Automatic Identification of Modal Parameters of Wind Turbine Towers Under Harmonic Excitation
The periodic excitation of wind turbines under operating conditions will generate harmonic disturbances with similar frequencies to the structural modes,which will affect the vibration level of the fundamental modes of the structure and the identification of the dynamic parameters.In order to effectively and continuously monitor the tower vibration status during operation,the covariance-driven stochastic subspace identification(Cov-SSI)method based on potential hierarchical agglomerative clustering(PHA)combined with the probability density function(PDF)was proposed for the automatic identification of modal parameters of wind turbine towers.Through the on-site vibration response test,the Cov-SSI method was firstly used to initially identify the tower structure modal parameters;secondly,the PHA method was introduced to improve the stability diagram,and the frequency and modal confidence criterion(MAC)distance matrix was defined for cleaning and clustering to automate the separation of different orders of modes;finally,the information of the clustered clusters was used to determine and eliminate the harmonic modes by the PDF method.The results showed that the proposed method could effectively separate and eliminate the harmonic components,realize the automatic identification of the modal parameters of wind turbine towers under operation,and provide a good engineering application value for the automated real-time monitoring of wind turbine safety operation.
NETL
NSTL
万方数据
维普
