| In order to realize the goal of net zero emission,our country vigorously develop the wind power industry,at present,it has become the world’s largest newly installed wind power.The blades are an important part of the wind turbine that converts wind energy into mechanical energy and withstand complex alternating loads,the random change of wind direction leads to the uncertainty of wind turbine blade load,which will reduce the operation efficiency of wind turbine and is unfavorable not conducive to the smooth and efficient operation of wind turbine.Therefore,this topic takes horizontal axis wind turbine blade as the research object,collects the strain signal of wind turbine blade under continuous crosswind operation condition through wind tunnel experiment.Aiming to optimize the noise reduction method and improve the reliability of blade strain signal analysis,combined with non-stationary signal noise reduction method,the noise reduction and analysis of wind turbine blade strain signal were studied.The main research contents include the following three aspects:(1)In view of the randomness of the natural wind,which is difficult to simulate in the wind tunnel experiment,the wind direction change is simplified reasonably,the dynamic yaw test bench is used to simulate the wind direction change in the horizontal direction.Five strain gauge measuring points were arranged on the broadwise aerodynamic center line of the blade.By adjusting the load resistance and dynamic yaw test bench,the strain data of wind turbine blades under continuous crosswind conditions with different tip velocity ratios and crosswind angular velocities were measured.(2)Aiming at the problem that the strain signal of wind turbine blade is non-stationary and doped with a lot of high-frequency noise interference,the combined method of Ensemble Empirical Mode Decomposition-Composite Multiscale Permutation Entropy and Wavelet threshold denoising is adopted to solve the problem of insufficient denoising of Ensemble Empirical Mode Decomposition method and achieve effective noise filtering.The simulation signal was constructed according to the non-stationary characteristics of the measured strain signal,the joint noise reduction method was used to denoise the simulation signal.The feasibility of the joint method to denoise the non-stationary signal was verified by evaluating the signal-to-noise ratio(SNR),root-mean-square error(RMSE),smoothness and determination coefficient.The joint noise reduction method is used to denoise the measured strain signal.The SNR and RMSE are used to evaluate the denoising performance.The frequency of the strain signal is fitted with the rotating frequency of the measured wind turbine,the relative error of the two frequencies is calculated,the accuracy of the joint noise reduction method to the measured strain signal is verified.(3)Explore the variation law of blade strain under continuous crosswind conditions:the strain distribution law of five positions along the aerodynamic center line of the blade is analyzed under different tip velocity ratio and crosswind angular velocity.The five positions are 0.20 R,0.35 R,0.50 R,0.65 R and 0.80 R.It is found that the strain from blade root to tip shows a decreasing trend,the strain amplitude decreases rapidly at 0.50R～0.65 R.The strain characteristics of wind turbine blade root position under different tip velocity ratio and crosswind angular velocity were analyzed.Breaking the limitation of only qualitative analysis of blade strain,the wind turbine blade strain signal is quantitatively analyzed,the research found that the strain at blade root position appeared hysteresis in the range from 0° to 10° crosswind angle under different conditions,the strain decreased by less than 10%.The strain signal processed by the joint noise reduction method can accurately show the relationship between strain frequency and amplitude with crosswind conditions,which lays a solid foundation for the characteristic analysis of wind turbine blade strain under continuous crosswind conditions. |
PDF Full Text Request