The Open-Source Approach for Computational Modeling and Simulation for Earthquake Engineering: History, Accomplishments, and Future Needs

50 %
50 %
Information about The Open-Source Approach for Computational Modeling and Simulation for...

Published on June 8, 2016

Author: AcademiaDeIngenieriaMx


1. 141 The Open-Source Approach for Computational Modeling and Simulation for Earthquake Engineering: History, Accomplishments, and Future Needs .4 Gregory L. Fenves ¿ Cockrell School of Engineering The University of Texas at Austin 1!Tih, 1 .....- •i E National Autonomous University of Mexico Mexíco City, Feb. 5, 2015 c

2. - i) i] - HP-1 i] - (L r l'LJ 1 Fundamentals Are Timeless 38 C. FLIOaBILIrÍ OF A i-p I'1BBa Considering the slice as the basic elenent of the 1-O camber, the sarne 000rdinate transformation procese discussed aboye can be used te c1cu1tta the flexibjljty of any arbitrary 1-D member. To demonetrate the procedure, the flexjbjljty of a straight, Uaiforni bar will be evaluated. 1. Unifona Bar y - ELAflC PRCPTI.Ç A-nAI P(cg - L -------__ (L OL Kc 4 flne first step ja any flexibility cnolysjs is the seleetion of the forne systen for whjch the fiexlbility is desired. Ifthebarjs te be used ja structure lying and loaded in one plane, the three forces ahovn ja the sketch wouJ.d be saltable. The flexibillty of the bar may now be cbtaned by su=ing up the internet york ja all Slices of length dx caused by unit values of these external jorcas, thus F = . 1 (bi)Tf1b1 JL(bs)Tfsbsdx The alice flexibiljty f 5 is given in Eq. 13. The jorge tran5formatjoa matrix b 5 represente the fornes ja the arbitrary slice at position "x" causeci by unit values of the externel meniber fornes, thus P 1 0 0 mz O1- p M o o o (15) y MIy L L -JBar y T o o Oj nr alice S5 = b5R (15a) Noy ja Calcndatisg the Internet work of t:e &ljce at position "X" Ray Clough, UC Berkeley, 1960s PLEKURAL / ELt)flt.

3. But Software Evolves Slow1y 1 Category 1 Command Basic k Gcometry larc circic 1 spline lcurv,2, 1, 2,3, 0. 0000 4 04084 0.00000000000 0000e+0 1. 000000000000000e+O 1. 000000000000000e+O 1. 000000000000000e+Ü 1. 000000000000000e+ioü 1.0000000 00000000e+000 6. 66666666()666667c-U(JI 1. 000000000000000e+000 Ii 14 &_I ;. !TT1JLS.

4. 0 Simulation State-of-the-Art 0 "The Good" - Linear structural analysis routine - Good commercial software widely used and ¡ntegrated with BlM - Nonlinear static analysis becoming important - Performance methods becoming more widely used, e.g. ATC 58 9 "The Bad" - Linear analysís with equivalent latera 1 loads is not simulation - Nonlinear static analysis uses very simplified models; it is not simulation - "Performance criteria" not thoroughly investigated, e.g. FEMA 356, nor modeis adequately developed - Long way to go in including uncertainty quantification

5. rs ro and the Ug1y • Many non linear analysis methods based on concepts from 1980sat often software architecture from 1970-1980s • Underinvestment in research in simulation, and what ¡s done is not well organized Poor Iinkages between fundamental experimental studies and modeling; ¡nsufficient validation and verification of models • Simulation modeis, methods, and computational procedures in earthquake engineering have not kept up with rapid advances in computing hardware, software engineering, databases, network communicatíons - Limited interaction with computer science - Inadequate education of students in computing

6. Observations on Historical Situation with Simulation Software • Tight binding of modeis in research and commercial codes is an impediment to new research and implementation of modeis for professional practice. • Embedding of computational procedures in codes makes it difficult to experiment and take advantage of computing technology: - Parallel and distributed computers - Computational grids - Now, cloud computing • "Closed-source" ¡s the norm, whereas other fields have adopted "open-source" software for communities users.

