On Health Monitoring

Abstract

Over the past twenty years, the health monitoring community has vastly extended its capability to monitor systems for operational health. Studies on various systems have revealed means of deducing the health of their parts such as specific engine components, traction and control systems including vehicle monitoring, aircraft parts such as landing gear, and many more. In most cases, researchers mapped the signatures of known areas of concern and determined methods to detect these signatures as they occurred. In safety critical applications, the emphasis became development of early detection techniques for termination of operation of the system before severe loss occurred. Existing health monitoring systems include spacecraft monitoring systems, aircraft in-flight monitors, and ground vehicle monitors, each designed to identify and notify the vehicle operation team of existing or pending maintenance and safety issues as they develop. In some systems, such as Boeing aircraft, health monitors allow the vehicle to call ahead to a ground station while en-route to report upcoming necessary maintenance based on known issues. Through the years of development, while health monitor systems have grown more complex and capable, the method of creating these systems has remained somewhat simplistic throughout the industry, using bottom up approaches or ad-hoc implementation in the design of these systems. In this paper, we will discuss another approach. We discuss a top-down functional health monitor design approach. We show that alternate methods are incapable of determining a provably closed-form and tractable health monitor system, and that application of this technique can help eliminate important oversight that threatens the safety of a vehicle. The result of its application is a far safer vehicle through a methodical application of this approach that addresses the various sources of loss within the system.