Loading...
Please wait, while we are loading the content...
Similar Documents
Polynomial Chaos Expansion for Global 1 Sensitivity Analysis applied to a model of 2 radionuclide migration in randomly 3 heterogeneous aquifers 4 5
| Content Provider | Semantic Scholar |
|---|---|
| Author | Ciriello, Vincenzo Federico, Vittorio Di Riva, Mario Cadini, Francesco Sanctis, Jacopo De Zio, Enrico Guadagnini, Alberto |
| Copyright Year | 2014 |
| Abstract | We perform Global Sensitivity Analysis (GSA) through Polynomial Chaos Expansion (PCE) on a 22 contaminant transport model for the assessment of radionuclide concentration at a given control 23 location in a heterogeneous aquifer, following a release from a near surface repository of 24 radioactive waste. The aquifer hydraulic conductivity is modeled as a stationary stochastic process 25 in space. We examine the uncertainty in the first two (ensemble) moments of the peak 26 concentration, as a consequence of incomplete knowledge of (a) the parameters characterizing the 27 variogram of hydraulic conductivity, (b) the partition coefficient associated with the migrating 28 radionuclide, (c) the effective dispersivity at the scale of interest. These quantities are treated as 29 random variables and a variance-based GSA is performed in a numerical Monte Carlo framework. 30 This entails solving groundwater flow and transport processes within an ensemble of hydraulic 31 conductivity realizations generated upon sampling the space of the considered random variables. 32 The Sobol indices are adopted as sensitivity measures to provide an estimate of the role of 33 uncertain parameters on the (ensemble) target moments of the variable of interest. The calculation 34 of the indices is performed by employing PCE as a surrogate model of the migration process to 35 reduce the computational burden. We show that the proposed methodology (a) allows identifying 36 the influence of uncertain parameters on key statistical moments of the peak concentration (b) 37 enables extending the number of Monte Carlo iterations to attain convergence of the (ensemble) 38 target moments and (c) leads to considerable saving of computational time while keeping 39 acceptable accuracy. 40 |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://hal-supelec.archives-ouvertes.fr/docs/00/92/63/41/PDF/Ciriello.pdf |
| Alternate Webpage(s) | https://hal-supelec.archives-ouvertes.fr/hal-00926341/document |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