7. Simulation Needs in Earthquake Engineering • Performance-based engineering depends on evaluation of damage and estimate of consequences • Rational, validated modeis of behavior of structural and geotechnical materlais, components and systems are needed for simulating performance • Simulation applications: - Assessment of performance - Design using parameterized modeis, including optimization with performance constraints - Reliability-based desígn - Regional loss estimation and disaster planning • Additional applications include structural and system health monitoring for control and operations

8. Simulation Has Transformed Other Engineering and Science Fie1ds £ • Computational chemisiry, Simulation- Based computational bioiogy Englneering Science • F/1 a te r ¡ a 1 s ci e11ce, p a rt ¡ c u 1 a rl y a t Sinn,lo,io,, nano-scale • Computational fluid dynamics S.dE,g,ong - Aerodynamics - Building interior environment - Virtual wind tunnels • Aircraftdesign • Automotive design • U.S. nuclear weapons stewardship (ASCI, PSAAP)

9. fN Vision for Earthquake Engineering Simulation (2007) NEES Computational modeling and simulation is central to the vision of NEES to transform the development of new earthquake engineering solutions from being primarily based on experiments to a balanced use of simulation and experimentation using computational modeis validated by experimental data. A close integration of modern computational modeis and simulation software with other NEES applications and services will provide the earthquake engineering community, and broad engineering users, new capabilities for developing innovative and cost-effective solutions. NEES George E. Brown. Jr Network for Earthquake Engineerrng Simuiation a-1: (op n riu Te±icic.rv V:ir cf tIe 3.rdfDtec:'r; c'EE". la:. Information Technology within the George E. Brown, Jr. Network for Earthquake Engineering Simulation: A Vision for an Integrated Community . 1 t'••_Ç, . nr; .-..•' - .•j . 4L 1

10. v

11. 09 Software Frarnework • A framework is a set of cooperating software components for buiiding appIications in a specific domain • Aframework dictates the architecture of the appiication - it represents the design decisions common to the appiication domain • A frameworks is based on the assumption that an architecture wili work for most appiications within the domain • Loose-coupiing of components within the framework is essentiai for extensibiiity and re-usability for appiications • Exam pies: Visualization (GLUT), Hadoop, Googie Apps, • A framework ¡s not a "code"

12. o . J1e Open System for Earthquake Engineering Simulation P.tcific 1trtIlthIht Iiigincenng Res..trc h Cc fltcr • OpenSees has been under development by PEER since 1998. NEES supported 2005-2014. PEER since then. • Windows application downloaded over 10,000 times a year. • Parallel Applications utilize over 1,000,000 CPU hours on NSF XSEDE compute resources yearly. • Open-source and royalty free license for non-commercial use and and interna 1 commercial use. • License must be obtained for software developers including OpenSees code in their applications. o Written in C++, C and Fortran (C++ being the main language) ~ ,j 'er) ca, Pr ±,trtJon HOME OpenSees 2.45 Released OPENSEESWIXI HSAGE BoRD 'ersion 2.4.5 of tieOe,Sees binarv ¡s now avatiable for Jownioad. Fere is the chanqe loo OpenSees Days Iy June 19-1 211, 215 )OWN LOAD An Opensees Days workshop will be heid at the Ce -npus of the Ulihersity of Sa eno in Fiscioro, to SOLRCE CODE corsttu:e a meeting point 'or researzhers and practitioners Qn topics elev a nt tD rnodding, aralvsis and clesign in the ficicis of structural and earthquake engineeriig. Deadi nes to relnemDer: BUG REPORT • 31 Deceínter 2014 - Aostract submissicn • 1 March 2015 . Papar submision • 15 Ma-ch 2015 Early bird egis:ration For rrore inforna:icxi, please coosult the arochure. Search Survey — OpenSees rvIissing Pieces Ir cuSIornzethe tO 5_C We are conducting a sJrvey to tdenty prabems ans shortcominc associated with OpenSees. 'lease help by lIIinç c*Jt the sicoy. OpenSces Days 2011 NEES ard PEER hs:ed this years :wo dab evelt 00 Sep:ernber 2526 at the Richmoid fleld Stat on, liC Berkeley. The prsetatiDns are now avai ¿tble onhine and can te 'ound .her3 'I .' ' pP .enSces in the CIouds! - ÚnpnSp'ç iç nrw nvaibh4p for ti-;P ro nl no 'j F"cht h rhrnich ftp OnenSeesLab tnP. rhoçp or vrij

