Author: W. V. Harper
Publisher:
ISBN:
Category : Geology
Languages : en
Pages : 28
Book Description
Sensitivity/uncertainty Analysis Techniques for Nonstochastic Computer Codes
Methods and Computer Codes for Probabilistic Sensitivity and Uncertainty Analysis
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :
Book Description
This paper describes the methods and applications experience with two computer codes that are now available from the National Energy Software Center at Argonne National Laboratory. The purpose of the SCREEN code is to identify a group of most important input variables of a code that has many (tens, hundreds) input variables with uncertainties, and do this without relying on judgment or exhaustive sensitivity studies. Purpose of the PROSA-2 code is to propagate uncertainties and calculate the distributions of interesting output variable(s) of a safety analysis code using response surface techniques, based on the same runs used for screening. Several applications are discussed, but the codes are generic, not tailored to any specific safety application code. They are compatible in terms of input/output requirements but also independent of each other, e.g., PROSA-2 can be used without first using SCREEN if a set of important input variables has first been selected by other methods. Also, although SCREEN can select cases to be run (by random sampling), a user can select cases by other methods if he so prefers, and still use the rest of SCREEN for identifying important input variables.
Publisher:
ISBN:
Category :
Languages : en
Pages :
Book Description
This paper describes the methods and applications experience with two computer codes that are now available from the National Energy Software Center at Argonne National Laboratory. The purpose of the SCREEN code is to identify a group of most important input variables of a code that has many (tens, hundreds) input variables with uncertainties, and do this without relying on judgment or exhaustive sensitivity studies. Purpose of the PROSA-2 code is to propagate uncertainties and calculate the distributions of interesting output variable(s) of a safety analysis code using response surface techniques, based on the same runs used for screening. Several applications are discussed, but the codes are generic, not tailored to any specific safety application code. They are compatible in terms of input/output requirements but also independent of each other, e.g., PROSA-2 can be used without first using SCREEN if a set of important input variables has first been selected by other methods. Also, although SCREEN can select cases to be run (by random sampling), a user can select cases by other methods if he so prefers, and still use the rest of SCREEN for identifying important input variables.
Methods and Computer Codes for Probabilistic Sensitivity and Uncertainty Analysis
Author: J. K. Vaurio
Publisher:
ISBN:
Category : Nuclear power plants
Languages : en
Pages : 10
Book Description
Publisher:
ISBN:
Category : Nuclear power plants
Languages : en
Pages : 10
Book Description
Sensitivity and Uncertainty Analysis, Volume II
Author: Dan G. Cacuci
Publisher: CRC Press
ISBN: 020348357X
Category : Mathematics
Languages : en
Pages : 367
Book Description
As computer-assisted modeling and analysis of physical processes have continued to grow and diversify, sensitivity and uncertainty analyses have become indispensable scientific tools. Sensitivity and Uncertainty Analysis. Volume I: Theory focused on the mathematical underpinnings of two important methods for such analyses: the Adjoint Sensitivity Analysis Procedure and the Global Adjoint Sensitivity Analysis Procedure. This volume concentrates on the practical aspects of performing these analyses for large-scale systems. The applications addressed include two-phase flow problems, a radiative convective model for climate simulations, and large-scale models for numerical weather prediction.
Publisher: CRC Press
ISBN: 020348357X
Category : Mathematics
Languages : en
Pages : 367
Book Description
As computer-assisted modeling and analysis of physical processes have continued to grow and diversify, sensitivity and uncertainty analyses have become indispensable scientific tools. Sensitivity and Uncertainty Analysis. Volume I: Theory focused on the mathematical underpinnings of two important methods for such analyses: the Adjoint Sensitivity Analysis Procedure and the Global Adjoint Sensitivity Analysis Procedure. This volume concentrates on the practical aspects of performing these analyses for large-scale systems. The applications addressed include two-phase flow problems, a radiative convective model for climate simulations, and large-scale models for numerical weather prediction.
Uncertainty and Sensitivity Analysis for Long-running Computer Codes
Author: Dustin R. Langewisch
Publisher:
ISBN:
Category :
Languages : en
Pages : 158
Book Description
This thesis presents a critical review of existing methods for performing probabilistic uncertainty and sensitivity analysis for complex, computationally expensive simulation models. Uncertainty analysis (UA) methods reviewed include standard Monte Carlo simulation, Latin Hypercube sampling, importance sampling, line sampling, and subset simulation. Sensitivity analysis (SA) methods include scatter plots, Monte Carlo filtering, regression analysis, variance-based methods (Sobol' sensitivity indices and Sobol' Monte Carlo algorithms), and Fourier amplitude sensitivity tests. In addition, this thesis reviews several existing metamodeling techniques that are intended provide quick-running approximations to the computer models being studied. Because stochastic simulation-based UA and SA rely on a large number (e.g., several thousands) of simulations, metamodels are recognized as a necessary compromise when UA and SA must be performed with long-running (i.e., several hours or days per simulation) computational models. This thesis discusses the use of polynomial Response Surfaces (RS), Artificial Neural Networks (ANN), and Kriging/Gaussian Processes (GP) for metamodeling. Moreover, two methods are discussed for estimating the uncertainty introduced by the metamodel. The first of these methods is based on a bootstrap sampling procedure, and can be utilized for any metamodeling technique.
