Chaotic oscillator based artificial neural network with LiDAR data for wind shear and turbulence forecasting and alerting

Abstract

Wind shear and turbulence, sudden changes in the wind direction and speed, are familiar danger to aviation as well as a complex and hard-to-predict phenomenon. Current research based on various approaches including the use of numerical weather prediction models, statistical models and machine learning models have provided some encouraging results in the area of long-term weather forecasting. But at the level of meso-scale and even micro-scale severe weather phenomena (involving very short-term chaotic perturbations) such as turbulence and wind shear phenomena, these approaches have not been so successful. The causes of wind shear may be different in different locations. In some places it is caused by microbursts, localized columns of sinking air, while in other places wind shear may result from meso-scale weather phenomena. Thus, algorithms and techniques used to predict wind shear in other places will not be applicable at an airport where wind shear and turbulence arise from larger-scale local conditions. This thesis focuses on the use of chaotic oscillatory-based neural networks (CONN) for predicting wind shear arising from meso-scale weather phenomenon at the Hong Kong International Airport. Using historical local data, the Hong Kong Observatory, simulations show that CONN is able to forecast wind shear events with a pleasing level of accuracy.