In the PAHMIR (Preventive Aircraft and Health Monitoring) project pattern recognition and signal analysis is used to support and simplify the monitoring of complex aircraft systems. The parameters of the signal analysis need to be chosen specifically for the monitored system to get the best pattern recognition accuracy. An optimization process was developed that uses global heuristic search and optimization to find a good parameter set for the signal analysis. The computed parameters deliver slightly (one to three percent) better results than the ones found by hand. In addition it is shown that not a full set of data samples is needed. Genetic optimization showed the best performance. The results show that an optimization can increase the performance of the signal analysis and pattern recognition. However the increase is less than 5 %. Still it is possible to push the accuracy up to 99.4 %. Even the short searches with a small population showed a good performance compared to choosing parameters by hand, which is nearly equal to chose a random parameter set. One goal of the project to reduce the amount of expert knowledge to use the system is achieved. The concept works well if a significant number of data samples are available. Another advantage of the concept is that it can be parallelized without much work, if a genetic algorithm is used. The members of the population can be spread over the available processor. With parallelization it is possible to reduce the computation time a lot and a much larger space can be searched in the same time.
Parameter optimization for automated signal analysis for condition monitoring of aircraft systems
Parameteroptimierung für die automatische Signalanalyse bei der Zustandsüberwachung von Flugzeugsystemen
2011
12 Seiten, 3 Bilder, 5 Tabellen, 3 Quellen
Conference paper
Storage medium
English
AIRCRAFT TIRE CONDITION MONITORING SYSTEMS
SAE Technical Papers | 2009
AIRCRAFT SYSTEMS AND METHODS WITH AUTOMATED RUNWAY CONDITION INFORMATION
European Patent Office | 2024
|AIRCRAFT SYSTEMS AND METHODS WITH AUTOMATED RUNWAY CONDITION INFORMATION
European Patent Office | 2024
|Condition Monitoring of Aircraft High Lift Systems
SAGE Publications | 1990
|