Accurate Estimation of Subcriticality Using Indirect Bias Estimation Method, (I) Theory.

Abstract

To estimate subcriticality of a system, a new method, Indirect Bias Estimation Method, is proposed. Instead of comparing a calculated keff with a "measured" one which cannot be measured directly, a bias in calculated keff is evaluated by direct comparison between measured and calculated quantities which are measureable in subcritical experiments. With these errors of measurable quantities, accuracy of a subcriticality calculation is indirectly derived. Three estimation methods based on the source multiplication method, the pulse neutron method, and the exponential experiment method are presented. For three experimental methods, neutron count rate distribution, prompt neutron decay constant, and spatial decay constant are compared between calculation and measurement, respectively, deriving biases in calculated keff's. With biases obtained by this procedures, namely Indirect Bias Estimation Method, subcriticality of a system can be estimated more precisely than by "measured" keff's.

Monte Carlo Alpha Iteration Algorithm for a Subcritical System Analysis

“…The dynamic MC calculations cover the estimation from numerical simulations of the pulsed neutron source (PNS) experiment by fitting time-dependent tally results of a neutron detector to the exponential function. However, it is well known that a starting time of fitting should be carefully determined by ensuring the convergence of estimated values [2,3]. The TART code [4] is equipped with a unique method to measure in time-stepwise MC simulations with controlling the neutron population by the combing algorithm [1].…”

Subcriticality Measurement by Neutron Source Multiplication Method Swith a Fundamental Mode Extraction

“…These factors can be evaluated from numerical analyses of the eigenvalue problem and the fixed source problem. The inverse bias estimation method 4) is based on the assumption that the "true" k eff of a subcritical system is a mathematical quantity and nonmeasurable variable. The bias in the calculated k eff is estimated from the deviation between a measurable variable, such as a neutron count rate or a fission reaction rate distribution, obtained from actual measurements and one determined by calculations of the neutron diffusion equation with an external neutron source.…”