Particle-clustering methods have been explored for application to the problem of predicting brownout conditions, which can occur when a rotorcraft operates over surfaces covered with loose sediment and stirs up large dust clouds that impede the pilot’s outward visibility. A significant issue in such simulations is the tracking of the very large number of particles needed to predict with acceptable fidelity the dual-phase flow properties and optical characteristics of the dust clouds. To this end, computationally efficient Lagrangian approaches were explored for the simulation of dilute carrier-particle suspensions at low Reynolds numbers of the relative particle motion. The algorithms examined were the Gaussian method, the k-means method, and Osiptsov’s method, which were all compared in terms of their accuracy versus computational cost against solutions obtained by directly solving for the individual particle motions. Specific results were computed for a prototypical flowfield that mimics the highly unsteady, two-phase vortical particle flow responsible for the development of brownout conditions. It is shown that although clustering algorithms can be problem dependent and have bounds of applicability, in some conditions they offer the potential to significantly reduce the computational costs of brownout dust cloud simulations while still retaining good accuracy.
Particle-Clustering Algorithms for the Prediction of Brownout Dust Clouds
AIAA Journal ; 51 , 5 ; 1080-1094
2013-05-01
Article (Journal)
Electronic Resource
English
Evaluation of Particle Clustering Algorithms in the Prediction of Brownout Dust Clouds
British Library Conference Proceedings | 2011
|Predictions of brownout dust clouds compared to photogrammetry measurements
Tema Archive | 2013
|Comparisons of Predicted Brownout Dust Clouds with Photogrammetry Measurements
British Library Conference Proceedings | 2011
|Predictions of Brownout Dust Clouds Compared to Photogrammetry Measurements 022001
Online Contents | 2013
|