Evaluating Computational Methods for Modeling Off-Normal Operation of Gas Centrifuge Cascades

This work compares and evaluates different computational approaches for modeling off-normal operation of a gas centrifuge enrichment cascade. The goal of this work focuses on developing the necessary understanding on potential misuse of enrichment cascade to design more effective and efficient international safeguards approaches.While it is straightforward to design an enrichment cascade under ideal conditions as a function of the theoretical feed, product and tails assays, it is very difficult to find reliable information about the behavior of a given cascade when the feed assay does not match the design value. Different methods have been developed to assess the behavior of an enrichment cascade in such circumstances, those methods evaluates the cut, the feed to product, feed to tail and the product to tail enrichment ratio, respectively, alpha, beta and gamma, as a function of the cascade feed assay. As those four parameters depends on each other, determining two of them allows to compute the other.The first approach consists of fixing the cut and alpha recomputing the corresponding assays at each stages of the cascade. The second one maintains the ideal condition of the cascade (alpha and beta fixed across the whole cascade), modifying the cut value at each stage accordingly. Both approaches have been implemented into the Cyclus fuel cycle simulator[1][2]. The third fixes the cut and gamma, using both alpha and beta at each stage as free parameters. The third method has been investigated in [3]. Following a description of each method and an evaluation of differences between each approach, this work compares the results produced by these methods within scenarios involving misuse of enrichment cascades simulated using the dynamic nuclear fuel cycle simulator, Cyclus.Reference:[1] K. HUFF, et al, “Fundamental concepts in the Cyclus nuclear fuel cycle simulation framework,” Advances in Engineering Software (April 2016). [2] M. MCGARRY and B. MOUGINOT, “mbmore”, (Jan 2018) https://doi.org/10.6084/m9.figshare.5097694.v3.[3] M.E. Walker and R.J. Goldston, “Timely Verification at Large-Scale Gas Centrifuge Enrichment Plants”, Science & Global Security, 25, 2, 59-79 (2017).