TY - GEN
T1 - Dynamic adaptive sampling based on Kriging surrogate models for efficient uncertainty quantification
AU - Shimoyama, Koji
AU - Kawai, Soshi
AU - Alonso, Juan J.
N1 - Funding Information:
This work was supported in part by the Young Researcher Overseas Visits Program for Vitalizing Brain Circulation, Japan Society for the Promotion of Science (JSPS), the JAXA International Top Young Fellowship Program, and JSPS Grant-in-Aid for Young Scientists (B) KAKENHI 24760670.
PY - 2013
Y1 - 2013
N2 - New Kriging-surrogate-model-based dynamic adaptive sampling methods are proposed for an accurate and efficient uncertainty quantification (UQ). The criteria for the proposed dynamic adaptive sampling are based on the combination of both the uncertainty and the gradient information of the Kriging predictors. The polynomial errors (related to Runge's phenomenon) appeared near the endpoints in the stochastic space are reduced by adding an extra error-estimate term (based on the difference of the Kriging predictors with different correlation functions) in the adaptive sampling criteria. The proposed Kriging-based dynamic adaptive sampling methods are tested on one-dimensional and two-dimensional analytic functions with smooth and non-smooth response surfaces. The method shows a superior performance to estimate the statistics of output solution in terms of efficiency, accuracy, and robustness regardless of the choice of initial samples and the smoothness and dimensionality of stochastic space compared to the existing criterion based on only the Kriging predictor uncertainty.
AB - New Kriging-surrogate-model-based dynamic adaptive sampling methods are proposed for an accurate and efficient uncertainty quantification (UQ). The criteria for the proposed dynamic adaptive sampling are based on the combination of both the uncertainty and the gradient information of the Kriging predictors. The polynomial errors (related to Runge's phenomenon) appeared near the endpoints in the stochastic space are reduced by adding an extra error-estimate term (based on the difference of the Kriging predictors with different correlation functions) in the adaptive sampling criteria. The proposed Kriging-based dynamic adaptive sampling methods are tested on one-dimensional and two-dimensional analytic functions with smooth and non-smooth response surfaces. The method shows a superior performance to estimate the statistics of output solution in terms of efficiency, accuracy, and robustness regardless of the choice of initial samples and the smoothness and dimensionality of stochastic space compared to the existing criterion based on only the Kriging predictor uncertainty.
UR - http://www.scopus.com/inward/record.url?scp=84880824578&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84880824578&partnerID=8YFLogxK
U2 - 10.2514/6.2013-1470
DO - 10.2514/6.2013-1470
M3 - Conference contribution
AN - SCOPUS:84880824578
SN - 9781624102233
T3 - 54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference
BT - 54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
T2 - 54th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
Y2 - 8 April 2013 through 11 April 2013
ER -