Simulation of a multivariate and multidimensional turbulent wind velocity, and the consequential effect on a moving ground vehicle

A multidimensional and multivariate turbulent wind velocity is computed for two points on a moving vehicle, and then used to determine the consequential aerodynamic side force and yawing moment. The main objective is to provide a subroutine, with the above capabi1ities, that may be used by the ADAMS code in simulatinq the dynamic response of a moving automobile in a turbulent atmospheric wind fîeld. The wind fluctuations are derived from a Fourier time series representation of the power spectral density function appropriate for a reference frame moving perpendicular to the mean wind speed. The side forces and yawing moments are calculated from the resultant relative wind speeds at points on the front and back sections of the vehicle cummulatively. The statistical characteristics of the simulated wind time series are evaluated for the mean wind speed, mean wind direction, root-mean-square variance, auto-correlation, and cross-correlation. It is demonstrated that all of these features compare favorably with their respective tarqet values when a sufficîent number of discrete frequency components is used in the Fourier time series.