Time-dependent Monte Carlo simulations for neutron noise in void-containing water flow

Abstract

Neutron noise measurements for obtaining void-related information of a void-containing water flow are simulated using the time-dependent Monte Carlo technique. In the simulations, circular bubbles move upward in a two-dimensional water channel. The Auto Power Spectral Density (APSD) and the Cross-Correlation Function (CCF) of the neutrons that penetrate the channel are obtained using the time series data of the detected neutrons. If the void velocity is not exceedingly high and the length of neutron detector is not exceedingly short, the APSD shows a characteristic structure where repeated dips and peaks appear in the low frequency region (f < 100 Hz). The Monte Carlo simulations reproduce the structure in the APSDs. The void velocity can be roughly estimated using the frequencies of the dips and/or the peaks if the detector length is known. A CCF between two axially displaced detectors presents a prominent peak at the time lag that corresponds to the void transit time. Because the maximum point of the CCF can be clearly identified, the CCF is more favorable than the APSD for the purposes of determining the void velocity. The properties of the APSD and the CCF are discussed for a water channel for which the void velocity varies in the horizontal direction.