Methods of Statistical Modeling of Atmospheric Turbulence

Abstract

AbstractThe paper studies the methods of statistical modeling of Gaussian random processes used in the modeling of atmospheric turbulence. The need to model atmospheric turbulence arises in many applied problems. Atmospheric turbulence is the main factor that determines the aircraft design resource, affects the assessment of aviation resource. Atmospheric turbulence affects the flight trajectory of unmanned aerial vehicles, the assessment of the variance of the deflection of ballistic projectiles during firing. For practical calculations, we use the implementations of the random field obtained as a result of statistical modeling. The study is based on the methods of statistical modeling of stationary Gaussian random processes and fields. When modeling stationary Gaussian processes, the shaping filter method can be used. The paper proposes a modeling method that uses spectral representations of a stationary Gaussian process and field. Using the spectral representation allows you to build models with a given accuracy and reliability. The paper presents new estimates of accuracy and reliability in the space of integrated functions. The obtained estimates depend on the parameters of the spectral decompositions. This allows us to take into account the properties of atmospheric turbulence in the construction of statistical implementations. The use of spectral images of Gaussian random processes and fields allows to obtain simpler statistical modeling algorithms. This is very important when organizing and conducting computational experiments to solve applied problems.KeywordsStatistical simulationGaussian random processesAtmospheric turbulence