Innovation in FDIR for aircraft and flight-critical systems

Auteurs : Ali Zolghadri
Publication MEA2017
contenu protégé  Document accessible sous conditions - vous devez vous connecter ou vous enregistrer pour accéder à ou acquérir ce document.
- Accès libre pour les ayants-droit


Innovation in FDIR for aircraft and flight-critical systems


1.73 Mo


Creative Commons Aucune (Tous droits réservés)





<resource  xmlns:xsi=""
        <identifier identifierType="DOI">10.23723/16236/18682</identifier><creators><creator><creatorName>Ali Zolghadri</creatorName></creator></creators><titles>
            <title>Innovation in FDIR for aircraft and flight-critical systems</title></titles>
        <resourceType resourceTypeGeneral="Text">Text</resourceType><dates>
	    <date dateType="Created">Mon 13 Feb 2017</date>
	    <date dateType="Updated">Mon 13 Feb 2017</date>
            <date dateType="Submitted">Mon 18 Feb 2019</date>
	    <alternateIdentifier alternateIdentifierType="bitstream">9c3f5f33cfca469cba227ba2c34d20dd2fb8ea1e</alternateIdentifier>
            <description descriptionType="Abstract"></description>

1 Innovation in FDIR for aircraft and flight-critical systems Ali Zolghadri Bordeaux University / CNRS  IMS lab France MEA 2017, 1-2 February, Bordeaux, France What is FDIR ? 3 SPACE SHUTTLE DISCOVERY J.L. Chrétien  Astronaute NASA / ESA FDIR 09/02/2017 5  Fault Detection & Diagnosis (FDD): Detect, Isolate and Estimate the severity of a fault  Fault Tolerant Guidance (FTG): if the available on-board control resources are limited, FTC would not be sufficient >> change the objectives Reconfiguration Level 2: reshaping outer guidance loops  Fault Tolerant Control (FTC): continue to “control” the faulty system: provide, at worst, a degraded level of performance in the faulty situations Reconfiguration Level 1: reshaping inner piloting loops Recovery FDIR FDD&FTC/ FTG Consolidated flight data SENSORS: Flight parameters  Air Data Inertial Reference System ACTUATORS: Propulsion system  Engines / Control surfaces Computed control signals GUIDANCE: Where am I going ? Required changes in velocity Required attitude … CONTROL: How do I get there ? Estimated states NAVIGATION: Where am I ? Fault Monitoring and Management 7 The theory offers a huge number of various designs, techniques and methods related to model-based FDIR. Moreover, many successful “demonstrations” exist… Aircraft and flight-critical applications provide numerous grounds where unconventional, smart and innovating FDIR is needed… However, today, few real model-based applications* can be identified… *Transfer of knowledge resulting in tangible and marketable aerospace technologies which can generate economic added value and benefits to society. Conventional versus advanced model-based FDIR 09/02/2017 8 Question (President): My question is simple: How did we go so wrong ? Answer (AG): I discovered a flaw in the model that I perceived is the critical functioning structure that defines how the world works … we went wrong because models, as we currently employ them, are not able to capture the full array of governing variables in extreme situations that drive global economic reality… Every model-based design/development for flight-critical systems should, of course, provide good “average” performance … It is necessary, but it is not sufficient at all ! The sizing element is the achievable performance, robustness and reliability in extreme, unusual and non-standard flight situations. It is all about model and model-based design / model-driven development Real world is not deterministic: An aircraft is a cyber-physical system combining physical dynamics with computational processes: multiple behavioral modalities interacting with each other that can change with context … In Engineering science, we love deterministic models as they can lead to deterministic model-based systems for monitoring, fault tolerance, reconfiguration, prediction… The issue is how to reconcile Deterministic Models with a Nondeterministic World 09/02/2017 11 A successful example: Model-based monitoring system for Airbus A350 09/02/2017 12 Aircraft control surfaces Two adjacent dissimilar actuators for moving a single control surface on the A380 On the left: an EHA, and on the right a conventional hydraulic actuator. Flight Control Law K COMmand MONitoring (Command) Actuator Flight Control Law Analogic Input Analogic Input Analogic Output Control surface sensor Analogic Input Rod sensor Servo - Valve sensor Monitored signal + Decision making Flight Control Computer (FCC) 09/02/2017 ∆(amplitude) ↔ weight saving  Structural local reinforcement 13 If not detected and not passivated, must be considered for load computations: 09/02/2017 14 The story begins with (2007): Actuators Servo command - Engines Sensors Flight mechanic Inboard computer Flight control laws Pilot inputs Evaluations on Aircraft Airbus Benchmark & In-flight data records… Basic research & lab investigations… 09/02/2017 15 Certification to fly: Europe: European Aviation Safety Agency (EASA): Sep. 2014 USA: Federal Aviation Administration (FAA): Nov. 2014 The story ends with: Entry into service and commercial flight (15-01-2015) 16 A350 FCC V&V activities Ground / Flight tests and evaluation: Actuator testbed platforms Aircraft 0 Aircraft 1 -3 -2 -1 0 1 2 0 1 2 3 0 0.2 0.4 0.6 q2 q1 =63.0957 q2 =0.1 min =0.26839 q1 Design optimisation Model-based design & development 09/02/2017 17 2007-2009: Ideas and concepts, design and theoretical developments, publications/patents, high fidelity simulations, tuning, proof of concept … TRL1–3 2012-2014: Flight V&V: A380 flight tests, adaptation to A350, tuning / initializations, extreme flight conditions, final implementation in FCC … TRL6–8 September 30, 2014: Certification, European authorities (EASA) November 12, 2014: Certification, US authorities (FAA) TRL9 January 15, 2015: First commercial flight of A350 XWB (Doha–Frankfort, Qatar Airways) 2009-2012: Ground V&V: Evaluation on Airbus testbed platforms, actuator bench, System Integration Bench , iron bird… TRL4–5 ≈8years 09/02/2017 18 Another example: under TRL5 investigations at Airbus‡  CR2 decision test† Flight Control Law K COMmand MONitoring (Command) Actuator Flight Control Law Analogic Input Analogic Input Analogic Output Control surface sensor Analogic Input Rod sensor Servo - Valve sensor Monitored signal + Decision making Flight Control Computer (FCC) RLS (k) CR2 Monitoring system α λ Runaway/jamming detection Tuning parameters - Estimation of parametric directions sensitive to faults - CR2 decision test to detect and confirm the presence of faults ‡ Zolghadri et al. “AIAA Journal of Aircraft”, 2016 † Zolghadri A. “IEEE Transactions on Automatic Control”, 1996 Thank You Minds are like parachutes, they work best when open A Scottish whisky distiller