Damage Characteristics Analysis Of Wind Turbine Blade

Blade is an important part of the wind turbine, which is an intermediary to convert wind power into electricity. During the turbine’s normal operation, the complicated external environment will threat the wind energy conversion system seriously, in which the wind is one of the most serious. When the wind scale is too high or the blades have been running in a constant fatigue, the blade will probably break off, which may strike the tower, and result in chain damages. Therefore, this paper mainly studies on the damage identification of the wind turbine blades, to put forward a new health monitoring method based on wavelet analysis and independent component analysis, the details are as follows:Firstly, PZT(Piezoelectric Ceramic Transducer), as the sensing element, has the advantages of small size, low weight, good piezoelectric property, small acoustical impedance, broad frequency response, etc. It is widely used in the vibration measurement. In order to improve the accuracy of the blade vibration data, the study adopts PZT sensor, as the sensing element of the blade vibration monitoring system, introducing the PZT sensor working principle, to test the performance through the changing of the actuation frequency. The experimental results show that the sensitivity of PZT sensor is valid for the study.Secondly, the study introduced the wavelet analysis theory to remove noise and other interferences, with detailed introduction of the wavelet analysis theory, the basic working principle of the multi-resolution analysis and wavelet thresholding denoising, and also putting forward a new thresholding denoising method.Then the free vibration simulation data can be got. Mix some noising signals into the simulated data, and then make a comparison with the traditional method and the new one, then it isobvious to see the signal-noise ratio has increased, which provides theoretical basis for the observation and analysis of the blade vibration.Thirdly, in order to effectively extract the blade damage identification signals, the study applied independent component analysis theory to extract the independent component collected by the monitoring system.The study introduced the basic algorithms of the independent component theory, and the procedure of the algorithms. Meanwhile, the study collected the healthy blade vibration signals to simulate, and to choose the components that can represent the periodic signals of the blade vibration,and then to conclude the valid of the theory.Fourthly, with the help of the wavelet analysis and the independent component analysis, the singular point appeared in the vibration can be observed. The damage part can be predicted through the analysis of the vibration frequency, distributed in the whole blade collected by experimental data.Finally, extract the independent component on different working conditions, like the healthy blade, blade with surface composite materials dropping, surface crack, or tip crack. When testing the samples, the correlation coefficient of the independent component can be measured on these four different working conditions. To be more accurate, the total feature of the sample should be gained, as well as the absolute number of the correlation coefficients. And then the damage type can be diagnosed easily with the identification of the sorter.The methods put forward in the study are very effective for settling damage identification problems, which is very significant for the safety operation of the wind turbine blade.