Presentation on theme: "1 Opportunities and Challenges in Damage Prognosis for Materials and Structures in Complex Systems AFOSR Discovery Challenge Thrust (DCT) Workshop on Prognosis."— Presentation transcript:
1 Opportunities and Challenges in Damage Prognosis for Materials and Structures in Complex Systems AFOSR Discovery Challenge Thrust (DCT) Workshop on Prognosis of Aircraft and Space Devices, Components, and Systems Cincinnati, OH 19-20 February 2008 Jim Larsen, Reji John, and Eric Lindgren AFRL/RX Materials & Manufacturing Directorate Wright-Patterson Air Force Base, OH
2 AFRL Focused Long Term Challenge (FLTC#8) on Affordable Mission Generation and Sustainment Engine Life Management DARPA program on Prognosis Long-term needs for basic research in Prognosis Aircraft Prognosis Challenges and Opportunities Supported by: AFRL; AFRL/RX AFOSR Structural Mechanics (Dr. Victor Giurgiutiu) DARPA Prognosis (Dr. Leo Christodoulou) Outline
3 Affordably Maximize Mission Capability, Readiness, Reliability, and Maintainability of Current and Future Fleets Provide Real-time Total Weapon System Health Status Predict Any System’s Mission Capability Autonomically Reconfigure Systems for Any Damage Condition Proactively Maintain Readiness Design for Integrated System Life Cycle Management & Intrinsic Reliability Design PredictSense Reconfigure Maintain Actual Predicted.............................................. Focused Long Term Challenge (FLTC #8) Affordable Mission Generation & Sustainment
4 Provide Real-time Total Weapon System Health Status Predict Any System’s Mission Capability Design for Integrated System Management & Intrinsic Reliability Autonomically Reconfigure Systems for Any Damage Condition Proactively Maintain Readiness 202520152005 Technologies Systems/Sub-Systems READINESS AFFORDABILITY SUSTAINABILITY SAFETY Conventional Fully Autonomous Aircraft Engines Missiles Spacecraft Electronics Spiral transition to legacy and future Air and Space systems Systems Design Technology Validation and InsertionBreakthrough Science & Technology Spiral Near-Term Technology Transition and Requirements Definition 8.1 8.2 8.3 8.5 8.4 Information Architecture Discovery Challenge Thrust True Game Changer Integrated Systems Health Management
5 Turbine Engine Life Management Driven by Uncertainty Database: Mission History and Design Yes Service NO Retire Decision Empirical Criteria (Cycles or Flight Hours) Inspect Usage (Duty Cycles) Failure Occurrences “Book Life” Today Book life: 1 out of 1000 damaged
6 Life Management of Turbine Engine Disks Condemned Engine Disks
7 Engine Structural Integrity Program (ENSIP) Design Philosophy Low-Cycle-Fatigue Design Criteria (safe life) Based on statistical lower bound 1 in 1000 components predicted to initiate a 0.8 mm crack Damage-Tolerant Design Criteria (fracture mechanics) Deterministic 1 or 2 safety inspections during service life Both design criteria are met at all critical locations on a component log Life (e.g. Cycles or TACs) Usage (e.g. Stress) Typical Mean Lower Bound Cycles (or Equivalent) Crack Length a i a C a *
8 050,000 Failure Occurrences Usage (Duty Cycles) 3 times useful life 50% OK 2 times useful life ~ 86% OK 1.5 times useful life ~ 97% OK Perceived useful life > 99.9% unfailed Log normal distribution viewed on a linear scale ~ one order of magnitude assumed for ±3 Median at 24,000 cycles, - 3 at 8,000 cycles Throw away 1000 components to remove the unknown one that is theoretically predicted to be in a “failed state” Current Safe-Life Approach for Low Cycle Fatigue 999 components are considered “not ready” Vast Majority of Life is Wasted 999 components are considered “not ready” Vast Majority of Life is Wasted
