Extraction of Shock Waves and Separation and Attachment Lines from Computational Fluid Dynamics Simulations using Subjective Logic
Computational fluid dynamics that accurately simulate highly complex fluid flow situations can generate terabytes of data, greatly increasing the time required for the researcher to fully analyze the data. Presented in this paper is a method to extract shock waves and separation and attachment lines using subjective logic. This method uses software agents that make decisions about extracted features in converging and converged data sets based on belief, disbelief, and uncertainty of extracted data sets. The method uses multiple algorithms for feature detection, accounting for the strengths and weaknesses of each. Using the subjective logic architecture, a final opinion of each extraction is developed. When evaluating an opinion, probability expectation is calculated to interpret how believable an extraction is. The method was validated with simulations of the Onera M6 wing and cylinder in a cross flow. Erroneous extractions have a low probability expectation and believable extractions have a high probability expectation.
M. C. Lively et al., "Extraction of Shock Waves and Separation and Attachment Lines from Computational Fluid Dynamics Simulations using Subjective Logic," Proceedings of the 50th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition (2012, Nashville, TN), American Institute of Aeronautics and Astronautics (AIAA), Jan 2012.
50th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition (2012: Jan. 9-12, Nashville, TN)
Electrical and Computer Engineering
Keywords and Phrases
Attachment Lines; Complex Fluid Flow; Computational Fluid Dynamics Simulations; Cross Flows; Data Sets; Feature Detection; High Probability; Low Probability; Multiple Algorithms; Subjective Logic, Aerospace Engineering; Computational Fluid Dynamics; Exhibitions; Software Agents, Shock Waves
Article - Conference proceedings
© 2012 American Institute of Aeronautics and Astronautics (AIAA), All rights reserved.