The Study Of Wind Shear Warning Algorithm And Type Recognition Based On Coherent Lidar

Low-level wind shear is an atmospheric phenomenon indicating that the wind speed or direction changes rapidly within a short time and small scale range in the height below600 m.It seriously affects the takeoff,landing and flight safety of aircraft.Coherent lidar can accurately measure atmospheric wind field in clear sky.Therefore it is an effective tool for detecting wind shear in civil aviation airports.Wind shear detection and wind shear type recognition are studied in the work,and the wind shear warning algorithm and type recognition algorithm of coherent lidar are designed.Firstly,the possible errors of traditional wind shear warning algorithm are analyzed,and the wind shear warning algorithm based on principal component analysis and phase difference correction method is proposed.The validity and feasibility of the algorithm are evaluated by using the experimental data of rotating motor and small aircraft respectively.The results show that the mean absolute error of the peak frequency of coherent lidar echo signal estimated by this algorithm is 0.26 MHz,and the warning rate of actual low-level wind shear is 92.31%.This algorithm can effectively reduce the frequency estimation error of coherent lidar echo signal and realize low-level wind shear warning.Then,considering the problem of the image types of low-level wind shear signal measured by coherent lidar are few and difficult to obtain,the fluid calculation software is used to simulate the three-dimensional wind field with low-level wind shear in the work.And then the wind shear signal images are obtained by using the radial velocity data obtained by coherent lidar scanning,which expands the image type of low-level wind shear signal.Finally,in order to effectively extract the image features of low-level wind shear signal of lidar,a multilayer feature extraction and adaptive fusion based on deep convolutional neural network is proposed.This algorithm is used to extract the image features from the data of the simulated low-level wind shear signal image sample library.The type recognition experiments are carried out by support vector machine.The results show that the average recognition rate of low-level wind shear image type recognition using thisalgorithm is 98.1%,and the average recognition time is 0.29 s.This algorithm can effectively extract image features and realize low-level wind shear type recognition.