Structural Health Monitoring Analysis for the Orbiter Wing ... › archive › nasa › casi.ntrs.nasa... · Structural Health Monitoring Analysis for the Orbiter Wing Leading Edge

Structural Health Monitoring Analysis

for the Orbiter Wing Leading Edge

Joint Army, Navy, NASA, Air Force (JANNAF) ConferenceSpace Vehicle & Propulsion Systems Wireless Sensors Workshop

Orlando, FLDecember 6-7, 2010

Keng C. Yap, Ph. D.Project/Technical Lead, Loads & DynamicsBoeing Defense, Space & Security (BDS)

[email protected]

https://ntrs.nasa.gov/search.jsp?R=20100042148 2020-07-05T16:46:52+00:00Z

Yap - 1Copyright © 2010 Boeing. All rights reserved.

Introduction

• Wing Leading Edge Impact Detection System (WLE IDS)

– Columbia re-entry breakup (STS-107) was caused by External

Tank (ET) foam release and subsequent impact on the WLE

– Structural health monitoring (SHM) system was developed

under Return-to-Flight (RTF) to monitor WLE debris threat

– System development led by NASA-JSC, supported by LaRC

& ARC, Invocon, USA, Boeing, LM, ESCG

– Goal is to detect foam/ice & micrometeoroid/orbital debris

(MM/OD) impacts, and help make critical mission decisions

• Impact Analysis Process

– Starts with searching for potential impacts in summary data

– G-time history data are then downloaded for detail analysis

– Impact criteria were established based on extensive impact testing conducted after the accident

– Seek for typical shock response with localized high-frequency transient and damped oscillation

– Primary impact criteria were extended to improve MM/OD monitoring

• Orbiter funded Boeing to explore new impact criteria (damping, multi-sensor, and nonlinear characteristics)

• The development had greatly enhanced the ability to discern MM/OD impacts from false positives

– Analysis capability was extended to provide severity assessment

• Helped establish reporting threshold and determine the level of concern

• Supported by elaborate Orbiter Vehicle testing and NASTRAN modeling efforts


Risk Management via SHM

• Risk Management

– Possible to prevent or reduce the occurrence of structural fault or hazard event

– It may not be feasible or cost-effective to completely prevent fault or eliminate hazard

– SHM can reduce the catastrophic failure risk after a fault condition or hazard event has occurred

• Risk Mitigation

– Goal is to mitigate risk between the time of detection and the time of potential catastrophe

– “. . . the reason for time is so everything doesn’t happen at once”

• Cost-benefit Study

– How much can you benefit from SHM? (trade study, design requirements, system goals)

– How much useable lead-time will you get? (application specific, instrumentation, analysis capability)

– What can you do within this limited amount of time? (repair options, operation changes)

Risk

Management

Anterior

Posterior

e.g., load, thermal, radiation, chemical

Fault Prevention

Hazard Prevention

Condition Monitoring

Hazard Monitoring

e.g., vibration/modal inspection, NDEs

e.g., change of operating environment, shielding

e.g., design change, structural reinforcement

SHM

load


0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45-6

-4

-2

0

2

4

6

Time (s)

Ac

ce

lera

tio

n (

g)

Hardware Overview


Sensors

Data

Acquisition

Units

Wing

Relay

Units

Cabin

Relay

Units

Laptop

Receiver

Unit

Laptop

Crew CabinWing Cavity/GloveWing Spar

RF RF

Instrumentation

Sensors • 132 accelerometers (66/wing) • 44 temperature sensors (22/wing) • Behind spar (RCC #1-19 & chine)

Wing Glove Sensor Unit Farm • 8 sensor units • 1 wing relay unit (WRU-A)

Wing Cavity Sensor Unit Farm • 14 sensor units • 1 wing relay unit (WRU-B)

Crew Cabin Equipment • Cabin relay unit (CRU) • Laptop receiver unit (LRU) • WLES laptop


