AN EXPERT SYSTEM BASED FAULT DETECTION AND ISOLATION SYSTEM TO MONITOR WEIGHT AND RADIATION SENSORS FOR INVENTORY MANAGEMENT

The Continuous Automated Vault Inventory System (CAVISTM) is a system designed to continually monitor the status of stored nuclear materials (SNM) at the Oak Ridge based Y-12 facility. CAVIS consists of an integrated package of low-cost weight and radiation sensors designed to continuously monitor the attributes of stored nuclear materials. The CAVIS system detects “changes-in-state” of the material's physical attributes and generates an appropriate alarm. Unfortunately, these types of monitoring systems can be subject to events that cause false alarms that do not coincide with the removal of nuclear material, but if not quickly reconciled may initiate an expensive and disruptive operational response. These false alarms may be due to the random stochastic nature of the measurements, due to failing components, or due to external radiation sources. The University of Tennessee has developed a monitoring system for CAVIS, which eliminates these types of alarms and their costly responses. The system merges advanced statistical algorithms to extract features related to changes in the CAVIS sensors with an expert system that forms a hypothesis on the root cause of any anomaly. This presents the development and integration of feature extraction techniques and expert diagnostic systems for monitoring stored nuclear materials.