[00275] A fast data-driven method for designing compressible shock dominant flows
Session Time & Room : 2E (Aug.22, 17:40-19:20) @F310
Type : Contributed Talk
Abstract : We will present a new class of high-order numerical algorithms for computational fluid dynamics. Called "GP-MOOD," the new finite volume method is based on the Gaussian Processes modeling that generalizes the Gaussian probability distribution. Solutions at shocks and discontinuities are handled by the improved Multidimensional Optimal Order Detection (MOOD) strategy, which controls numerical stability and accuracy in an "a posteriori" shock-capturing formalism. We also introduce a new data-driven "a priori" MOOD method.