13. United States 264,504 .. a China 138,279 r%) = Irar 80,406 11 Italy 61,838 S. 1'1 Canada 52,019 •3J%) India 39,757 34%) • Japan 31,569 21%) United Kinpdom 30,081 6%) íE Greece 28,644 i%) ;. South Korea 23,567 3%) 0 Turkey 17,957 Germany 12,957 2%) 0 Taiwan 12,380 6%) 14, a New Zealard 10,691 L. Cnie 10,507 Í%) Portuga 10,297 3%) Australa 8,865 .3%) El Hong Kong 7,948 1%) (not set) 7,939 1%) .. France 7,098 2%) El Mexico 6,272 vv) = Spain 6,216 Pageviews 1,514,946 4f .4 264,504 Visitors 94,120 Visits 266,239 ¿ río OpenSees Woridwilde Usage (2014)

14. 1

15. Ir OpenSees Approach to Simulation c Basic approach: - Modular software design for implementing and integrating modeling, numerical methods, and IT for scalable, robust simulation - Focus on capabilities needed for performance-based engineering - Programmab!e interfaces 11 Most users: a "code" for nonlinear analysis. Fully scriptab!e. • Generaily: a software framework for developing simulation applications.

16. Structural Modeling and Simulation Non-ductile RC frames and calíbration of buliding code provisions C. Haselton, G. Delerlein, Stanford - Joints with both bond-slip springs and shear springs fl.. o 1Column base ond-sIip springs Corotational geometric transformations

17. - 11 JU 111 UC San Diego, 2008

18. wi lil Examples of OpenSees Applications • Parametric studies to examine relationship between intensity and damage for PBEE and design procedures • Computational reliability for PBEE • Soil-structure-foundation interaction • Spatial distribution of damage • Simulation of bridge performance

19. Conceptual Approach for Simulation Algorit 1 Solvers nework, ParaIIe sualization, compu com putation nent, system

20. 1 Simulation Software Architecture Traditional Code Fra mework of Components User Interface Input Language Base Code Compute Technology Applícation Program Interface (API)

21. Software Depends on Expressiveness of Language and Power of Processor(s) rJJ2 - Brov,,ser operEdit5tnng Disp Browser operMessa9ecstEdItStrr: Browser openOnCIassWithEditString 8rowse operysternCatEdit5trinq sendets em 'o - send - - - Tool-Inspertor CbangedMessaqeset mesoaqe Iist defneMesqe.00tityu Tocs-Exorer Fakeçlasspool defineMessageFromrs Toals-Debugger I1ierorchyBroeser codo pene inspectinstances To -Fkliçt LinkedMessageSet metadaes inspectSubjnstances (. TooJs-Fe Contente Brc Meseageset copying message1istMenu;sh1t Tods-Changes PackageBrowser drag and drop removeMeesage Syetem-Compiler case removeMeesegeFromB (: eedei'e inplernentois versons ehentance hierarchy instvars daes vare (: IJOptbiFeittoroVie W?j rid again (g) set search etrino (h) W onVie, do again (j) lt fwet previousView 1 eVi,,-VÍew nos, undo(z) bt$eiit '- se opt °0PY (c) L,, CV,ew windoos: (O pate()eight), aV,ew borderWkjt paste,. bottoçn: 1 offset-O, doit(d) ______________________ print it (p) firet .-true. inspect it (i) pres'iouçS'iew ti4 accept (e) -- more,..

