Detection of oscillatory actuator failures in passenger airliners

Els, Dylan (2019-04)

Thesis (MEng)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: This project investigates and develops techniques to detect oscillatory failure cases (OFCs) in aircraft control surface actuators. Oscillatory failures induce additional loads on the structure of the aircraft, requiring additional structural support to withstand these loads, increasing the overall mass of the aircraft. If oscillatory failures can be detected and pacified quickly, then the additional structural support would not be required, and the mass of the aircraft can be reduced, resulting in improved fuel efficiency and aircraft performance. Oscillatory failure case (OFC) detection is performed by evaluating the difference (residual) between the measured behaviour of the real actuator and the simulated behaviour of a fault-free analytically redundant actuator model running in parallel with the real actuator. An OFC detection system must generate a residual signal using the analytically redundant actuator model, and evaluate the residual signal to determine whether an oscillatory failure is present. The challenge for the residual evaluation stage is to distinguish between the components of the residual signal resulting from modelling uncertainty and sensor noise, and the components resulting from an actual oscillatory failure case. The OFC detection system must detect oscillatory failures within a maximum allowable detection time, but must not produce false alarms. Five different oscillatory failure detection techniques are investigated and developed, namely oscillation counting, integrated absolute error (IAE), discrete Fourier transform (DFT), multi-window Fourier transform (MWFT), and phase-locked loop (PLL) detection. Oscillation counting is an existing OFC detection technique that was developed by Goupil [1] and is currently in service on the Airbus A380 passenger airliner. The other four techniques are new OFC detection techniques that are developed in this project. A simulation framework is created to serve as a testbed for the training and testing of the different OFC detection techniques. The simulation framework contains models for the physical actuator, the analytically redundant actuator, the oscillatory failures (both liquid and solid failures), the flight control system, and the aircraft longitudinal dynamics. The simulation models the aircraft’s response to an oscillatory failure, since it affects the performance of the OFC detection. The five OFC detection techniques are trained and rigorously tested using training and testing data generated with the simulation framework. The detection thresholds for each technique are “trained” on fault-free data to determine the lowest detection thresholds that do not produce false alarms. The detection techniques are then tested using testing data to determine the smallest amplitude oscillatory failure that each technique can detect within the specified maximum allowable detection time. The number of false alarms for each technique is also determined. The results show that DFT, MWFT, and the PLL outperform oscillation counting and IAE by detecting smaller amplitude oscillatory failures and with shorter detection times, with MWFT providing the most promising results. However, oscillation counting and IAE are the most computationally efficient techniques, while DFT, MWFT, and PLL are more computationally expensive. Overall, the multi-window Fourier transform (MWFT) technique is the recommended approach for OFC detection, offering the best detection performance with only a small increase in computational complexity.

AFRIKAANSE OPSOMMING: Hierdie projek ondersoek en ontwikkel tegnieke om ossillatoriese faling gevalle (OFGs) in vliegtuig beheeroppervlak aktueerders te bespeur. Ossillatoriese falings induseer bykomende ladings op die struktuur van die vliegtuig, en vereis dus bykomende strukturele ondersteuning om hierdie ladings te weerstaan, wat die algehele massa van die vliegtuig verhoog. Indien ossillatoriese falings bespeur en vinnig gepassifiseer kan word, dan sou die bykomende strukturele ondersteuning nie benodig word nie, en die massa van die vliegtuig sou verminder kon word, wat sou lei tot verbeterde brandstofverbruik en werkverrigting. Ossillatoriese faling geval (OFG) bespeuring word uitgevoer deur die verskil (residu) te evalueer tussen die gemete gedrag van die werklike aktueerder en die gesimuleerde gedrag van ’n foutvrye analities-oortollige aktueerder model wat in parallel met die werklike aktueerder uitvoer. ’n OFG bespeuringstelsel moet ’n residu sein genereer deur gebruik te maak van die analitiese-oortollige aktueerder model, en moet die residu evalueer om te bepaal of daar ’n ossillatoriese faling teenwoordig is. Die uitdaging vir die residu evaluasie stadium is om te onderskei tussen die komponente van die residu sein wat afkomstig is van model onsekerheid en sensor ruis, en die komponente wat afkomstig is van ’n werklike ossillatoriese faling geval. Die OFG bespeuringstelsel moet ossillatoriese falings bespeur binne ’n maksimum toelaatbare bespeuringstyd, en moet nie vals alarms gee nie. Vyf verskillende ossillatoriese faling bespeuringstegnieke word ondersoek en ontwikkel, naamlik ossillasie telling, geïntegreerde absolute fout (IAE), diskrete Fourier transform (DFT), multi-venster Fourier transform (MWFT), en fase-sluit lus (PLL) bespeuring. Ossillasie telling is ’n bestaande OFG bespeuring tegniek wat ontwikkel is deur Goupil [1] en word tans gebruik op die Airbus A380 passassiersvliegtuig. Die ander vier tegnieke is nuwe OFG bespeuringstegnieke wat ontwikkel is in hierdie projek. ’n Simulasie raamwerk is geskep om te dien as ’n toetsplatform vir die opleiding en toets van die verskillende OFG bespeuringstegnieke. Die simulasie raamwerk bevat modelle vir die fisiese aktueerder, die analities-oortollige aktueerder, die ossillatoriese falings (beide vloeibare en soliede falings), die vlugbeheerstelsel, en die vliegtuig se longitudinale vlugdinamika. Die simulasie modelleer die vliegtuig se reaksie op die ossillatoriese faling, aangesien dit die prestasie van die OFG bespeuring beïnvloed. Die vyf OFG bespeuringstegnieke is opgelei en volledig getoets deur gebruik te maak van opleiding en toets data wat genereer is met die simulasie raamwerk. Die bespeuring drempels vir elke tegniek is “opgelei” op foutvrye data om te bepaal wat die laagste bespeuringsdrempel is wat nie vals alarms gee nie. Die bespeuringstegnieke is dan getoets op toets data om te bepaal wat die kleinste amplitude ossillatoriese faling is wat elke tegniek kan bespeur binne die maksimum toelaatbare bespeuringstyd. Die aantal vals alarms vir elke tegniek is ook bepaal. Die resultate wys dat die DFT, MWFT, en PLL tegnieke oortref die ossillasie telling en IAE tegnieke deur kleiner amplitude ossillatoriese falings te bespeur in korter bespeuringstye, met die MWFT wat die mees belowende resultate lewer. Die ossillasie telling en IAE tegnieke bly egter die mees berekeningsdoeltreffende tegnieke, terwyl die DFT, NWFT, en PLL meer berekeningskoste dra. Algeheel, word die multi-venster Fourier transform (MWFT) tegniek aanbeveel as die voorkeurtegniek, omdat dit die beste bespeuringsprestasie bied met net ’n klein verhoging in berekeningskoste.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/106204
This item appears in the following collections: