Facility Scale In-situ Source Localization And Assay Via A Sparse 3He Neutron Detector Array: Enhancing Nuclear Material Control And Accounting In Nuclear Fuel Cycle Facilities

At Los Alamos National Laboratory (LANL), the Dynamic Material Control (DYMAC) project has been established to enhance the LANL Plutonium Facility’s manufacturing, research agility, efficiency, and to improve nuclear security by modernizing, streamlining and optimizing quantitative nuclear material measurements and Nuclear Material Control and Accounting (NMC&A). Challenges such as the complex dynamic nature of the radiation background, due to the continuous movement of nuclear items between glove boxes, can limit the efficacy of in-line NMC&A techniques in a nuclear production environment. In this paper, we develop an approach to in-situ facility scale source tracking to address this dynamic environment by developing a test bed measurement system which consists of a sparse array of 3He neutron detectors located in proximity to an arrangement of glove boxes. Real time continuous neutron count rate data from the detectors are used for tracking the position and neutron emission rate of a number of 252Cf items simultaneously. Several algorithms have been developed to analyze these in-situ measurements including: 1) a template matching algorithm which uses a set of calibration measurements to characterize the radiation field produced by a source without high-fidelity modeling, 2) an iterative deconvolution algorithm which uniquely separates the radiation of each source from their combined background to improve the solution accuracy, and 3) a reduced order model which quickly projects the radiation field produced by the combined sources to the rest of the room. The system’s performance was evaluated on its ability to detect a number of 252Cf items emitting neutrons with their neutron emission intensity and position within the test bed facility. Results such as minimum detectable activity, localization spatial resolution, and neutron emission rate dynamic range will be presented. Acknowledgments. This work was supported by the Laboratory Directed Research and Development (LDRD) program at Los Alamos National Laboratory.