Air vehicle diagnostic system rotor head load and global fault evaluation as installed on the SH-60F helicopter

This study covers SH-60F rotor head load estimation, track and balance fault detection, and degraded component diagnostics of the Air Vehicle Diagnostic SystemAdvanced Technology Demonstration. The purpose of this study was to acquire load data from selected,SH-60F main rotor components; provide "proof of concept" for the use of neural networks in load estimation; acquire SH-60F main rotor track and balance and vibration data in nominal (no known defect) condition and with intentionally induced known out of track and balance conditions in the main rotor; and acquire SH-60F main rotor track and balance and vibration data in nominal condition and with degraded rotor head components at varying flight conditions. The method of data collection was a survey recorder which houses a memory unit, a neural network and the maneuver recognition algorithms. The memory unit was used to record the load estimation,maneuver recognition algorithms, track and balance, and main rotor degraded component detection system results. Data signals are received directly from the aircraft 1553 databus, existing aircraft systems, and special sensors designed for each task. Strain gauges placed at four different rotor head locations to provide truth data for comparison with the neural network data. The instrumentation monitors 37 parameters for three testing phases totaling over 50 flight hours. Results indicate that the neural network estimation are within 10 percent of the actual strain gauge loads and that the main rotor track and balance capability of the system was able to quickly and accurately track the rotor system and lower vibrations to acceptable levels. The main rotor head fault detection data provided further support for the use of neural networks in vibration analysis.