22

CChhiinnee

88 77 1100 99 1111

11

22

11 22

22

11

22

22

11 11

11

22

22

22 11 44 33 66 55

22 11 11

11

22

22

1122 13 1144 1155 1166 1177 1188 1199 2200 2211

11

2222

Ascent Summary

Download Order

Chan #1 Chan #2 Chan #3

Chine

Sensor

Unit

Wing Glove Units (Fwd) Wing Cavity Units (Aft)

• Accelerometer Locations

– 3 channels per data acquisition (sensor) unit, typically distributed 2 panels apart

– Sensor units are mounted at two separate “farm” areas (wing glove and cavity)

• Ascent Summary Download Priority

– 3 groups of data are downloaded according to a prioritized order

– Download priority is based on the criticality of re-entry aeroheating of the panels monitored

Sensor Configuration


Debris Hazard Monitoring

• Ascent Monitoring

– Debris (foam & ice)

• Foam insulates ET, protects it from ascent

aeroheating, and reduces ice formation

• Study conducted after STS-107 prompted

bipod redesign and NDE closeout

• Foam shedding from multiple locations

reduced but continued to occur

– Ascent Operation

• WLEIDS continued to operate with 10-min

data take through ascent flight monitoring

• Main challenge is to determine when and

where an impact occurred and its severity

• On-orbit Monitoring

– Micrometeoroid & Orbital Debris (MM/OD)

• Micrometeoroids are interplanetary particles

broken off from larger debris

• Man-made orbital debris (e.g., fragments

from satellites/rockets) also pose serious risk

• Small MM/OD damage craters are commonly

found (e.g., RCCs, thermal tiles, radiator)

– On-orbit Operation

• After ascent analysis is completed, sensor

units cycle through idle and trigger modes

intermittently for the remaining flight

• Main challenge is impact discernment

Various Foam Types Ballistic Impact Test Micrometeoroid Crater Damage


ET Sep

SRB Sep

ME On

SRB Ign

Time (sec)

Port RCC

Interface

Max Q

G

rms

• Summary Data

– Data summarized to optimize storage & downlink time

– 312.5 Hz high-pass periodic Grms summary

– 10-min 20 kHz data down to 1200 points

– ME & SRB ignition are most pronounced

– SRB and ET separations are distinctly seen

– Chine shows higher response sensitivity

– Higher noise at certain panel interfaces(foil-wrapped spar insulation batting)

• Summary Analysis

– Screened data for potential impacts

– Process can be slow & labor intensive

– Auto-detection

• Tried using data mining methods

• Adopted expert systems† approach

• Incorporated test, simulation, and flight experiences

• Resulted in significant savings in time and resource

• Safeguards against possible visual prevalence

† An artificial intelligence (AI) approach that captures the expert's knowledge base via representation formalism,

so that the engineered system can serve as an aid to human in the same problem solving setting as the expert

Ascent Response Summary


G Time History Grms Time History

Power Spectral Density G — Noise G o Peak G █ Noise Band G — Noise G o Peak G █ Noise Band

PSDs — Noise PSD — Peak PSD █ Noise Band

Spectrogram

MET (s) Frequency (Hz)

P

SD

(d

B)

Frequency (Hz)

P

SD

(d

B)

G-time History

G Time History Grms Time History

Power Spectral Density G — Noise G o Peak G █ Noise Band G — Noise G o Peak G █ Noise Band

PSDs — Noise PSD — Peak PSD █ Noise Band

Spectrogram

MET (s) MET (s)

G

rms

G

Grms-time History

Power Spectral Density Spectrogram

Response Signal


Click on Image to Play

20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 0

5

10

15

20

25

30

35

40

Mission Elapsed Time, MET (sec)

Nu

mb

er

of

Su

sp

ec

ted

Im

pa

cts

, N

>20 Grms

10-20 Grms

3-10 Grms

1-3 Grms

0-1 Grms

0

0.5

1

1.5

2

2.5

He

ati

ng

Ra

te (

BT

U/f

t2-s

ec

)

Simulated Aerodynamic Heating – Reference Body Point 7000 (1 ft sphere @ stagnation) – Performance Enhancement design trajectory

