Speaker
Description
Authors (affiliation): 1. Hiroki Hashimoto (Hirosaki Univ., Japan), 2. Ryohei Yamada (Hirosaki Univ., Japan), 3. Kouichi Sasaki (JNFL, Japan), 4. Kanna Yamaguchi (JNFL, Japan), 5. Yasutaka Omori (Hirosaki Univ., Japan), 6. Masahiro Hosoda (Hirosaki Univ., Japan), 7. Shinji Tokonami (Hirosaki Univ., Japan).
Early detection of the abnormal release of artificial radionuclides from nuclear facilities requires discriminative measurements of natural and artificial radionuclides. Currently, there are measuring instruments with discrimination functions, but false alarm events have been reported. Thus, a reliable detection system with as few false alarms as possible is required. Since false alarms may occur when judgments are made based on only one method, we are constructing some discriminative methods and developing a system that combines them to detect the abnormal release of artificial radionuclides quickly. In this study, we examined two discriminative methods as a system component.
Discriminative methods were developed based on the aerosol monitor using a silicon semiconductor detector. The first method is based on the counting ratio in the region of interest (ROI). In this method, the ROI for plutonium and the ROI for 218Po and 212Bi, which are decay products of 222Rn, were set. The counting ratio within these ROIs was evaluated in the presence of only natural radionuclides. Artificial radionuclides were considered to have been detected when the counting ratio exceeded 3σ of this ratio. The mean and standard deviation of the counting ratio were evaluated to be 0.45 ± 0.55 using the radioactive aerosol chamber at Hirosaki University. Although no change in this ratio was observed at several concentrations in the chamber, a significant variation in the ratio was observed under the environment due to a few counts. A one-month measurement in September 2021 in Rokkasho Village, Aomori Prefecture, Japan, the mean and standard deviation of the ratio were evaluated to be 0.54 ± 0.28, and the number of times above the 3σ range (0 - 1.38) was 1.6% of the total.
The second method uses the relationship between 214Po counts and gross alpha counts. Taking advantage of the fact that previous studies have shown a correlation between gross alpha counts and alpha-beta coincidence counts from 214Bi and 214Po, 214Po counts were directly obtained and correlated with gross alpha counts. As a result, a strong positive correlation was found between them, and a regression equation was obtained. Based on the regression equation, gross alpha counts based on 214Po were calculated and subtracted from the measured gross alpha counts to reduce the influence of natural radionuclides. However, since they could not be removed entirely, any deviation from these 3σ values was considered to detect artificial radionuclides. As a result, the number of times above the 3σ range was 0.57% of the total.
The timing at which the two methods exceeded the reference values was different. This finding implies that applying a combination of these two methods can reduce the false alarm rate.
