Interpretation of Field Test Data from UF6 Cylinder Accountancy Scales

Uranium enrichment service providers are expanding existing gas centrifuge enrichment plants and constructing new facilities to meet demands resulting from the shutdown of gaseous diffusion plants, the completion of the U.S.-Russia highly enriched uranium downblending program, and the projected global renaissance in nuclear power. The International Atomic Energy Agency (IAEA) is responsible for conducting verification inspections at safeguarded facilities to provide assurance that signatory States comply with their treaty obligations to use nuclear materials only for peaceful purposes. The ability for the IAEA to monitor UF6 cylinder accountancy scales at an enrichment plant would provide independent verification of declared cylinder weights, provide assurance that operations at the plant are ‘normal,’ supply authenticated information that could make safeguards more effective and efficient, and enable information-driven inspections. In the summer of 2009, researchers at Oak Ridge National Laboratory (ORNL) conducted a field test of an accountancy scale monitoring system at the Global Nuclear Fuel (GNF) fuel fabrication facility in Wilmington, North Carolina, owned and operated as a joint venture of General Electric, Hitachi, and Toshiba. Two separate data sets were collected from two scales, one in the receiving area and the other in the dry conversion process area. The data collection spanned six weeks from June 27–August 7, 2009. The data analysis consists of analyzing (1) individual ‘events’, such as check weighs, empty and full cylinder weighs, and operator activities; (2) event statistics; and (3) plant operational behavior. Researchers were able to deduce ‘normal’ operation of the plant from the data collected from the process area, such as that the plant operates continuously, 24 hours per day, 7 days per week, and does not always shut down for holidays (e.g., July 4th). Also, full cylinder weighs occurred in clusters during the daytime shifts, while empty cylinder weighs appeared uniformly throughout the day, consistent with accountancy procedures that call for weighs immediately after a cylinder is detached from the process; check weigh events occurred throughout the day with greatest frequency at the beginning of the inferred work shifts or coincident with empty cylinder weighs.