STS-107

≈45 Grms ST-107

N

um

ber

of

Su

sp

ecte

d Im

pacts

• Data Trend

– System detects as many as 100 indications (low energy, non-damaging, small “popcorn” foam)

– Distribution shows high correlation with ET aero-heating (second hump is less pronounced)

– ET aero-heating causes internal pressure build-up and burst of small pores in foam insulation

• Significance

– Provided the first strong evidence of the system registering real impacts

– Helped establish confidence in the system’s sensitivity to detect more severe foam impacts

– The discovery confirmed the well-known ascent flight phenomenon of popcorn foam release

Distribution of Flight Indications

Nu

mb

er

of

Imp

act

Ind

icati

on

s


• Ascent PRA

– Analysis Goals• Discern impacts from aero-acoustic loads

• Address situational risk due to an indication

• Determine the level of concern

– Analysis Process

• Characterize impact indications by time,

location, and severity

• Use PRA to determine severity and produces “decisionable” information

• Account for varying response sensitivity

across the wing, and uncertainty

(location, angle, velocity, debris type)

• Elaborate effort involving vehicle thumper

testing, model simulation, and risk analysis

• On-orbit PRA

– Analysis Goals• Discern impacts from spurious triggers

• Address situational risk due to an indication

• Determine the level of concern

– Analysis Process

• Estimate impact and damage probability

• Relate flight response to damage from test

• Scale flight response to account for higher

test article response sensitivity

• Model the statistics of these scaling factors

Gmax from Flight Indication

Test Article Scaling

Measurement Scaling

Damage-Gmax Correlation

Panel-to-Panel Scaling

Sensor Configuration Scaling

Damage

Probability

Maximum

Grms

Maximum

Stress

Risk Analysis

Model Runs

Damage Threshold

Model Runs

PRA

Probabilistic Risk Analysis (PRA)


Structure

Test

Test Article Operating Structure

Ballistic Impact Test• Correlate test article model at high loads

• Realistic test conditions

• Full-scale test article used

• No damage risk to operating structure

• Correlate hi-fi model at high loads

• Most realistic test conditions

• Potential damage to the structure

• Prohibitive cost & damage risk

Thumper Test• Correlate test article model at low loads

• Thumper simulated impact conditions

• Full-scale test article used

• No damage risk to operating structure

• Correlate hi-fi model at low loads

• Thumper simulated impact conditions

• Minimal risk of damaging the structure

• Manageable cost & damage risk

Model Correlation


Test Article ThumpingBallistic Impact Test

Thumper Test on OV-105 at KSC


Impact Tests

Hypervelocity Impact Test


Click Image

to Play

WLE IDS Vehicle Hi-fi

Port Wing Model w/

Spar, Fittings, RCC #4-18

Click Image

to Play

Enlarged View of

WLE IDS Vehicle

RCC/Fitting/Spar Model

Wing Leading Edge Modeling

Foam Impact Ice Impact


5 6 7 8 9 10 11 12 13 14 15 16 170

5

10

15

20

25

11 11 10

3.8

6.45.8

5.2 5.44.4

6.0

8.8 8.49.5

Nonuniform, BX-265 Foam, Apex, P*=1/100

Panel

Grm

s*

5 6 7 8 9 10 11 12 13 14 15 16 170

5

10

15

20

25

8.4 8.2 8.2

2.5

4.7 4.23.6 3.9

3.24.5

6.4 6.16.9


Panel

Grm

s*

5 6 7 8 9 10 11 12 13 14 15 16 170

5

10

15

20

25

6.0 5.7 6.0

1.5

3.1 2.7 2.2 2.5 2.13.0

4.2 4.0 4.5


Panel

Grm

s*

5 6 7 8 9 10 11 12 13 14 15 16 170

5

10

15

20

25

4.6 4.3 4.7

1.02.2 1.9 1.5 1.8 1.5

2.23.0 2.9 3.3


Panel

Grm

s*

Ascent Debris PRA Results

Foam Impact on Apex, P=1/500

█ – Nominal █ – 95th Percentile

Foam Impact on Apex, P=1/1000

Foam Impact on Apex, P=1/200Foam Impact on Apex, P=1/100

Impact Location (Panel) Impact Location (Panel)

