Application of global sensitivity analysis approach to exercise I-1 of the OECD LWR UAM Benchmark (original) (raw)

Abstract

Models of varying complexity are developed to approximate or mimic systems and processes in different aspects of the real world (e.g. physical, environmental, social, or economic). Applying models in such domains inevitably involves uncertainty in both the model representation and in the input data. Uncertainty and sensitivity analysis techniques can be applied to study uncertainty in model predictions arising from imprecisely-known processes and input data. Sensitivity analysis involves determining the contribution of individual inputs to the uncertainty in model predictions. Global sensitivity analysis deals with uncertainty sources spanning over finite or infinite ranges of uncertainties and with the simultaneous variation of such sources. This in turn enables the identification of high-order interactions among inputs in determining the uncertainty in the output of interest. This paper applies global sensitivity analysis to the modeling of nuclear reactor applications using the Monte Carlo method. In particular, the tests of Exercise 1 (I-1), “Cell Physics”, of the OECD Benchmark for Uncertainty Analysis in Best-Estimate Modeling (UAM) for Design, Operation and Safety Analysis of LWRs (OECD LWR UAM Benchmark), are analyzed.

Shadi Ghrayeb hasn't uploaded this paper.

Let Shadi know you want this paper to be uploaded.

Ask for this paper to be uploaded.