Downloaded from https://www.everyspec.com ADS-79D-HDBK In the bottom portion of Figure D-9, the steep increase in CI value between 3 and 4 on the horizontal axis could provide sufficient detection with high confidence. Both CIs demonstrate one reality: the fault progression may result in CI values remaining nearly constant even though the fault is growing; this is clearly not ideal, and an indication that more than one CI may be required for detection with high confidence. 0.24 Feature Value vs. crack length (93/100% torque) 0.22 0.2 0.18 0.16 0.14 0.12 0.1 0.08 1 2 3 4 5 6 7 8 0.06 FIGURE D-9.
Two examples of CI plots to compare detectability For the purpose of this example, we assume that comparison of the CIs selected from Appendix E and the technical literature indicate that this CI has the best available performance in detectability, accuracy, and identifiability for this particular fault. When performance criteria are met with the sample data sets, the selection process shifts to validation of a flight qualified system.
This entails the process of moving the preliminary software code from the laboratory environment to flight qualified hardware for the portions of the process to be accomplished on board, and moving the other portions of the algorithms to the ‗ground station‘ or post flight processing. Once the performance of the algorithms has been validated in this environment, they may proceed to implementation as directed by the aircraft program manager. D.6 APPROACH: CI/HI DEVELOPMENT FOR NEW DEVELOPMENTAL AIRCRAFT D.6.1 Initial situation. The following section provides an example of a new development aircraft which is an evolutionary design from a previous design.
The acquisition strategy and PM guidance mandate the use of CBM for critical systems. The requirements include a target availability of 85% and mean time to repair (MTTR) of under 3 hours. D.6.2 CI development process D.6.2.1 Understand the failure mode. Reliability and Maintainability (R&M) studies typically allocate ―not-mission–capable‖ fractions to various systems based on past practice, modified by new design data. Vendors supplying the new designs have some modeling and testing to substantiate R&M 90