22. What is Object-Oriented SM411I4LK-80 Programming? • Object-oriented programs are composed of objects that bind data and operators on data • Objects are operated upon by sending messages to it. The public ¡nterface defines the operations on an object • Object's internal state is encapsulated ¡n the object. The implementation is private • Classes define the software behavior of objects. • Classes and their objects are designed to represent key abstractions • A programmer should be able to use a class through the interface independently of the implementation

23. Object-Oriented Finite Element Frametwork r14'Iltvcl I' nil 4 E 1v In4IIt P rogralnllki ng: Fra IIk4'<)rk-i for i IIi Iv-i-. A l!.!pri 1 IIH4 a 11(1 Pa ial tul ( )IJhJ)lI 1 III LZ 1•. e. 1 . '1 I.; I. 1 Uhi% C'Ill'-L I11L1)lIU. III. !III 1 1TI$, (u11I'I F?I1_.1,!I.1'I.1.-i.I l'I Domm 1 Males LodCas .W_CoaamT SP_Consti.irnt 1 Ld ' Trus CorntumEimut J ' ¡li_u. lii, u_ii 1•'l1 1 1- "1 od1L1 EktevLiLLd eaniPLod • • i _ —E —L - iur lI IlI_1IIjI;,JIl' i 1 1 i II:, 1 'i . 1IJ1•.I 1 .4MC#S?J ........... - ls_-»..._............. - U..-. t 4 E.. 4..ii.. ... Li. - £ £tdO*iS4E 1997

24. . Structural Modeis as Aggregation Pattern 1I 1u E r

25. Analysis Class for Simulation Analysi for per domair

26. 1 Pattern Example of Analysis Class

27. El ernent u p GeometricTran Basic System p = af q 715Linear LinearPD Corotational q2,t 1 '0 u=au Beam-Column Modeis 1

28. "11i q=j aTscix q2,t 71,V1 Displacement e=av Beam-Column Modeis II v1 q Basic System e lJsS ection Force s=bq y =fbbTedx L e = ase s =fA' Material o = G(E ) No assumptions are made on section or material behavior; each level in the hierarchy can be defined independently of other leveis

29. Form Follows Mechanics acrdA LVf z]

30. Types of Behavior Ip AHA 120 (O :2 100 Cu ci 80 u- -a 60(T5 —J 40 mun 0 Fiexure Orily Vu = 125 kips = 70 kips V1 =4Bkips 1 2 3 4 5 6 Lateral Dispacernent (iri) [n(er - 1asticity • Ductile and brittle modes represented • Soution method converges rapidly even with strong softening 160 140

31. 26 26 24 26 26 18 16 14 12 10 O 4 2 24F1 24F2 24F3 24F4 24F5 2426 3403 3404 3405 3608 Expk 2601 2602 2603 2604 2605 2606 3603 3604 3605 3608 Errp1 .. Machine&oker startActor(chax *thePro Chanrtel &heChneI mt comp Demend) OpenSees Parallel Processing - NP3 NP4 ~ NP5 ~ NP7 T • • Ácb.QmeiL :1 • ÁdiIe:l 1 DomainCompon&t Matrix • mP Saek€tI a a [ IO3 •00l - _____________ 1 Subdomain 1 ______........:4

32. rmo Large-Scale Computing and Visualization /7•__• E3 , --v ° 30,237 nodes • 1,140/280 linear/nonlinear BC elements * 81 linear shell elements • 23,556 solid brick elements • 1,806 zero-length elements

