurrence (such as cracks) and subsequent failure. The wide-spread of such materials used in their construction and the very variable loading to which blades are subject in operation makes it difficult for predictions on remaining lifetime to be made, thus the importance of condition monitoring (Borum et al., 2006, McGowan et al., 2007, Esu et al., 2013a).

CREST in conjunction with colleagues in the Communications Division of the School of Electronic, Electrical and Systems Engineering have been researching into a vibration-based compact condition monitoring system, comprised of low-cost consumer electronics capable of detecting variations in the dynamic properties of a turbine. Such a system could potentially be used to detect changes in the material properties of the blade as well as changes in loading (Esu et al., 2013b). This has potential for aiding advances load control and the detection of icing as well as blade condition monitoring.

At marginal cost, the in situ condition monitoring system could be integrated easily either retrofitted externally or embedded in turbine blades at manufacture. Research work has been done through the testing of a system on a 4.5m long Carter wind turbine blade under different loading conditions. Early results suggest that the system can satisfactorily characterise the bending modes of a blade and could potentially be used to detect changes in blade properties.

References

Borum, K.K., McGugan, M., and Brøndsted, P., “Condition monitoring of wind turbine blades,” in 27th Risø International Symposium on Materials Science: Polymer Composite Materials for Wind Power Turbines, 2006, pp. 139–145.

McGowan, J.G., Hyers, R.W, Sullivan, K.L. Manwell, J.F. Nair, S.V., McNiff, B. and Syrett, B.C., “A review of materials degradation in utility scale wind turbines,” Energy Mater. Mater. Sci. Eng. Energy Syst., vol. 2, no. 1, pp. 41–64, Mar. 2007.

Esu, O.O, Flint, J.A. and Watson, S.J., “Integration of Low-cost Accelerometers for Condition Monitoring of Wind Turbine Blades,” in European Wind Energy Conference, Vienna, 2013, pp. 1–4.

Esu, O.O, Flint, J.A and Watson, S.J., “Condition Monitoring of Wind Turbine Blades Using MEMS Accelerometers,” in Renewable Energy World Europe, 2013, pp. 1–12.