Impact Location (Panel) Impact Location (Panel)

Resp

on

se T

hre

sh

old

, G

rms*

Resp

on

se T

hre

sh

old

, G

rms*

Resp

on

se T

hre

sh

old

, G

rms*

Resp

on

se T

hre

sh

old

, G

rms*



0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Flight Gmax

Pro

ba

bilit

y

4-sensor

#1-5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15

#16

#17-22

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Flight Gmax

Pro

ba

bilit

y

1-sensor

#1-5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15

#16

#17-22

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Flight Gmax

Pro

ba

bilit

y

2-sensor

#1-5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15

#16

#17-22

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Flight Gmax

Pro

ba

bilit

y

3-sensor

#1-5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15

#16

#17-22

MM/OD PRA Results

4-sensor 3-sensor

2-sensor 1-sensor

Cri

tical D

am

ag

e P

rob

ab

ilit

y

Cri

tical D

am

ag

e P

rob

ab

ilit

yC

riti

cal D

am

ag

e P

rob

ab

ilit

y

Cri

tical D

am

ag

e P

rob

ab

ilit

y

Flight Response, Gmax Flight Response, Gmax

Flight Response, Gmax Flight Response, Gmax


Summary

Key Success Drivers Significance

Common Analysis Tool Provided a unified analysis software for the mission support team

Auto-detection Saved valuable analysis time & resources while improving the quality of results

Impact Criteria Allowed rigorous quantitative & qualitative evaluation of impact indications

Analysis Procedure Guaranteed consistent results by formalizing the analysis steps

Knowledge Integration Developed strong knowledge base from testing, modeling, & flight experience

Aeroheating Correlation Demonstrated high sensitivity, built confidence in detecting damaging impacts

Reporting Threshold Enhanced operational feasibility & sustainability by setting a minimum threshold

Computational PRA Extended the analysis capability to severity determination

Project Elements Lessons Learned

System DevelopmentSHM helps manage risk of operating structures under a hazardous environment

A deployed system can continue to evolve through on-going operation & analysis

InstrumentationWireless instrumentation provides a practical solution for a retrofit design

Power source affects utilization of wireless transmission and monitoring duration

Testing & ModelingExtensive testing provides valuable data for model development

Test & model development is most meaningful when driven by analysis goals

SHM AnalysisComplete SHM analysis involves identification, localization, & severity assessment

Probabilistic analysis is useful for handling many issues involving uncertainty


Conclusions

• System Role

– Debris Risk Management

• Debris hazard environment experienced by the Orbiter presented a challenging risk management problem

• SHM reconditioned this problem, as hazard monitoring made the pertinent flight risk more manageable

– Mission Highlight

• During STS-132 early inspection, OBSS could not properly position the LDRI due to a snagged cable

• EVA was planned to fix the snag, but RCC could not be cleared for re-entry per flight rule

• WLEIDS analysis helped determine that RCC was unlikely to have sustained unacceptable damage

• Future Development

– Wireless Instrumentation

• Overcame many difficulties associated with incorporating the system into an entrenched structure

• Provided a practical platform for an integrated impact sensing, signal processing, and analysis operation

– Future SHM

• Enhance the safety of human space transportation, exploration, and habitation

• Focus on MM/OD monitoring instead of ascent due to in-line design of future launch vehicles

• Medium size particles large enough to cause damage despite shielding and yet too small to be tracked

• Advanced impact criteria developed for MM/OD monitoring will contribute to a more reliable SHM system

• Build on previous technology concept (instrumentation, interface firmware, impact analysis tools)

• Perform cost-benefit study by assessing risk mitigation options within a certain lead-time

• Pursue severity assessment to help realize the risk buy-down from SHM

• Simultaneously monitor for multiple hazards and conditions to get the most bang for your buck


