Surveillance plays a crucial role in safeguards. Reviewing surveillance data requires a significant number of inspection manpower. As the number of surveillance cameras increases, the demand for such manpower is expected to grow even more. Recently, in the field of security, there has been a development of deep learning models that automatically detect abnormal events from video images, and their usage is expanding. In this study, we used an AutoEncoder-based semi-supervised learning model, which can detect unexpected abnormal events, to detect anomalies in the UCSDped2 dataset and in simulating safeguards-related event videos taken at Dry Mockup facility of KAERI. To improve the model performance, we transformed the video images into two parts: the appearance part, which are sequences of video image frames, and the motion part, which are the pixel value differences of consecutive video frames. In addition, we added memory module to the bottle neck of the AutoEncoder model, and skip connection to enhance the model performance. To evaluate the model performance, we proposed a new evaluation index, which is adequate to the video images of safeguards surveillance in addition to the widely used AUC (Area Under the ROC Curve).

The nuclide management technology for separating high-heat generating/high-mobility/long-lived nuclides from high-level wastes based on the chemical reactions is under development. In order to secure the reliability of nuclear non-proliferation and to implement the effective safeguards, it is necessary to consider the safeguards from the conceptual design phase of the novel technologies. However, there was no experience and research on safeguards for the chemical reaction based nuclide management technology. In order to development the available monitoring techniques for the safeguards of nuclide management technology, the possible diversion scenarios were developed and the material flows of major nuclear materials were analyzed according to the various diversion strategies for each unit process in this study. The diversion strategies in this study is limited to the diversion of nuclear materials according to the change of operational parameters (temperature, chemical reagents, pressures, etc). The nuclear material distribution behaviors under the abnormal conditions were analyzed and compared with normal conditions using the HSC Chemistry. The results will be used to determine the proper signals and feasible techniques to monitor the abnormal operations.

Pyroprocessing is a promising technology for managing spent nuclear fuel. The nuclear material accounting of feed material is a challenging issue in safeguarding pyroprocessing facilities. The input material in pyroprocessing is in a solid-state, unlike the solution state in an input accountability tank used in conventional wet-type reprocessing. To reduce the uncertainty of the input material accounting, a double-stage homogenization process is proposed in considering the process throughput, remote controllability, and remote maintenance of an engineering-scale pyroprocessing facility. This study tests two types of mixing equipment in the proposed double-stage homogenization process using surrogate materials. The expected heterogeneity and accounting uncertainty of Pu are calculated based on the surrogate test results. The heterogeneity of Pu was 0.584% obtained from Pressurized Water Reactor (PWR) spent fuel of 59 WGd/tU when the relative standard deviation of the mass ratio, tested from the surrogate powder, is 1%. The uncertainty of the Pu accounting can be lower than 1% when the uncertainty of the spent fuel mass charged into the first mixers is 2%, and the uncertainty of the first sampling mass is 5%.

한국원자력연구원은 IAEA에서 권고하고 있는 안전조치기반설계(SBD)에 입각하여 파이로 안전조치 기술을 개발하고 있다. 한국원자력연구원은 파이로 안전조치접근방안 개발을 위한 IAEA 회원국지원프로그램(MSSP)을 수행하였다. IAEA 회원국 지원프로그램을 통하여 기준파이로시설(REPF) 개념을 설계하고, 이 시설에 대한 안전조치시스템을 개발하였다. 최근에 기 준파이로시설은 용량이 증대된 REPF+로 업데이트 되고 있다. 핵물질계량관리시스템 성능평가를 위하여 전산코드 PYMUS 를 개발하였으며, PYMUS는 전용탐지획률 통계평가 방안을 포함하여 업그레이드하고 있다. 파이로 입력물질 계량을 위한 비파괴분석장비로 ASNC가 개발되고 있으며, 파이로 출력물질인 U/TRU 잉곳을 계량하기위한 비파괴분석장비로 HIPAI가 개발되고 있다. 또한 컴프톤 억제 감마선분광기술, LIBS 기술, 균질화 공정의 샘플링 오차에 대한 평가도 진행 중이다. 이러 한 노력들은 국내에서 선진핵주기기술 실현에 크게 기여할 것이다.