A Control Variate Method for Probabilistic Performance Assessment. Improved Estimates for Mean Performance Quantities of Interest
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
We present a method of control variates for calculating improved estimates for mean performance quantities of interest, E(PQI) , computed from Monte Carlo probabilistic simulations. An example of a PQI is the concentration of a contaminant at a particular location in a problem domain computed from simulations of transport in porous media. To simplify the presentation, the method is described in the setting of a one- dimensional elliptical model problem involving a single uncertain parameter represented by a probability distribution. The approach can be easily implemented for more complex problems involving multiple uncertain parameters and in particular for application to probabilistic performance assessment of deep geologic nuclear waste repository systems. Numerical results indicate the method can produce estimates of E(PQI)having superior accuracy on coarser meshes and reduce the required number of simulations needed to achieve an acceptable estimate.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE Office of Nuclear Energy (NE), Fuel Cycle Technologies (NE-5)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 1254272
- Report Number(s):
- SAND2016-4679; 640344
- Country of Publication:
- United States
- Language:
- English
Similar Records
Probabilistic calculations of groundwater travel time in heterogeneous three-dimensional porous media
Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators: Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators