U.S. Army ADS-79D-HDBK — ADS-79D-HDBK.pdf

Downloaded from https://www.everyspec.com ADS-79D-HDBK The data acquisition system should have a built-in test capability to check the validity of the incoming sensor data. The system should notify maintenance personnel of any data that is suspected of being faulty. The integrity of an engine monitoring system should be ensured by having the ability to detect and flag faulty sensor data. Typical fault detection techniques for sensors include, but are not limited to: a. Out of range – reading is outside the range physically possible for the parameter being measured b. Rate of change – rate of change in the sensor reading exceeds what is physically possible for the parameter c.

Parameter interrelationships – cross checking between redundant sensors or between dissimilar sensors using simple models of the parameter relationships If faulty data are detected, a process should be in place to restore the missing information or to account for the lost data. Linear interpolation between good values is one approach. Another would be to use substitute calculations which may be based on a model using other signals. Wherever processing is carried out, whether onboard or ground-based, the onboard unit should be able to record one or more flights of raw data, depending on scheduled maintenance intervals, for independent processing.

This will provide a routine system check and may be used to investigate suspicious results. Again, segments of raw data that precede anomalies or actual faults detected should be captured and available for downloading to enhance algorithm development and validation. G.7.1.3 Data Transfer. Data transfer should be quick, easy to operate, robust, and user friendly. When damage data or flight data is transferred from the onboard system to the ground system, the system should unambiguously identify the monitored components with the associated data. The engine configuration should be included with the data base of the reduced engine model.

Data transfer should be carried out with a high level of automation. Manual support or even manual transcription should be kept to a minimum. Provisions should be made to ensure data integrity. The downloading device should be able to download and manage data from several aircraft, preferably a whole operational unit, before the operator has to return to the ground station. Data from different engines should be clearly separated and uniquely tagged to avoid misinterpretation and faulty results. In addition, data should be uniquely tagged with an engine run number / time / date.

38 RTO TR 28, ―Recommended Practices for Monitoring Gas Turbine Engine Life Consumption‖, North Atlantic Treaty Organization, Research and Technology Organization, April 2000, pg. 8-8. 126 TABLE G-XI. Typical EMS sensor update rates 38 Vibration Monitored Component Update Rate Compressor disks 1 to 2 Hz Turbine disks 1 to 4 Hz Shafts 1 to 2 Hz Turbine blades 2 to 8 Hz