9 Yes Service NO Retire Usage (Duty Cycles) Failure Occurrences “Book Life” Today Prognosis will Enable Transformation in Asset Management Database: Mission History, Maintenance, Life Extension, and Design Prognosis Failure physics, damage evolution, predictive models State Awareness Interrogation Prognosis Translates Knowledge and Information Richness to Physical Capability Reduce and Manage Uncertainty “Book Life” Today “Book Life” Tomorrow Dr. Leo Christodoulou
10 Reduced logo Increasing Age and Degradation Mission Requirement Dislocation density saturation Microcrack formation Local heating / hotspot Macrocrack formation Changes in blade-tip timing/displ. Vibration changes Degraded module efficiencies (temperatures, pressures, speeds) Depot level inspections (eddy current, FPI, etc.) Prognosis: Fatigue Damage Characteristics
11 Reduced logo Increasing Age and Degradation System Monitoring Continuous Rate of damage progression When CAPABILITY approaches MISSION REQUIREMENT Short-term asset management (months / weeks) Material Monitoring Global characteristics / continuum damage Identify tails of life distribution Long-term asset management Mission Requirement Component Monitoring Scheduled inspections Looking for local damage / cracks Mid-term asset management (years) Prognosis: Health Monitoring & Asset Management
12 Needs for Basic Research in Prognosis Reliable, cost effective, methods for characterizing the current state and remaining lifetime of components Damage precursor and cracking detection and signature analysis Microstructural characterization and analysis methods for defining life-limiting material and component features Effects of microstructure on uncertainty in life prediction under service conditions – Focus on life-limiting behavior (worst-case behavior) – Fatigue, creep, environment, mission loading Develop efficient approaches to demonstrate and validate new technologies for life management – Efficient validation methods Key to success is transition to service
Typical Aircraft Geometry Approximately 80% of aircraft structure that needs to be inspected are either: –Multi-layered structures, or –Vertical stiffeners with holes Focus on detection of fatigue cracks –Additional factors will affect detection of Corrosion Stress Corrosion Cracking Damage in Non-metallic Material Ref: QNDE 2006: Aging Aircraft NDE: Challenges and Opportunities, Lindgren et.al.
Multi-layer Structures Can become very complex In this example the objective is to detect fatigue cracks in Wing Skin (red) without disassembly Avoid disassembly due to extra time, cost, and risk of additional damage Y Rib (Y)–0.310” 7075-T73 Al Shim– 0.094” Al Laminate Spar Cap (Bl)–0.344” 2024-T3511 Al Skin (Red)–0.300” 2024-T3511 Al Shim– 0.125” Al Laminate Lt G Attach Fitting (Lt G)–0.320” 4340 Steel Shim– 0.094” Al Laminate Longeron(DkG)G)–0.200” 9Ni-4Co-.03C Steel Layer: H G F E D C B A Holes 2,4 Holes 1,3 Ref: QNDE 2006: Aging Aircraft NDE: Challenges and Opportunities, Lindgren et.al.
Challenge: Defect Characterization in Two Layer Wing Structure Desired: flaw location, orientation, and size Typical flaw detection requirements*: –Each layer can be 0.10" to 0.40" thick (total thickness of 0.20" to 0.80") –1000 to 10000 fasteners –Detect 0.07" corner cracks around fasteners at each surface of each layer –90/95 POD for detection –False call rate between 0% and 1.0% –Inspection process: must be reliable, fast, and should be inexpensive * Lindgren, et. al., "Enhanced NDE Techniques for the C-130 Center Wing", presented at 2002 Aging Aircraft Conference, San Francisco, CA Ref: QNDE 2006: Aging Aircraft NDE: Challenges and Opportunities, Lindgren et.al.
