Applications of Virtual Monitoring of Loads to Engineering Decision Making

A virtual tail rotor torque load model was applied to over 500,000 flight hours of operational parametric data from the fleet of S-92® rotorcraft and was used to develop design load spectrum prorates for the design, qualification, and certification of drivetrain components. Fleet parametric data was recorded by an onboard Health and Usage Monitoring System and archived at Sikorsky Aircraft for all active tail numbers since inception. Model training and blind validation was performed using production structural flight load survey test data. To extract meaningful and reliable loads information from such a large fleet database, a multistep process was developed to ensure data quality, identify questionable torque events for engineer review, and reduce the calculated time histories into individual tail number torque spectrums. All tail number torque spectrums were combined into a single composite worst case torque spectrum, which was compared against the existing design load spectrum to identify areas where the design load spectrum is overly conservative and where usage prorates could be developed. A usage monitor reliability factor was then applied against the fleet composite worst case load spectrum to ensure the reliability of the resulting usage prorates. The newly developed usage prorates increased the critical component retirement time by more than a factor of 3. As the first in-depth application of virtual monitoring of loads to affect the design or qualification process, this project paves the way for more sophisticated fleet and tail number based credits and design methods that more fully exploit the ability to monitor key loads.