Wind Shear Prediction from Light Detection and Ranging Data Using Machine Learning Methods

The main aim of this paper is to propose a statistical indicator for wind shear prediction from Light Detection and Ranging (LIDAR) observational data. Accurate warning signal particularly important aviation safety. challenges are that may result sustained change the headwind possible velocity have wide range. Traditionally, models based on terrain-induced setting used detect phenomena. Different traditional methods, we study which measure variation headwinds multiple profiles. Because value nonnegative, decision rule one-side normal distribution employed distinguish cases non-wind cases. Experimental results real data sets obtained at Hong Kong International Airport runway presented demonstrate proposed quite effective. performance method better than by supervised learning methods (LDA, KNN, SVM, logistic regression). This model would also provide more accurate warnings pilots improve Wind Turbulence Warning System.