Details
| Original language | English |
|---|---|
| Pages (from-to) | 19-38 |
| Number of pages | 20 |
| Journal | International Journal for Uncertainty Quantification |
| Volume | 11 |
| Issue number | 5 |
| Publication status | Published - 2021 |
Abstract
Computational cost often hinders the calibration of complex computer models. In this context, history matching (HM) is becoming a widespread calibration strategy, with applications in many disciplines. HM uses a statistical approxi-mation, also known as an emulator, to the model output, in order to mitigate computational cost. The process starts with an observation of a physical system. It then produces progressively more accurate emulators to determine a non-implausible domain: a subset of the input space that provides a good agreement between the model output and the data, conditional on the model structure, the sources of uncertainty, and an implausibility measure. In HM, it is essential to generate samples from the nonimplausible domain, in order to run the model and train the emulator until a stopping condition is met. However, this sampling can be very challenging, since the nonimplausible domain can become orders of magnitude smaller than the original input space very quickly. This paper proposes a solution to this problem using subset simulation, a rare event sampling technique that works efficiently in high dimensions. The proposed approach is demonstrated via calibration and robust design examples from the field of aerospace engineering.
Keywords
- Gaussian process emulation, History matching, Robust design, Subset simulation
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
- Mathematics(all)
- Modelling and Simulation
- Mathematics(all)
- Discrete Mathematics and Combinatorics
- Mathematics(all)
- Control and Optimization
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: International Journal for Uncertainty Quantification, Vol. 11, No. 5, 2021, p. 19-38.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - History matching with subset simulation
AU - Gong, Z. T.
AU - DiazDelaO, F. A.
AU - Hristov, P. O.
AU - Beer, M.
N1 - Publisher Copyright: © 2021 by Begell House, Inc. www.begellhouse.com.
PY - 2021
Y1 - 2021
N2 - Computational cost often hinders the calibration of complex computer models. In this context, history matching (HM) is becoming a widespread calibration strategy, with applications in many disciplines. HM uses a statistical approxi-mation, also known as an emulator, to the model output, in order to mitigate computational cost. The process starts with an observation of a physical system. It then produces progressively more accurate emulators to determine a non-implausible domain: a subset of the input space that provides a good agreement between the model output and the data, conditional on the model structure, the sources of uncertainty, and an implausibility measure. In HM, it is essential to generate samples from the nonimplausible domain, in order to run the model and train the emulator until a stopping condition is met. However, this sampling can be very challenging, since the nonimplausible domain can become orders of magnitude smaller than the original input space very quickly. This paper proposes a solution to this problem using subset simulation, a rare event sampling technique that works efficiently in high dimensions. The proposed approach is demonstrated via calibration and robust design examples from the field of aerospace engineering.
AB - Computational cost often hinders the calibration of complex computer models. In this context, history matching (HM) is becoming a widespread calibration strategy, with applications in many disciplines. HM uses a statistical approxi-mation, also known as an emulator, to the model output, in order to mitigate computational cost. The process starts with an observation of a physical system. It then produces progressively more accurate emulators to determine a non-implausible domain: a subset of the input space that provides a good agreement between the model output and the data, conditional on the model structure, the sources of uncertainty, and an implausibility measure. In HM, it is essential to generate samples from the nonimplausible domain, in order to run the model and train the emulator until a stopping condition is met. However, this sampling can be very challenging, since the nonimplausible domain can become orders of magnitude smaller than the original input space very quickly. This paper proposes a solution to this problem using subset simulation, a rare event sampling technique that works efficiently in high dimensions. The proposed approach is demonstrated via calibration and robust design examples from the field of aerospace engineering.
KW - Gaussian process emulation
KW - History matching
KW - Robust design
KW - Subset simulation
UR - http://www.scopus.com/inward/record.url?scp=85107430258&partnerID=8YFLogxK
U2 - 10.1615/Int.J.UncertaintyQuantification.2021033543
DO - 10.1615/Int.J.UncertaintyQuantification.2021033543
M3 - Article
AN - SCOPUS:85107430258
VL - 11
SP - 19
EP - 38
JO - International Journal for Uncertainty Quantification
JF - International Journal for Uncertainty Quantification
SN - 2152-5080
IS - 5
ER -