03.4 - Applied Aerodynamics
SPARSE POLYNOMIAL CHAOS EXPANSION BASED ON FULLY ADAPTIVE FORWARD-BACKWARD SELECTION METHOD
H. Zhao, Sun Yat-sen University, China; Z.H. Gaoš, L. Xiaš; šNorthwestern Polytechnical University, China
A novel fully adaptive forward-backward selection (FAFBS) algorithm has been developed for reconstructing sparse PC. The FAFBS method is investigated by several applications. The results reveal that the proposed FAFBS method can efficiently identify the significant PC contributions for QoI's, and reproduce sparser PC metamodel and more accurate UQ results than classical OMP and full PC methods.rn