Progress Toward Efficient Compression Algorithms for Application in Flight Simulation Environments

Abstract

Modern computational fluid dynamic simulations of flows about naval vessels produce an enormous amount of flow-field data. The computations are performed in order to model details of the erratic unsteady flows that can occur about naval superstructures. The flow-field data can then be used in flight simulators for naval pilot training purposes. Often, however, far too much data are generated for the flight simulators to process in real-time. This paper demonstrates the use of both proper orthogonal decomposition and Fourier series decomposition approaches for airwake dataset compression. The proper orthogonal decomposition method is used for compressing airwake data in the time domain, and Fourier series decomposition is used for compressing airwake data in the spatial coordinates. The approaches are applied to airwake data for a simplified frigate vessel model. A separate aircraft tanker configuration is also examined in order to demonstrate that the level of numerical precision can have an effect on the dataset compression results. Both approaches are effective at reducing airwake dataset size while preserving the dominant flow-field features, and both approaches are inexpensive and straightforward to implement. The level of dataset compression ultimately depends on the level of accuracy required by the user.