Condition Monitoring and Fault Diagnosis of a Wind Turbine with a Synchronous Generator using Wavelet Transforms Wenxian Yang 1 , P. J. Tavner 1 , Michael Wilkinson 2 1 New & Renewable Energy Group, School of Engineering, Durham University, Durham DH1 4RL, UK 2 Garrad Hassan & Partners Ltd, St Vincent Works, Bristol, BS2 0QD Keywords: wind turbine, condition monitoring, fault diagnosis, wavelet transforms. Abstract Some large wind turbines use a low speed synchronous generator, directly-coupled to the turbine, and a fully rated converter to transform power from the turbine to mains electricity. This paper considers the condition monitoring and diagnosis of mechanical and electrical faults in such a variable speed machine. The application of wavelet transforms is investigated because of the disadvantages of conventional spectral techniques in processing instantaneous information in turbine signals derived from the wind, which is variable and noisy. A new condition monitoring technique is proposed which removes the negative influence of variable wind in machine condition monitoring. The technique has a versatile function to detect mechanical and electrical faults in the wind turbine. Its effectiveness is validated by experiments on a wind turbine condition monitoring test rig using a permanent-magnet synchronous generator, which can be driven by aerodynamic forces from a drive motor controlled by an external model, representing wind and turbine rotor behaviour. Within the technique wavelet transforms are employed for noise cancellation and are extended to diagnose faults by taking advantage of their powerful capabilities in analysing non-stationary signals. The diagnosis of wind turbine rotor imbalance in the will be used as an illustrative example, heralding the possibility of detecting a wind turbine mechanical faults by power signal analysis. 1 Introduction The development of wind turbine technology has benefited from the government decisions favourable to ‘green’ or renewable power. Wind turbines are becoming economically a viable alternative to conventional fossil-fuelled power generation. In some countries, notably Germany and Denmark, wind turbines have been playing a vital role in the power network, although not without some problems [1]. However, wind turbines do experience failures [2], due to their variable load condition and aggressive operating environment, however, turbines are beginning to show a reliability that is better than other forms of power generation, for example diesel generators. So developing economic condition monitoring and fault diagnosis techniques for them would be highly desirable and this will be especially important if they are deployed offshore. SCADA techniques are being applied widely to wind turbines but the data rate, once every 5-10mins, is too slow for most rotating machine fault diagnosis. There are many techniques developed in electric power production, aerospace, marine propulsion, and other process industries that could be applied to wind turbines [3]. However, the results to date have not proved satisfactory, due to the peculiarities of the wind turbine, that is slow and variable speed, at least for the larger types. In recent years, some efforts have been made to improve this situation [4]. However, the majority of wind turbine condition monitoring and fault diagnosis techniques proposed have used the Fourier Transform (FT), which is less capable of solving the problem due to its shortcomings in dealing with non-stationary signals. In view of this, the potential application of the wavelet transform to the condition monitoring and fault diagnosis of wind turbines is investigated in this paper as an extension of the work described in [6].The Discrete Wavelet Transform (DWT) is used for noise cancellation as the signals from the wind turbine contain noise which is difficult to remove by using a conventional filter with fixed cut-off frequencies. To be presented at IET International Conference, PEMD, York, April 2008