Publisher:
ISBN:
Category :
Languages : en
Pages : 158
Book Description
This thesis presents a critical review of existing methods for performing probabilistic uncertainty and sensitivity analysis for complex, computationally expensive simulation models. Uncertainty analysis (UA) methods reviewed include standard Monte Carlo simulation, Latin Hypercube sampling, importance sampling, line sampling, and subset simulation. Sensitivity analysis (SA) methods include scatter plots, Monte Carlo filtering, regression analysis, variance-based methods (Sobol' sensitivity indices and Sobol' Monte Carlo algorithms), and Fourier amplitude sensitivity tests. In addition, this thesis reviews several existing metamodeling techniques that are intended provide quick-running approximations to the computer models being studied. Because stochastic simulation-based UA and SA rely on a large number (e.g., several thousands) of simulations, metamodels are recognized as a necessary compromise when UA and SA must be performed with long-running (i.e., several hours or days per simulation) computational models. This thesis discusses the use of polynomial Response Surfaces (RS), Artificial Neural Networks (ANN), and Kriging/Gaussian Processes (GP) for metamodeling. Moreover, two methods are discussed for estimating the uncertainty introduced by the metamodel. The first of these methods is based on a bootstrap sampling procedure, and can be utilized for any metamodeling technique.
Sensitivity & Uncertainty Analysis, Volume 1
Author: Dan G. Cacuci
Publisher: CRC Press
ISBN: 0203498798
Category : Mathematics
Languages : en
Pages : 304
Book Description
As computer-assisted modeling and analysis of physical processes have continued to grow and diversify, sensitivity and uncertainty analyses have become indispensable investigative scientific tools in their own right. While most techniques used for these analyses are well documented, there has yet to appear a systematic treatment of the method based
Publisher: CRC Press
ISBN: 0203498798
Category : Mathematics
Languages : en
Pages : 304
Book Description
As computer-assisted modeling and analysis of physical processes have continued to grow and diversify, sensitivity and uncertainty analyses have become indispensable investigative scientific tools in their own right. While most techniques used for these analyses are well documented, there has yet to appear a systematic treatment of the method based
Technical Progress Report for the Quarter ...
Author:
Publisher:
ISBN:
Category : Radioactive waste disposal
Languages : en
Pages : 84
Book Description
Publisher:
ISBN:
Category : Radioactive waste disposal
Languages : en
Pages : 84
Book Description
Uncertainty Analysis
Author: Ralph E. Thomas
Publisher:
ISBN:
Category : Probabilities
Languages : en
Pages : 76
Book Description
Publisher:
ISBN:
Category : Probabilities
Languages : en
Pages : 76
Book Description
A Sensitivity Analysis Technique for Application to Deterministic Models
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :
Book Description
An important element of any uncertainty analysis consists of evaluating the sensitivity of the output uncertainty distributions to the input assumptions. These sensitivity analyses require extensive information regarding mathematical correlations between input and output variables which are generally obtained through repeated computer runs of a physical model. However, in predicting uncertainties associated with severe accident sources terms using contemporary methods, techniques must be devised which reduce the need for extensive commutation using large computer codes. The purpose of the current paper is to propose a new method for sensitivity analysis which does not utilize response surface methods, but instead relies directly on the results obtained from the original computer code calculations. 3 refs., 1 fig.
Publisher:
ISBN:
Category :
Languages : en
Pages :
Book Description
An important element of any uncertainty analysis consists of evaluating the sensitivity of the output uncertainty distributions to the input assumptions. These sensitivity analyses require extensive information regarding mathematical correlations between input and output variables which are generally obtained through repeated computer runs of a physical model. However, in predicting uncertainties associated with severe accident sources terms using contemporary methods, techniques must be devised which reduce the need for extensive commutation using large computer codes. The purpose of the current paper is to propose a new method for sensitivity analysis which does not utilize response surface methods, but instead relies directly on the results obtained from the original computer code calculations. 3 refs., 1 fig.