Geotechnology: Paradigm Shifts in the Information Age J. Carlos - - PowerPoint PPT Presentation

GeoCongress 2006 - Geotechnical Engineering in the IT Age

Geotechnology: Paradigm Shifts in the Information Age

• J. Carlos Santamarina

Georgia Institute of Technology

Information Technology - Synergism:

microelectronics computers data storage and display sensors digital data analysis inverse problem solving numerical methods communications (cell phones - internet)

Interwoven History

Submicron electronic devices More than 30 nano-technology research centers in the US. 2000's Digital memory and storage IBM Deep Blue defeats G. Kasparov (1997) World wide web 1990's Personal computers & CD players, commercial cellular phones Texas Instrument: single-chip digital signal processor Micromachining 1980's Microprocessors: computers = chip Consumer electronics begin transition to digital Computerized tomography 1970's Computers emerge Growth of digital signal processing: FFT algorithm 1960's Sony pocket-size transistor radio Integrated circuits at Texas Instruments Feynman: nano-technology 1950's Digital computer Transistor at Bell Labs Digital signal processing starts 1940's Car radios and portable radios 1930's Consumer electronics (radios, electronic phonographs) 1920's Fredholm: generalized inverse 1910's

Microelectronics – Moore's Law

1.E+03 1.E+05 1.E+07 1.E+09 1970 1980 1990 2000 2010

Year Transistors per Chip

4004 8080 8086 80286 80386 80486 Pentium & 80786 Pentium III Pentium IV doubles: 24 months data from Birnbaum and Akinwande

Storage

0.00001 0.001 0.1 10 1000 100000 1950 1960 1970 1980 1990 2000 2010

Year Kilobytes per dollar

doubles: 14 months

2006: < $10/GB

data from Kurzweil

The brain - Storage

each neuron stores 1 bit brain ~1 TB 10,000 $ each synapses stores 1 bit brain ~100 TB 1 million$ each molecule stores 1 bit brain ~107 TB 100 billion$

2006 Computer Capabilities

Calculations per second

0.000001 0.0001 0.01 1 100 10000 1000000 100000000 1900 1920 1940 1960 1980 2000

Year (Calculations/second) / $1000

doubles: 19 months

2006: 104 MIPS computers Brain: 108 MIPS

data from Kurzweil; Moravec

Communications

0.0001 0.01 1 100 10000 1000000 100000000 1940 1950 1960 1970 1980 1990 2000 2010

Year MBytes per second

0.0000001 0.00001 0.001 0.1 10

MBytes per second per $

doubles: 10 months doubles: 7 months wireless data from Kurzweil

Lenses: Paradigm Shifts

Galileo Telescope geocentric (pre-Copernicus) heliocentric Leeuwenhoek Microscope biotic sterile

Observations

Underlying technology: doubles every 7-to-24 months At present rate: computers ~ brain in 10-to-20 years How is our field changing? What are possible paradigm shifts?

Building Blocks

Sensors Signals Inversion Content Databases

Nano and Micro Technology Sensors - MEMS

Nano-Control

Nano-manipulation (Eigler 1990)

9.6 Å

Montmorillonite (MDL)

C H O H Na C H C O C H O H Na C H C O

Surface control NaPAA

Micro-electrical mechanical systems MEMS

Cantilever displacement sensor

Yaralioglu et al

Micro-electrical mechanical systems MEMS

Micro-mirror array

Bell Labs

Fiber optic based pressure transducer

0-to-70 kPa to 0-to-7 MPa

www.fiso.com

Distributed Optical TDR Sensors

laser Signal Processing

Strain (Dowding) Pore fluid chemical properties Moisture content (Brillouin - Pamukcu) Temperature (Raman - SENSA) 30 km … every 1 m … 1oC resolution

Soil = innate sensing system

100 200 300 400 500 600 700 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Time [microsec] Meausred Sign al 100 200 300 400 500 600 700 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Time [microsec] Meausred Sign al

(N. Skipper – UCL 2002)

Signal Processing Data Fusion

Signals → Information

• 1

1 2 10 20 30 40 50 60 70 80 90 100

Days Water Level [m]

Dennis (7/4) Emily (7/9) Katrina (8/29) Rita (9/23) 7/1/05 9/30/05

Pilots Station, Louisiana – NOAA

Before Katrina

After Katrina

NSF - D. Fratta

NSF - D. Fratta

Massive data → Display → Information

Biloxi D’Iverville

I-110 Bridge

Pile 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Bathymetry: 200 kHz Sub bottom profiling: 20 kHz

NSF - D. Fratta

Data Fusion: Same Mode

Fuse multi-sensor data to gain new information

http://www.pc.rhul.ac.uk/zanker/teach/PS1061/L6/braille.JPG

Data Fusion: Multi Mode

Navigational Homing in

www.moorhen.demon.co.uk sunsite.tus.ac.jp/multimed/pics/animals/bat.jpg

Cementation - Elastic waves

σ’= 70 kPa

100 200 300 400 500 600 700 10 100 1000 10000

Time (min) Vs (m/sec)

sand-cement

Cementation - Electromagnetic waves

bentonite-cement

0.2 0.4 0.6 0.8 1 1.2

k'/k'

• 0.5

1 1.5 0.0001 0.001 0.01 0.1 1 10

k"eff/k"effo time [days]

Observations

Signal processing = information extraction noise control similarities between signals simple algorithms may be sufficient Data fusion = new information from: multiple same-mode sensors multi-modal sensors spatially distributed sensors concurrent or time-shifted data streams

Inversion

Sensing at boundaries … learning about the body

From CAUSE to EFFECT

d z q t

forward

From EFFECT back to CAUSE

?

d z q t

inverse

Tomography

Unknown internal conditions

?

