Advanced Analysis of Nuclear Waste Storage Health

The potential longevity and number of dry storage systems employed for storage of used nuclear fuel necessitate a deliberate consideration of long-term data management approaches. Necessary considerations include approaches to ensure archival of data over multiple centuries and data fusion methods to facilitate decision making based on multiple streams of information. Also important is establishing an advanced data analyses methodology for characterizing the physical condition of dry-storage systems in order to understand how their condition changes over time. The proposed "Cask Analysis and Degradation Evaluation System CaskADES) tool will combine information from various data streams to help assess Nuclear Waste container/cask material degradation status. The data analysis methodology will ensure nuclear waste containers or casks structural integrity by detecting and predicting early indications of degradation from multiple heterogeneous sources of information. The system will provide status indicators/features using combined model-based and data-driven approaches from a structural and material point of view. These indicators/features will be fused using hybrid modeling based on Bayesian theory to provide an overall safety of a container or a cask. In Phase I, we will make an assessment of the type of data that is currently available and use it demonstrate the feasibility of the CaskADES health analysis architecture. Recommendations for additional instrumentation/monitoring and models will be made to improve the predictive capability of the tool. Additional data, including from the planned DOE-sponsored High Burnup Fuel Demonstration Program, will be obtained to validate the tool. The resulting technology from this effort can be applied to wide range of applications employed for monitoring of machinery and structural components. The commercial intent is to produce a beta-tested software tool for commercial markets. Potential commercial applications include diagnostics/prognostics of machinery and structural health monitoring.