Dynamic Signal Analysis and Neural Network Modeling for Life Prediction of Flight Control Actuators

Abstract – The authors have developed a Condition-Based Maintenance (CBM) methodology to detect faults and predict failures in flight control actuation systems using dynamic signal analysis and neural network modeling. This advanced processing scheme provides the maintainer of the system with a clear vision of the current actuator health state as well as the useful life remaining. The developed approach utilizes command/response signals and hydraulic pressure data from the actuators to monitor the system and reveal evidence of wear or failure. This evidence is evaluated using a sophisticated automated reasoner that integrates advanced knowledge fusion, classification, and probabilistic failure mode progression (prognostic) algorithms to provide a real-time assessment of the current and future health of the actuator. This methodology is applicable to both fixed-wing and rotorcraft platforms and has been demonstrated using data from F/A-18 stabilator electro-hydraulic servo valves (EHSV). Excellent health state classification was obtained and results are presented. Prognosis was also implemented but data was not available to validate the prediction. Because of the adaptable nature of the core technologies incorporated within this approach, there is significant potential for implementation on a broad range of platforms, specifically helicopters such as the H-60 and H-1 and tilt-rotor aircraft such as the V-22 Osprey.

[1]  R. C. Kryter,et al.  Condition monitoring of machinery using motor current signature analysis , 1989 .

[2]  E. Lavretsky,et al.  Health monitoring of an electro-hydraulic system using ordered neural networks , 2002, Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290).

[3]  H. D. Haynes.  Electrical signature analysis (ESA) developments at the Oak Ridge Diagnostics Applied Research Center , 1995 .

[4]  J. Urnes,et al.  Reconfigurable control and fault identification system , 2004, 2004 IEEE Aerospace Conference Proceedings (IEEE Cat. No.04TH8720).

[5]  br,et al.  How Hums Systems Have The Potential Of Significantly Reducing The Direct Operating Cost For Modern Helicopters Through Monitoring , 1995 .

[6]  P. W. Stevens,et al.  A Multidisciplinary Research Approach to Rotorcraft Health and Usage Monitoring , 1996 .

[7]  Michael J. Roemer,et al.  Assessment of data and knowledge fusion strategies for prognostics and health management , 2001, 2001 IEEE Aerospace Conference Proceedings (Cat. No.01TH8542).

[8]  Miguel,et al.  Neural Network Estimation of Low Airspeed for the V-22 Aircraft in Steady Flight , 2003 .

[9]  K. F. Fraser.  An Overview of Health and Usage Monitoring Systems (HUMS) for Military Helicopters , 1994 .

[10]  Neil Davey,et al.  Time Series Prediction and Neural Networks , 2001, J. Intell. Robotic Syst..

[11]  Matthew J. Watson,et al.  In-line health monitoring system for hydraulic pumps and motors , 2003, 2003 IEEE Aerospace Conference Proceedings (Cat. No.03TH8652).