Results of Mining Data Features During Computational Fluid Dynamics Simulations

Computational Fluid Dynamics (CFD) simulations provide a variety of data mining challenges and present an opportunity for novel solutions. A core challenge is that CFD computations can require weeks of computation on expensive high performance clusters, delaying investigation of results until a fully converged solution is obtained. 21st Century Systems, Inc. and Brigham Young University have been collaborating on a concurrent agent-enabled feature extraction project designed to mine feature data while a CFD simulation is executing. This paper summarizes the work and presents the combined results from a unified Industry/Application perspective. Empirical results show the concept validity and capability of obtaining CFD feature information much earlier than waiting for complete simulation convergence, which may ultimately save extensive computational resources and provide much quicker turnaround in development requiring CFD modeling.