Ship Airwake Turbulence Modeling Using Moving Stochastic Spectral Filters

This paper describes an ongoing study of atmospheric turbulence modeling using stochastic spectral filters derived from CFD data. The filters, when excited by white noise, generate velocity perturbations for inclusion in the aerodynamic models of desktop and real-time helicopter simulations operating in a ship airwake. The filter parameters were identified from air velocity time history CFD data using system identification techniques. The filter performance was evaluated using a Simulink® model with a moving filter whose coefficients were variable depending on the location of the filter on the airwake grid. The moving filter location corresponds to an aerodynamic computation point of interest for inclusion in blade element and/or aerodynamic panel models. Translational velocity over the grid was considered, along with a harmonic perturbation which is representative of the motion of a rotor blade element. The resulting turbulence velocities compare favorably with interpolation within the CFD data, in both the time and frequency domains. Further, the filters represent a substantial increase in computational efficiency over the method of preloading the CFD time histories into computer memory.