Monitoring System for Storm Readiness and Recovery of Test Facilities: Integrated System Health Management (ISHM) Approach

Severe weather events are likely occurrences on the Mississippi Gulf Coast. It is important to rapidly diagnose and mitigate the effects of storms on Stennis Space Center's rocket engine test complex to avoid delays to critical test article programs, reduce costs, and maintain safety. An Integrated Systems Health Management (ISHM) approach and technologies are employed to integrate environmental (weather) monitoring, structural modeling, and the suite of available facility instrumentation to provide information for readiness before storms, rapid initial damage assessment to guide mitigation planning, and then support on-going assurance as repairs are effected and finally support recertification. The system is denominated Katrina Storm Monitoring System (KStorMS). Integrated Systems Health Management (ISHM) describes a comprehensive set of capabilities that provide insight into the behavior the health of a system. Knowing the status of a system allows decision makers to effectively plan and execute their mission. For example, early insight into component degradation and impending failures provides more time to develop work around strategies and more effectively plan for maintenance. Failures of system elements generally occur over time. Information extracted from sensor data, combined with system-wide knowledge bases and methods for information extraction and fusion, inference, and decision making, can be used to detect incipient failures. If failures do occur, it is critical to detect and isolate them, and suggest an appropriate course of action. ISHM enables determining the condition (health) of every element in a complex system-of-systems or SoS (detect anomalies, diagnose causes, predict future anomalies), and provide data, information, and knowledge (DIaK) to control systems for safe and effective operation. ISHM capability is achieved by using a wide range of technologies that enable anomaly detection, diagnostics, prognostics, and advise for control: (1) anomaly detection algorithms and strategies, (2) fusion of DIaK for anomaly detection (model-based, numerical, statistical, empirical, expert-based, qualitative, etc.), (3) diagnostics/prognostics strategies and methods, (4) user interface, (5) advanced control strategies, (6) integration architectures/frameworks, (7) embedding of intelligence. Many of these technologies are mature, and they are being used in the KStorMS. The paper will describe the design, implementation, and operation of the KStorMS; and discuss further evolution to support other needs such as condition-based maintenance (CBM).