33. Click en a directory to enter that directory. Click en a file te display ita revision history and te çjet a chance te display diffs between revisions, Current directory: [local] / OpenSees / SRC Ry Acie API Last loa entry l3SflU • r 0000k — 0000000 Weloomo! TOis pago conteino (orne 00001,1 :nt,:rnaOoi br IGL] b,e,o 001,10. ono 10000000 olLed 11 tOe codo deveboornent of Openoees ObOWPLQOQ Documerttation Broese the Source Code Refute vou begin and lOO cOno no ,et stuck 000000 3Ia 100 0:1 GEn 00€ o-:o-ti- o- - ,0000 '1 urnentatOn FO, fleO uSor: 00 1010,0 0[000 01000 001JG e codo o- /,-' 005ees, hav 3 00k 00 Iho OflL'flc — niceos to gel vourselytmni cro y' ccc lCr wOk tOe 0:01311 000 , 00, :000 orograrnm000 who necd :0 understand toe tone, uork:ngo of toe (lesees 0303 00k 00 tOe Cipos Seeç'fcçancr 5- Downiogd CYS Donboad tf,e source 113 000 licOse Ocio; 3:0 , 0 deslaprnor,t Oil IEh;0 :° oo— NEESI ce: Bu,lds Contribute 10,0 of tOe buid wcçtructcon, fo T. Contoibute codo icbrn,t -i al' finid out bnw 0€ ooecp,le ibiS 0000gev 00 luIsito; bh bonO en ceo' pIptfo,rn. Reos ,r,000uQutt II tOe codo chpnçlei are neelcing en e nno pietfenn ,e epproced tOco libo - cb;00 1 Fi'e Parent Directory Ü AtJ/ íDont hidej converenceTesti/ D coordTransformation/ O damaae/ O deta base,! docZo ci __ OQLDQb O handIr O machine/ O materiai: O modelbui[der.í O nDarray/ O otimization/ O poickacie/ O recorder/ . OpenSees as Open -Source Software Open Source ls... • Roadmap • Architecture • Program Interfaces (APFs) • Code repository — Checkout/in — Branch/merge — Versioning • Release engineering • Testing • Validation & Verification

34. # set sorne variables set gMotion el centro 2 D Steel Moment Frame set scale 1. set roofWeight [expr 80*120.*72./l000.]; #kips; set floorWeight [expr 95*120 . *71/1000 . 1; set nurnFrameResisting 2.0; #load resisting frames set percentLoadFrame [expr 15 .1120.] set dampRatio 0.03 set model 1 ROOF set mode2 3 set Fy 60. 6 set E 30000. 5 set b 0.03 4 #setuprnylists 3 set floorOffsets {216. 150. 150. 150. 150. 150 set colOffsets {288. 288. 288.}; #inches 2 set colSizes {W30X173 W30X173 W27X11 Load: 95psftypical, roof8Opsf E=2990009 Fy=50.0, b =0.003 3% Rayleigh Damping lst and 3rd Modes (1k) (1íi) W24 x 76 (1;), 6h Siory JW24 x 76 W27 x 94 1W27x94 24 fi - W30_x_99 n. 2nd Story W30 x 99 setcoitxizes 1vvI4ÁI9 VV1'4AJi VV14Á1 ' ' •' - '' '' '_•' set bearnSizes {W30X99 W30X99 W27X94 W2/X94 W24X/b W24X/bÍ; # build and run the model using standard template file source SteelMomentFrame2d_UniformExcitation.tcl 5ft IIr(_ - A 1.1 Af1 '. A II AF-1 A 1.1 AJI r fi

  • 35. Concrete Building Study 113 records, 4 intensities 3 hour a record, 1356 hours or 565 days. Ran on 452 processors Qn XSEDE in Iess than 5 hours. - - - 01 , set pid [getPIDJ set np [getNP] set count O; source parameters.tcl source ReadSMDFieNewFormaticI: foreach GMfIle $ ( foreach Factcrl248 $iFactorl248 if ([expr $count % $np] == $ph) { set inFile $G.Md...nS.GMf'ie.AT2 set outFile $$. .M.flç.g3; ReadSMDFileNewFormat Sin File $outFile dt nDts; wipe sou rce GravityAn alysisS cripttc loacCons: .::me 0.0; wipeAnalys is source EQRecorder,tcl source EQAnaIyssScript.tcl if($oko{ us "Process $GMfiIe x $Factorl 248 FINSHED OK modelTime [getTirnejj' } else Duts Process $GMfile x $Factorl248 FINSHED FAIL modeTíme [getTirne] desredTime S.a.xA..aJysi } incr count 1 1

