Crack Trajectory Prediction in Thin Shells Using FE Analysis 6 th International Conference on Computation of Shell and Spatial Structures Cornell University and NASA Langley Research Center A.D. Spear 1 J.D. Hochhalter 1 A.R. Ingraffea 2 E.H. Glaessgen 3 1 Graduate Research Assistant, Cornell University 2 Principal Investigator, Cornell University 3 Grant Monitor, NASA Langley Research Center
16
Embed
Crack Trajectory Prediction in Thin Shells Using FE Analysis 6 th International Conference on Computation of Shell and Spatial Structures Cornell University.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Crack Trajectory Prediction in Thin Shells Using FE Analysis
6th International Conference on Computation of Shell and Spatial Structures
Cornell University and NASA Langley Research Center
A.D. Spear1
J.D. Hochhalter1
A.R. Ingraffea2
E.H. Glaessgen3
1 Graduate Research Assistant, Cornell University2 Principal Investigator, Cornell University3 Grant Monitor, NASA Langley Research Center
Outline
• Motivation & objectivesi. Point-source damage: HOW TO LAND SAFELY?ii. Fatigue damage: HOW MANY MORE FLIGHTS?
• Relevant past work• Improvements in physics-based modeling
– Incorporating the nano- & micro-scales
• Current technical challenges
2
Point-source damage: HOW TO LAND SAFELY?
www.youtube.com/watch?v=DUstvXSytRc
Airbus A300 damaged by surface-to-air missile
3
1) Develop finite element-based analyses to predict growth of point-source damage within airframe structures under realistic conditions and in real-time
2) Interface real-time damage assessment with control systems to provide a damage-dependent flight envelope to restrict structural loads in the presence of severe damage
4
Point-source damage: Objectives
Response surface
Point-source damage: Technical approach
Airbus A300 damaged by surface-to-air missile
Stiffeners
Skin
IdealizedDamage
Generic aircraft component damaged by surface-to-air missile
5
• Integrate information from on-board sensors to characterize damage
• Develop interface with control system
• Parameterize damage configurations
• Store a response surface of computed allowable load given the damage configuration and query in real-time
Response surface• Recast structural
component as a lower order model (i.e. equivalent plate)
• Get the sensor description of inflicted damage and compute updated allowable load in real-time
T. Krishnamurthy and J.T. Wang, NASA Langley Research Center
Projectile Projectile
Original Load Allowable
Parameterized Damage State
Global Finite Element Model
Local Finite Element Model
Catastrophic Crack Growth?
Extract Local Boundary Conditions
Explicit Crack Growth Simulation
Decrease Load Allowable
Store New Load Allowable in Response Surface
YES
NO
7
Point-source damage: Response surface method
Outline
• Motivation & objectives– Point source damage: HOW TO LAND SAFELY?– Fatigue damage: HOW MANY MORE FLIGHTS?
• Relevant past work• Improvements in physics-based modeling
– Incorporating the nano- & micro-scales
• Current technical challenges
8
Fatigue damage: HOW MANY MORE FLIGHTS?
The plane, a B-737-200, had flown 89,680 flights, an average of 13 per day over its 19 year lifetime. A “high time”
aircraft has flown 60,000 flights.
April 28, 1988. Aloha Airlines Flight 243levels off at 7,000 meters...
25 mm
Small cracks start at each rivet hole…
…then link to form a lead crack
9
ui10
Global-Local Hierarchical Modeling
ui
Internal cabin pressure, P
Initial crack
Fatigue damage: Relevant past
workFRANC3D-ABAQUS interface
for crack growth simulation
Displacement in z-direction
-1.10
-0.16[inch]
z
measured
predicted
What about -slanted crack growth?-influence of fundamental fatigue damage mechanisms?-the inherent stochastic nature?
Maximum tangential stress for crack trajectoryExperimental determination of phenomenological material
constants:- crack tip opening angle, CTOA- critical radius, rc
11
Improvements in physics-based modeling: Modeling crack front with 3D finite elements
Shell-to-solid couple
12
Improvements in physics-based modeling: Modeling crack front with 3D finite elements
SEM’s of 7075-T651 (R. Campman, CMU)
RDRD
Improvements in physics-based modeling: Considering common damage mechanisms
• (a) Incubation – the process that leads to the first appearance of a cracked particle
• (b) Nucleation – the appearance event of a crack in the matrix
• (c) Propagation – the process of crack extension governed by microstructural heterogeneities– Stage I – Slip along a single band– Stage II – Slip along multiple bands, causing crack
propagation subnormal to the global tensile direction
250 m
a
b
c
SEM/OIM courtesy of Northrop Grumman Corporation
Stage I/II illustration from:
C. Laird, 1967.
0Cycle: 1Cycle:
(a)
100Cycle:
(b)
3000Cycle:
(c)
10 m
Loading Direction
7075-T651
13
– FCC polycrystal plasticity for grains & linear elastic, isotropic for particles
– 1 Cycle @ 1% strain in simple tension, along RD-axis
Improvements in physics-based modeling: Incorporating the nano- & micro-scales
14
Improvements in physics-based Modeling: Incorporating the nano- & micro-scales
15
Incubated crack
Grain Boundary
Molecular dynamics simulation
Crack
Technical Challenges
• Incorporating nano- & micro-scale simulation in a computationally feasible manner
• Determination of damage configurations and assessment during flight
• Better physical understanding of the governing mechanisms for crack growth– Why does CTOA appear to work?
• Interpolating between damage states• Development of real-time interface with control