Multisphere neutron spectrometers are pivotal in accurately measuring neutron flux across various fields. The inherent complexities and non-linearities in their calculation processes, such as correlated variables, make traditional uncertainty analysis methods based on theoretical models and empirical formulas (e.g., the GUM (Guide to the Uncertainty in Measurement) method) unsuitable.

This study aims to demonstrate the application of the Monte Carlo (MC) method as an effective tool for evaluating measurement uncertainty in multi-sphere spectrometers, addressing the challenges posed by complex systems and non-linear problems.

First of all, the MC method was employed to conduct probability density sampling of input variables to obtain the probability density distribution of the output variables. Detailed statistical characterization of the calculation results was allowed to provide a more comprehensive understanding compared to conventional methods. Then, numerous simulations were performed to take into account of variability and uncertainty in the input parameters, hence the robustness of the analysis was enhanced. Consequently, this technique overcome the limitations of traditional deterministic approaches, offering more reliable and nuanced insights into the system's behavior. Finally, experiments were carried out using multisphere neutron spectrometers and neutron field standard device (including 4 neutron sources), and the measurement results of neutron flux spectrum under 30 cm iron ball shielding were evaluated using above-mentioned method.

The spectrum unfolding results obtained by this method have a total uncertainty of 5% in the energy group of interest after being passed by the spectrum unscrambling program.

The application of the MC method offers a robust framework for the assessment of measurement uncertainties in multi-sphere neutron spectrometers. This study not only enhances the accuracy and reliability of spectrometer measurements, but also contributes to the broader field of neutron measurement techniques by providing a reference for the evaluation of complex systems.