S3 S4 ∗ ∗ S1 ∗ S2 ∗ 1 4 2 3

            ⋅             =            

4 3 2 1 4 , 4 2 , 4 3 , 3 1 , 3 4 , 2 3 , 2 2 , 1 1 , 1 4 3 2 1

V / 1 V / 1 V / 1 V / 1 h h h h h h h h t t t t

invert

R1 R2 R3 R4

Micro Computed Tomography

Alshibli - www.eng.lsu.edu

Inversion: Ubiquitous in Geotechnology

evolution of G during event ground vibration kh(z) along the pile δh(z) along a pile Cv Cs settlement f(t) Vs(z) from SASW VRayleigh(ω) location and timing of leak pollutant c(z,t) Cv k

• edometer u(t)

constitutive model parameters triaxial F-δ Inverted Values Measured Values

Conceive all experiments within inverse problem solving framework

Distributed Content Development

many + internet = collective intelligence

Great Backyard Bird Count

Northern Cardinal

(2/17/06 - 2/20/06) Responses: 31,515

www.birdsource.org/gbbc

Community Internet Intensity Map

Nisqually 2/28/2001

did you feel it? www.usgs.gov

Wiki-Geo-Pedia?

"Thousands of people, all over the world, from all cultures, working together in harmony to freely share clear, factual, unbiased information… [with the] simple and pure desire to make the world a better place." Wikipedia Founder Jimmy Wales

Observations

Distributed sensing Many not necessarily "sophisticated sensors" Specific task / protocol Proper data gathering / transfer Distributed content development Unprecedented opportunities Development of large databases New information… new understanding… new questions…

Databases

From data to knew understanding

To identify the critical parameters

Risk of heart complications (Database: 10,682 patients - 7 hospitals) Q-waves in electrocardiograms low systolic blood pressure abnormal respiratory sound with fine crackles exacerbation of known reduced blood flow to the heart Better practice/diagnosis Lower cost Enhanced understanding Guide to further research

coefficient of uniformity, Cu 1 2 3 4 5 6 10

1.4 1.2 1.0 0.8 0.6

minimum emin maximum emax

0.8 0.6 0.4 0.2 Youd

To identify the nth control variable

To explore causal relations

rotational frustration (e↑) vs. chain collapse (e↓)

?

10 20 30 40 50 0.2 0.4 0.6 0.8 1

Roundness R CS friction angle cv

cv

42 17 R φ = − ⋅

Spatial Systematic Organization

Mendeleev (1860's)

Bronowski

Spatial organization + analyses: GIS

Paris

www.brgm.fr

Paradigm Shifts

The future ain’t what it used to be …

Yogi Berra

"inert soils" → "self-sensing media"

100 1000 10 100 1000

σ’v [kPa] Vs [m/s]

100 1000 10 100 1000

σ’v [kPa] Vs [m/s]

β

        σ + σ α

a y x S

P 2 ' ' = V

β

        σ + σ α

a y x S

P 2 ' ' = V

A1 A2 A8

B1 B2 B6 C1 C2 C8

Vs (m/s)

35 50 65 80 95 110 >125

Pixel Parametric (RLSS) (L-norms)

Fernandez, Lee

"n-simple tests" → "one information-rich test"

p' q p' q p' q p' q

See also A. Rechenmacher: spatial variability

• ne may be sufficient

many Number of tests as much as needed very limited Information per test comprehensive inversion simplest inversion Interpretation many (x,y,z,t) multisensor very few Measurements complex simplest possible Boundaries a few, information-rich tests many simple tests Philosophy

New Paradigm Old Paradigm

"site investigation" → "model confirmation"

Seoul - GIS

Stratigraphy

gis.seoul.go.kr

based on extensive database printed correlations Design parameters model testing/updating, @site simple, @office Interpretation probabilistic; spatial variability none Real time optimization extensive, multisensor minimal Tools GIS-based model of the site limited Starting point test a model "go and see" Philosophy

New Paradigm Old Paradigm

"design+build" → "predesign+build+monitor+adapt"

forward

strut loads stratigraphy (+ deformation history)

inverse

known, adequate safety "probably" over-designed Safety comprehensive inversion just measured data Inferred infor. continuous - extensively used minimal - limited use Interpretation continuous monitoring sporadic measurements During constr. spatially distributed, multi-mode minimal Sensor system

New Paradigm Old Paradigm

Observational Method in the information age

Closing Thoughts

During the last 30 minutes…

You have received 2 phone calls in your cell phone 1 voice mail in your fixed phone 5 e-mails (2 spam) Your students or employees spent 10 min in instant messenger Decision on your BlackBerry is still on hold… "Digital Attention Deficit Disorder": a real concern !

IT revolution: it's here ! Embracing IT affects: teach, learn, research, solve problems Time for best engineering skills and ingenuity to explore new problem solving strategies IMAGINE A RENEWED GEOTECHNOLOGY inexpensive sensors, unlimited data (z,t) readily searchable comprehensive databases powerful user friendly analysis and simulation software …

Thank you

Organizers Comments and suggestions by: A Bayoumi G Cascante GC Cho A Fernandez D Fratta E Kavazanjian HK Kim TH Kwon JY Lee G Narsilio V Rebata-Landa A Schofield HS Shin JR Valdes TS Yun The Goizueta Foundation