Factors that will Affect NDE of Two-layered Structure A. NDE method: 1.NDE technique 2.Transducer/probe design 3.Contact condition with part (direct, immersion, air-coupled) 4.Scan plan (directions, resolution, orientation) B. Part geometry, material and condition: 1.Layer material, number, and thickness (shims) 2.Outer layer surface condition (paint, very thick coatings, corrosion) 3.Fastener material / type / head condition 4.Hole geometry (oblong, off-angled, surface conditions, scratches, chatter, tool marks) 5.Fastener hole fit (asymmetric fit, irregular contact conditions / loading, sealant) 6.Gaps / sealant between layers (aging) 7.Presence of metal shavings 8.Bushings, residual stress around holes 9.Proximity of adjacent fasteners and edges 10.Presence and condition of repairs C. Flaw characteristics: 1.Flaw number (number of cracks per fastener site) 2.Flaw type (cracks, EDM notch) 3.Flaw location (layer, location in layer: surface, mid-bore, eye-brow cracks) 4.Flaw orientation (around fastener site, skew angle from normal) 5.Flaw dimensions 6.Material within flaw (none, use of filler material, filled with sealant/paint/fluids) 7.Flaw morphology (regular, irregular) 8.Flaw conditions at faces (contact conditions, residual stress) Ref: QNDE 2006: Aging Aircraft NDE: Challenges and Opportunities, Lindgren et.al.
Modeling of Plate Waves Explored effect of interface condition on energy transfer from one layer to another Modeling done with FEM package to solve 2D explicit finite element model for elastodynamic problem (FEAP) Integrated with MATLAB (FEAPMEX) to implement: –Sophisticated transducer models –Arbitrary interface conditions between plates in contact –Facilitated parametric studies Absorbing boundary conditions implemented to reduce the size of the meshed domain and minimize solution time A parametric model constructed to simulate either adhesive or sealant layers between plates –Finite material layer using finite elements, or –Stiffness interface modeled using longitudinal (K L ) and tangential (K T ) springs between adjacent nodes Ref: SPIE 2007: Ultrasonic Guided Waves for Fatigue Crack Detection in Multi-layered Metallic Structures, Lindgren et.al.
5 MHz Results main pulse returns to transducer (very weak case) very stiff fluid very weak Interface Conditions: Observations: Very complex signals for stiff and fluid interfaces will be difficult to interpret - need research to improve inspection methods At t = 48.0 s Ref: SPIE 2007: Ultrasonic Guided Waves for Fatigue Crack Detection in Multi-layered Metallic Structures, Lindgren et.al.
Representative Multi-layered Geometric Configurations Most multi-layered structure located at joints of two skins, e.g. lap-splice However, most joints reinforced by third and/or fourth layer This is critical, as it means the layer of interest typically has material on both surfaces Two-layered lap-splice Three-layered butt-splice Two-layered lap-splice with reinforcing stringer Two-layered lap-splice with reinforcing stringer and tear straps Ref: SPIE 2007: Ultrasonic Guided Waves for Fatigue Crack Detection in Multi-layered Metallic Structures, Lindgren et.al.
Experimental Data Acquisition 5 MHz transducer, coupled via 62° wedge Center sheet: 350 x 50 x 3.2 mm, 10 fastener holes 4.75 mm in diameter separated by 25 mm Through-wall EDM notches measuring 2.5 mm at holes 1 and 6 relative to transducer location oriented perpendicular to ultrasonic wave propagation Two outer sheets either both 1.6 mm thick or both 3.2 mm, without and with liquid at interface Location of EDM notches in red Ref: SPIE 2007: Ultrasonic Guided Waves for Fatigue Crack Detection in Multi-layered Metallic Structures, Lindgren et.al.
Reflections from EDM Notch at Fastener Hole 6 Reflection from the EDM notch at the sixth hole clamped by two thin sheets. Reflection from the EDM notch at the sixth hole clamped by two thick sheets Reflection from EDM notch at sixth hole clamped by two thick sheets with mineral oil at both interfaces Reflection from the EDM notch at the sixth hole with free surfaces Reflection from the EDM notch at the sixth hole clamped by two thin sheets with mineral oil at both interfaces Ref: SPIE 2007: Ultrasonic Guided Waves for Fatigue Crack Detection in Multi-layered Metallic Structures, Lindgren et.al.