    36. zone f, ': i /,,ne 4 PGA4OOp1 - /onc 2 -044 7-7 1 r -... jprf Prof Xin-Zheng Lu Tsinghua University ¡ u How Does OpenSees Compare With Commercial Software? ! !. 1:12 :lo - PGA 40Og - so 60 40 Shanghai Tower u H632m, 124 stories u 53,006 nodes u 88,089 elements u 48,774 fiber beam elements u 39,315 multi-layer shell elements u Memory used: 8.9GB Similar Resul t -(Results PGA 4O0g1 100 so 0.2 025 03 0.35 Ii with commercial applications the same íff model and analysis are the same)

    37. Lessons and Observations • Many see the benefits of exchanging research and ideas through software • Success depends on ability of developers to understand abstractions in the software design • "Not invented here" is sometimes an issue • Computing education and experience of civil engineers makes the learning curve look steep • Many users just want the code and are not interested in open-source • Documentation is never good enough • Long-term support of an organization is necessary • Innovation is possible, but it takes long-term commitment

    38. . Using the Internet for Simulation (2001) MOdl Bud LMateriais lements Other o Solution Procedures N.J ca Solvers (1) > Compute Technology Internet API's Data bases

    39. Schematic of Simulation in the Future (2002) aci . modeis org modelBuiiders cern E o o -H o o buildingcode. org oomputatiori. com usgs . gov

    40. o __ o Cloud Computing Servers Virtua' Desktop Software Platform e. a.... . Router IL í o o o

    41. líl 101 Cloud as Deployed Services Sofare as a Service (1) 4) Platform as a Service o 4) (1) Infrastructure as a Service D o o > 1 ! 11 U1 I II 1 1 Onf,pp 'oc jj Ir 00, 1 //t?9 a. V o o 4) > a. o o >' c E E o o Deployment modeis

    42. Key Question for the Future • How will nonlinear simulation modeis based on fundamentais be developed for use in performance-based design? • How will validation, verification, and uncertainty quantification (VVUQ) be incorporated in earthquake engineering simulation? • How will the earthquake engineering industry use transformational cloud-based services?

    43. t c u-. - 1 - Mi j ;l'.1• 1. -_ a..- r - : IS .. 1 - 1 - : 1

  • #load presentations

    Add a comment

    Related presentations

    Related pages

    The Open-Source Approach for Computational Modeling and ...

    The Open-Source Approach for Computational Modeling and Simulation for Earthquake Engineering: History, Accomplishments, and Future Needs
    Read more

    “The Open for Earthquake Engineering: History ...

    ponencia “The Open-Source Approach for Computational Modeling and Simulation for Earthquake Engineering: History, Accomplishments, and Future Needs”.
    Read more

    Wood, Fenves Inducted Into Mexico's Academy of Engineering ...

    ... on “The open-source approach for computational modeling and simulation for earthquake engineering: history, accomplishments and future needs. ...
    Read more

    Visualization Techniques for Computational Mechanics

    computational modeling and simulation of ... we provide an open source visualization ... works in tandem with the OpenSees earthquake engineering ...
    Read more

    Jinchi Lu - Publications - ResearchGate

    ... Open System for Earthquake Engineering Simulation, ... simulations. The open-source computational ... modeling approach for earthquake ...
    Read more

    Natural Hazards Engineering Research Infrastructure (2015 ...

    ... Jr. Network for Earthquake Engineering Simulation ... C. Natural Hazards Engineering Research ... Computational Modeling, Simulation, ...
    Read more


    “The Open-Source Approach for Computational Modeling and Simulation for Earthquake Engineering: History, Accomplishments, and Future Needs”.
    Read more

    Thrust Area IV – Simulation & Information Technologies

    Thrust Area IV – Simulation & Information ... Open System for Earthquake Engineering Simulation, ... by loading history. Year 10 modeling research ...
    Read more

    Earthquake engineering - Wikipedia, the free encyclopedia

    ... Facilities for Earthquake Engineering Simulation ... computational and hybrid simulations that may ... buildings in future earthquakes. ...
    Read more

    Miembros por profesión: Ingeniería Civil | Academia de ...

    Usted está aquí. Inicio » Miembros de la Academia » Miembros. Miembros de la academia
    Read more