Year
1979
Abstract
This paper reviews some of the computational tools developed by Lawrence Livermore Laboratory (LLL) for the assessment and design of Material Control and Accounting (MC&A) process monitoring components and illustrates their application to a Pu evaporator/concentrator unit operation. The codes which we have developed include: (1) a general-purpose dynamic simulator for modeling the physical phenomenology of various chemical unit operations and their associated measurement systems, (2) an estimation code for simulating the operation of some modern signal processing algorithms (Kalman filter formulation), and (3) a set of detection algorithms for simulating on-line material loss detection algorithms for simulating on-line material loss detection. These codes can be used to address the issues of on-line material accounting and diversion detection for safeguarding SNM, and specifically with respect to arriving at meaningful performance measures. They can be used to compare state-of-the-art with state-of-the-practice and to study cost benefit tradeoffs. They are capable of treating stochastic models with nonlinear process and measurement dynamics and as a result should provide means for better designs of MC&A process monitoring components.