Flight regimes recognition in actual operating conditions: A functional dataanalysis approach

Helicopters need adequate monitoring to prevent dynamic failures from excessively affecting components'health status, increase the level of safety, and reduce operative costs. Health and Usage Monitoring Systemshave been developed to monitor helicopters during their lifetime in the last few decades. Recent worksdemonstrated that despite analyzing physical components' behavior over time, tracking the regimes performedduring each flight contributes to estimating the aircraft's health and usage status, paving the way for designingaccurate prognostics algorithms. However, today, most regime recognition systems rely on data recordedduring certification flights. It follows that the training regimes differ from the ones proposed in the predictionphase, which are acquired during helicopter actual operating conditions. This affects these recognition systemperformances. Aiming at overcoming this limitation, in this work, we proposed an unsupervised regimesrecognition system capable of better handling the actual helicopter usage spectrum. In detail, we proposeda system based on an unsupervised learning paradigm, which leverages a soft-membership classificationtechnique to account even for mixed regimes and transitions. In addition, the system represents data accordingto functional data analysis theory, which allows for considering the temporal relationship between samples inthe classification process, often neglected in state-of-the-art approaches. The proposed system was tested onexperimental data, collected by Leonardo Helicopter Division, assessing outstanding capabilities in recognizingcorrectly standard and mixed regimes and transients. Also, the presented results demonstrate the approachcapabilities in paving the way for the definition of new regimes, more consistent with the actual helicopterusage spectrum.

Automated summary

This paper presents an unsupervised regime recognition system for helicopters that can better handle the actual usage spectrum. The system demonstrates outstanding capabilities in recognizing standard, mixed regimes, and transients based on experimental data.