Reduction of statistic error in Mihalczo subcriticality measurement

Hazama, Taira 

The theoretical formula for the statistical error estimation in Mihalczo method was derived, and the dependence of the error were investigated on the facility to be measured and on the parameter in the data analysis. The formula was derived based on the reactor noise theory and the error theory for the frequency analysis, and found that the error depends on such parameters as the prompt neutron decay constant, detector efficiencies, and the frequency bandwidth. Statistical errors estimated with the formula was compared with experimental values and verified to be reasonable. Through parameter surveys, it is found that there is a setting in the data analysis to optimize the magnitude and reliability of the errors. In the experiment performed in DCA subcriticality measurement facility, it is estimated experimentally that the measurement requires 20 minutes to obtain the statistic error of 1% for the keff 0.9. According to the error theory, this should be reduced to 3sec in the aqueous fuel system typical in fuel reprocessing plant.