Study On Identification And Prediction Of Aircraft Wake Vortex Using Doppler Lidar

The aircraft wake turbulence is inevitable product of the lift.It forms a pair of vortices with opposite rotation directions behind the wings,and can exist for a relatively stable period of time,which has an impact on the safety of civil aviation.The wake turbulence separation standards currently in use have gradually failed to meet the development of the civil aviation industry.Both domestic and international are studying how to optimize the wake turbulence separation and the dynamic wake turbulence separation,which raises the demand for the identification and prediction of aircraft wake vortices.This paper focuses on the identification and prediction of aircraft wake vortices,researches the characteristics of the radial velocity field of lidar detection of aircraft wake vortices,and combines environmental parameters to study the identification and prediction methods.The main research work of the paper includes:First,the method of lidar detection of aircraft wake vortices is studied.By analyzing the detection principle of lidar and combining the aircraft's wake vortex field area,the effects of different detection modes on the aircraft's wake vortices detection are analyzed.After determining that the range height indicator scanning mode has the best effect on the aircraft's wake vortices detection,The model of the conversion of the tangential velocity of the aircraft's wake vortex and the radial velocity of the lidar detection in this mode is established,and the lidar placement index is proposed.Through simulation calculation,the optimal placement position range of the lidar is determined.The effectiveness of the method was confirmed through field detection experiments,and data collection was completed.Secondly,according to the characteristics of the aircraft wake vortex,the characteristics of the aircraft vortex are extracted from the data of the lidar collected on the field experiment,and the environmental parameters are fused as the data set,and a method of aircraft vortex identification based on Support Vector Machines is proposed.Through the method of feature scaling,grid search and cross-validation,the best parameters of the model are determined.The test results show that this method can effectively identify the aircraft wake vortex.Then,the aircraft wake vortex characteristic parameters are further extracted from the radial velocity field with wake vortices.According to the feature changes between adjacent scan periods,an aircraft wake vortex prediction method based on XGBoost is proposed.The optimal parameters of the model are determined by a step-by-step search method.The test results show that the method has a good predictive ability for the aircraft wake vortex,and the importance of different features for the prediction results are analyzed an effectively predict the aircraft wake vortex and testing.The test results show that the method can effectively predict the aircraft wake.Finally,on the basis of the proposed aircraft wake vortex identification and prediction method,the aircraft wake vortex identification and prediction system are established and developed,and the historical data is used to verify that the system functions properly.The system can identify and predict aircraft wake vortices accurately and give corresponding prompts to user.The lidar aircraft vortex detection,identification and prediction methods proposed in this paper provide support for optimized wake turbulence separation and dynamic wake turbulence separation and are of great significance and application value for improving flight safety factors and airport capacity and efficiency.