Uncertainty in unprotected loss-of-heat-sink, loss-of-flow, and transient-overpower accidents.

Morris, E. E.

doi:10.2172/948804

Cited by 5 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using this new definition of regression parameters, one can predict the mean value of the model as [28] 18) where w * is the vector of covariances between the test point function values and the gradient at each training point. The variance associated with this prediction can now be calculated as:…”

Section: Universal Kriging Model With Derivative Valuesmentioning

confidence: 99%

“…A description of the underlying mathematical model is available, but the complete process of obtaining the numerical solution is documented only in the code comments, and sparsely at that. MATWS was used in combination with another simulation tool, GoldSim, to model unprotected loss-of-heat-sink, loss-of-flow and transient-overpower accidents [18].…”

Section: Matwsmentioning

confidence: 99%

“…We note that if the maximum is unique, then its value is differentiable with respect to the input parameters. In our case the inputs (corresponding to physically significant sources of uncertainty) are the radial core expansion, control rod driveline expansion, fuel axial expansion, and Doppler reactivity feedback coefficients [1,18].…”

Section: Matwsmentioning

confidence: 99%

See 2 more Smart Citations

Gradient-Enhanced Universal Kriging for Uncertainty Propagation

Lockwood

Anitescu

2012

Nuclear Science and Engineering

View full text Add to dashboard Cite

In this work, we investigate the issue of providing a statistical model for the response of a computer model-described nuclear engineering system, for use in uncertainty propagation. The motivation behind our approach is the need for providing an uncertainty assessment even in the circumstances where only a few samples are available. Building on our recent work in using a regression approach with derivative information for approximating the system response, we investigate the ability of a universal gradientenhanced Kriging model to provide a means for inexpensive uncertainty quantification. The universal Kriging model can be viewed as a hybrid of polynomial regression and Gaussian process regression. For this model, the mean behavior of the surrogate is determined by a polynomial regression, and deviations from this mean are represented as a Gaussian process. Tests with explicit functions and nuclear engineering models show that the universal gradient-enhanced Kriging model provides a more accurate surrogate model when compared to either regression or ordinary Kriging models. In addition we investigate the ability of the Kriging model to provide error predictions and bounds for regression models.

show abstract

Section: Universal Kriging Model With Derivative Valuesmentioning

confidence: 99%

Section: Matwsmentioning

confidence: 99%

See 1 more Smart Citation

Gradient-Enhanced Universal Kriging for Uncertainty Propagation

Lockwood

Anitescu

2012

Nuclear Science and Engineering

View full text Add to dashboard Cite

show abstract

“…The feasibility of such approaches has been demonstrated for uncertainty quantification in unprotected loss-of-heat sink, loss-of-flow, and transient overpower accidents. In the particular case outlined by Morris, [9] the commercial stochastic simulation code GoldSim (developed originally for geological repository modeling) was coupled to simplified point-kinetics and thermalhydraulics models extracted from the SAS4A/SYSSYS-1 computer code and used to rapidly simulate 10,000 realizations in order to characterize the uncertainty of transient responses within the first hour of a given triggering event. While the large number of realizations used in this demonstration is impractical for high-resolution approaches (e.g., using LES-or RANS-based thermal-hydraulics analysis) it is feasible to consider using MC to simultaneously drive hundreds or thousands of realizations in parallel of existing or future neutronic, thermal-hydraulic, and safety analysis codes.…”

Section: Monte Carlo Methodsmentioning

confidence: 99%

“…[8] This comparison led to the conclusion that the development and implementation costs for DST were likely to be significant in comparison to the development of forward solutions. Both Vaurio and Morris [9] developed techniques for statistically sampling input parameters of large codes to propagate and measure the impact of uncertainty on selected output parameters. Vaurio developed the PROSA code [10,11] to evaluate probabilistic response surfaces to obtain probability distributions for consequences of reactor transients.…”

Section: Transient Analyses and Statistical Quantificationmentioning

confidence: 99%

Methods for quantifying uncertainty in fast reactor analyses.

Fanning¹,

Fischer²

2008

View full text Add to dashboard Cite

Liquid-metal-cooled fast reactors in the form of sodium-cooled fast reactors have been successfully built and tested in the U.S. and throughout the world. However, no fast reactor has operated in the U.S. for nearly fourteen years. More importantly, the U.S. has not constructed a fast reactor in nearly 30 years. In addition to reestablishing the necessary industrial infrastructure, the development, testing, and licensing of a new, advanced fast reactor concept will likely require a significant base technology program that will rely more heavily on modeling and simulation than has been done in the past. The ability to quantify uncertainty in modeling and simulations will be an important part of any experimental program and can provide added confidence that established design limits and safety margins are appropriate. In addition, there is an increasing demand from the nuclear industry for best-estimate analysis methods to provide confidence bounds along with their results. The ability to quantify uncertainty will be an important component of modeling that is used to support design, testing, and experimental programs. Three avenues of UQ investigation are proposed. Two relatively new approaches are described which can be directly coupled to simulation codes currently being developed under the Advanced Simulation and Modeling program within the Reactor Campaign. A third approach, based on robust Monte Carlo methods, can be used in conjunction with existing reactor analysis codes as a means of verification and validation of the more detailed approaches. Methods for Quantifying Uncertainty in Fast Reactor Analyses vi

show abstract