Vibration Monitoring And Early Warning Method For Wind Turbine Structure

Recently,new energy is the world’s development and research hot spots.With the continuous development of wind power,the number of wind turbines has gradually increased.The wind turbines are developed towards large capacity,which the capacity and height of the wind turbines tower are on the rise.The traditional monitoring of wind turbines is mainly focused on themonitoring of mechanical parts of wind turbines,while ignoring the structure monitoring.However,most of the accidents have been caused by the collapse of the tower.As the main loaded structure of the entire wind turbine,the safety of wind turbine tower is important.This paper studies the monitoring methods of dynamic characteristics of wind turbines and the early warning methods.Main contents are included as follows:A set of structural health monitoring system is designed for the structure of 1.5MW large-scale wind turbine.The components of the whole monitoring system and the specific equipment are introduced.The working principle of the system is explained.At the same time,the layout of the whole system is designed considering the situation of the wind turbines.Finally,the implementation of the monitoring system is carried out to a wind turbine as an example.The data collected by the monitoring system is analyzed by the power spectrum method and the stochastic subspace identification method.The relationship between the modal parameters of wind turbine structure and wind speed,temperature,wind angle,blade speed and yaw angle is investigated.At the same time,an finite element model of the wind turbine structure is established by using ANSYS,and the modal analysis is carried out to verify the accuracy of the identified modal parameters from the measured data by monitoring system.An early warning method of the tower-based on time series analysis and statistical process control is proposed.First,the original acceleration data is processed by the Butterworth filter,and the processed data is integrated to obtain the displacement data.Then data prediction is carried out by using the autoregressive model,and early warning of the displacement is done by using the model prediction error combined with the statistical process control.Finally,the data of the wind turbine in each wind speed are analyzed and the validity of the method is verified.