AUTOMATED HEALTH MANAGEMENT FOR GAS TURBINE ENGINE ACCESSORY SYSTEM COMPONENTS

摘要

The authors have developed model-based and data-driven techniques to provide reliable health assessments of hydraulic pumps and valves, which are essential components in aircraft fuel and lubrication systems. As part of this effort, a physical model of an aircraft fuel/lubrication system was created and parameters indicative of pump and valve faults were derived. Data-driven routines were also developed to analyze dynamic pressure signals and detect faults that would be difficult to incorporate into physical models. Evolutionary prognostics methods were also implemented to track faults across feature space and indicate time to failure for each component. The approach was demonstrated using an experimental setup representative of aircraft fuel and lubrication systems. Pump leakage, pump gear damage, and valve blockage were seeded on the setup, and the developed routines were trained with experimental data. The developed routines yielded parameters that were uniquely correlated to particular faults. These parameters allowed wide separation between baseline and faulted cases, yielding negligible missed detection rates for moderate faults and reasonable missed detection rates for an incipient valve blockage fault. The demonstration produced a quantifiable estimate of achievable performance using the hybrid techniques compared to traditional performance monitoring on accessory components.