Acknowledgements

• NASA

– George Studor (Principal Investigator), Eli Rayos (led development of modeling and mission operations),

Chip McCann (led development of impact criteria and operations), Josh Johnson (analysis procedures),

Tammy Gafka (project/engineering lead) (NASA-JSC)

– Eric Madaras, Bill Prosser (supported instrumentation and analysis developments) (NASA-Langley)

– Dave Iverson, Ashok Srivastava (application of data-mining techniques to impact detection) (NASA-Ames)

• Invocon

– Mark Kuhnel, John Sumners, Mike Walcer, Kevin Champaigne

• USA

– Jesse Macias & Brian Walker (Project/Subsystem Engineer)

• Boeing

– Keng Yap (Project/Analysis), Jennifer Hodge (Project/Ops), Max Maynard (Project), Abel Noah, Nestor Dub,

Jennifer Dean, Eugene Wong, Clint Stephenson, Michelle Grotts, Nathan Rosnow, Sujatha Sugavanam (Boeing-Hou)

– Mike Dunham (Subsystem Manager), Bill Walls, Darwin Moon, Mo Kaouk (former) (Boeing-Hou Managers)

– Jerry Bohr (Testing), Ed Tong (Modeling), Robert Stephenson, Dan Takahashi, Chhour Thong (Boeing-HB)

• LM & ESCG

– Ken Schultz, Bill Marak, Todd Isaacson, Ernesto Alvarez, Zeeaa Quadri, John Coates

• Panels & Groups

– Structures, Loads & Dynamics (SLD), Leading Edge Subsystem (LESS)

– Micro-wireless Instrumentation (MWIS), Debris Integration Group (DIG), Safety & Mission Assurance (S&MA)


WLEIDS Publications

• Yap, K. C., Macias, J., Kaouk, M., Gafka, T., Kerr, J.

“Probabilistic Structural Health Monitoring of the Orbiter Wing Leading Edge”

The 19th AIAA/ASME/AHS Adaptive Structures Conference, Denver, CO, 2011

• Bohr, J., “Preparation and Support of a Tap Test on the Leading Edge Surfaces of the Space Shuttle”

SEM Winter Test and Measurement Conference, Mesa, AZ, 2009

• Iverson, D. L., Martin, R., Schwabacher, M., Spirkovska, L., and Taylor, W., Mackey, R. and Castle, J. P.

“General Purpose Data-Driven System Monitoring for Space Operations”

AIAA Infotech@Aerospace Conference, Seattle, WA, 2009

• Iverson, D. L., “Data Mining Applications for Space Mission Operations System Health Monitoring”

AIAA SpaceOps Conference, Heidelberg, Germany, 2008

• Studor, G., “Fly-by-Wireless: A Revolution in Aerospace Architectures for Instrumentation and Control”

NASA/CANEUS Workshop, Grapevine, TX, 2007

• Studor, G., “JSC Micro-Wireless Instrumentation Systems on Space Shuttle and International Space Station”

CANEUS Workshop, Toulouse, France, 2006

• Yap, K. C. and Johnson, J.

“Wing Leading Edge Impact Detection System – Impact Data Analysis Tool”

MATLAB & Simulink Technology Seminar, Houston, TX, 2006

• Madaras, E. I., Winfree, W. P., Prosser, W. H., Wincheski, R. A., and Cramer, K. E.

“Nondestructive Evaluation for the Space Shuttle’s Wing Leading Edge”

The 41st AIAA/ASME/SAE/ASEE Joint Propulsion Conf. & Exhibit, Tucson, AZ, 2005

• Prosser, W. H. et al., “Structural Health Management for Future Aerospace Vehicles”

The 2nd Australasian Workshop on Structural Health Monitoring, Melbourne, Australia, 2004

Documents

Structural Health Monitoring Analysis for the Orbiter Wing ... › archive › nasa › casi.ntrs.nasa... · Structural Health Monitoring Analysis for the Orbiter Wing Leading Edge