Prognostics technology was recently demonstrated to Space Shuttle operations team members and Ground Operations launch control system representatives at Kennedy Space Center (KSC). Focusing on a pneumatic valve in the Space Shuttle liquid hydrogen refueling system, and using historical valve degradation data extended with simulated data to the end of valve life, prognostic algorithms automatically determined the health state, damage progression rates, and the remaining useful life of the valve as it cycled.
The algorithm employs a model-based approach in conjunction with a particle filter and measured-valve timing data for the remaining life prediction. Algorithm results were displayed on a prototype prognostics interface that is part of a comprehensive simulation of cryogenic propellant refueling, providing damage progression estimates, measured and predicted valve open and close times, and remaining useful life assessments with confidence bounds. The work was developed as a technology proof of concept for the Fault Detection, Isolation, and Recovery (FDIR) project, which provides systems health management to ground support of launch control systems.
BACKGROUND: The Fault Detection, Isolation, and Recovery Task focused on maturing, integrating, and certifying health management techniques and tools for use with complex subsystems in space vehicles and ground support systems. The primary goal of the FDIR Task was to formulate a certifiable configuration and architecture for providing the health and status of ground systems in order to decrease the time it takes for ground personnel to recognize and respond to anomalous behavior, isolate faults, and recover from failures. By decreasing isolation and recovery times, the launch availability of the ground systems would be improved, allowing greater chances for a successful launch.
As a part of the FDIR Task, a study to determine the feasibility of developing and certifying prognostics technology for launch vehicle ground support systems was also explored in an attempt to avoid or lessen the impact of hardware failures during operations. Prognostics technology, if successful, can significantly lower operating costs while increasing system safety, reliability, and availability.
NASA PROGRAM FUNDING: This work was funded under the Fault Detection, Isolation, and Recovery Task, which was managed under the Exploration Systems Mission Directorate’s Exploration Technology Development Program and the Intelligent Systems Health Management Project.
PROJECT TEAM: Matthew Daigle (UC Santa Cruz), Kai Goebel (NASA Ames), and Barbara Brown (NASA Ames @ KSC)
Contact: Matthew Daigle