Research On The Influence Of Terrain Conditions On Aircraft Turbulence

Safety is a serious problem of civil aviation transportation,and the turbulence of aircraft seriously affects the safety and comfort of flight.When the aircraft is flying near the ground where the altitude is higher and the terrain changes greatly,the aircraft turbulence is very frequently.Therefore,it is necessary to combine the specific characteristics of the terrain to study the influence of terrain conditions on the aircraft over larger terrains.Firstly,This paper combines the areas with dense aircraft turbulence shown by AMDAR data and the topographic and climate characteristics of South Africa,then selected three regions with obvious characteristics in South Africa.Secondly,rasterized the terrain DEM data,the ECMWF reanalysis data and the aircraft turbulence data,analyzed the temporal and spatial distribution characteristics of the aircraft turbulence over the three terrains,and the physical characteristics of the turbulence areas.Finally,the five terrain-related parameters of the rasterization are substituted into the support vector machine model to obtain the four parameters with the highest correlation with the turbulence of the aircraft.The research results show that the aircraft turbulence in South Africa mostly occurs in winter.Among the three terrains,the most frequent aircraft turbulence is in the eastern part of the Great Escarpment,where the highest altitude is 3,500 m,and the terrain contours are dense.It has a significant increase,so there is often greater turbulence.There are multiple consecutive peaks in the Cape Fuld area,and it is obvious that topographic waves propagate on the leeward side of the mountain,which coincides with the location of the high-prone area of aircraft turbulence.The South African plateau region,where terrain changes are not obvious,has the least number of aircraft turbulences,and the turbulence is mainly concentrated in the aircraft take-off and landing segments.Substituting the rasterized five terrain-related parameters into the support vector machine model of different kernel functions to predict the high-prone area of turbulence.The results show that the polynomial kernel function can be optimized for this training set.Use the support vector machine of the polynomial kernel function The prediction of the high-prone area of turbulence with different parameter combinations shows that terrain undulation,maximum altitude,wind horizontal velocity and vertical velocity are the best,up to 94.44%,which proves that aircraft turbulence is mainly related to these four terrain parameters.