| As one of the important components of wind turbine,blade plays an important role in wind energy capture.It bears the role of periodic load and extreme load in operation.For a long time,blade may be damaged in different degrees.At the same time,once suffering from bad weather,it is very easy to aggravate the damage of wind turbine blade,resulting in bad safety accidents.If the damage of blade can be detected in time,the safety accident can be avoided to a great extent.By online monitoring the changes of blade modal parameters,blade damage can be detected in time,which has important research significance and engineering practical value.According to the structure and operation characteristics of wind turbine blade and the actual situation in the experiment,an identification method of working modal parameters,enhanced power spectrum transmissibility method,which is suitable for blade modal analysis,is studied.Moreover,a small blade test-bed is built in the laboratory to carry out blade vibration experiments.The modal analysis method based on the transmissivity is used for blade modal analysis,and compared with the traditional experimental modal analysis method.The experimental results verify the accuracy of the method.Then,the wind turbine simulation software fast developed by the National Renewable Energy Laboratory(NREL)is used to simulate the blade damage.Through the analysis of the simulation results of various blade damage degrees,the influence law of blade damage on modal parameters is found.Finally,based on the unit operation data and modal identification parameters,the support vector machine classification algorithm is applied to realize the blade damage fault diagnosis.The diagnosis results show the effectiveness of the blade damage diagnosis method. |