Background of Component Based Finite Element Method Frantiiok| Motivation + Summarise current design of structural connections. * Focus to FEM features + Explain importance of Validation & Verification Research model Design mode! List of contents + Connection design ~ Models ~ FE analyse + Validation & verification + Components modelling ~ Slender plates ~ Botts + Connection behaviour = Open sections — Hollow sections + SummaryDescription of behaviour for design by moment-rotation characteristic Connection exposed to bending + Rotational stfiness + Moment resistance il + Rotation capacity nome tin jal sess sort eae \Expetmetal cave Baste ime 23 Mm 1 Roaton, mad Cefomrabon capaci a 4 Deformation/rotation capacity *For safety = Seismic design ~ Plastic global analyses 4 a M # * Ductile components — Plate in bending 00°. # * % ~ Column web in shear + Brittle components ~ Bots, welds Rotation capacity Upper material properties (© Question of the Actual yield strengthJoint design European standards | + EN1993-1-8, Eurocode 3, Design of stee! structures, Part 1-8, Design of joints, CEN, Brussels, 2006. + EN1904-1-1, Eurocode 4, Design of composite steel and ‘concrete buildings, CEN, 2010 —z| Rage Ws Design approaches for structural joints “ee | Models SES SR| + Experimental SSa| + Curve fitting “| + Analytical én - Component Method CM us| + Finite element analysis | = Researen Sara: - Design finite etement analysis Re CBFEM MWe Curve fitting modelSurrogate models (Meta models) + Numerical experiments + Standardised mathematical procedure Hollow section joints + Design by mixture of — Curve fitting ~ Analytical models * Failure modes =z — CHS, RHS asin ~ Open and hollow Piasbe failure ofthe chord faceHollow section joints wZ SZ ee ‘Moue F: Local Duckling of me Mote G: Local bucking of chord walis the choc face CIDECT materials + Background CIDECT materials = — ISO/FDIS 14346 ~ EN 1993-1-8 Chapter 7 = + Uni-planar and multi-plangr joints. = Circular, square or rectangular hollow sections + + Uni-planar joints — Combinations of hollow sections with open sections + Detailed application rules to determine the static resistances of joints in lattice structures: 'SOIFDIS 14046; sate Dasgn Prot fr Welded Hotow Section dots - Recommends, 80, IWXV-428-13, 2012, Geometrical types of basic joints Typnt Yiont © Class 1 and 2 cross sections © Limits in geometryGeometrical types of complex joints DY joint Failure modes — chord, shear Chords offer Brace failure Local bucklingComponent based aproach for holow section design + CIDEC project + 7 failure modes = 7 components + Defidend lever arm + The same equations x enginering frendly aproach *+ ky factors transfered to by, elfective widts Component model Procedure § Decomposition of joint § Component description coix § Joint assembly jy § Classification Representation § Modelling in analyses poe meouene | et Component model Prediction accuracy + Good accuracy + MAN Interaction — Tests in PragueComponent model Prediction accuracy mee| * Good accuracy | + M.N Interaction == | _—_ - Testsin Prague Component model Prediction accuracy + Good accuracy + MAN Interaction ~ Tests in Prague Component Model Application § Design tables =z “=| § Green book § Blue book k = § Computer programs § Simplified hand calculationComponent Model Design tools Component Model Design tools For cases ae worw fine cz Component model Background references + Zostemeljer P: Summary of the research on bolted beam-to- ‘column connections, TU-Dait report 25-6:90-2, 1990 + Zostemaljr P: Summary of the Research on Bolted Beam-to- ‘Column Connections (period 1978 - 1983) Ref. No, 6.85-M, Steven Laboratory, Delt. 1983, + Zoatemalier P: Proposal for Standardisation of Extended End Plate Connection hased on Test results -Test and Analysis, Ref No. 6-83-23, Steven Laboratory, Delft, 1983. + Jaspart J.P, Design of structural joints in building frames, Prog ‘Stuct. Engng Mater, 4, 2002, 18-34. + Wald F, Sokol Z., Steenhous M. and Jaspar, JP. Component Method for Stee! Column Bases, Heron 53, 2008, 3-20. + Da Siva Simoes L, Towards a consistent design approach for steel joints under generalized loading, JCSR, 64, 2008, 059-1075. + Beg D., Zupantié E.,Vayas | On the rotation capacity of moment connections, JCSR, 60, 3-§, 2004, 601-620,Finite Element models of joint + Research oriented + Design oriented 11 Research model Design model Bolt FEM research models + 30 Complex E Material ~ Plasticity criteria ~ Demage model f j Wu,2, Zhang, 8. ant arg, 8, Simulaticn of tonsil bolts faite eamont modeling of semixgid beam-to-column connections, Intanatonal Jounal af Stas! Siuetuves, 2012 1203 339360 Bolts FEM design models © Contact model = Matricks description ” |Razavia H. , Abolmaalia A., Ghassemiehd M., Invisible elastic bolt model concept for finite element analysis of boted connections, JCSR, 63, 2007, 47-657 ‘aFEM modelling of bolts Bolts | > Fan model Bearing FEM modelling * von-Mises yield criterion + Demage models Mote, Bea D.. A complete study of bearng stress in single bolt Contevuuns: FE research models +N. Krishnamurthy (1978) ~ 3 models = 2D mode!Connections FE research models wees | « Bahaari and Sherbourne (1996) comecneesy | — 6 fests, 3D finite element model using ANSYS — — One of “=a first component based finite element models Connections FE research models semen "+ Bursi and Jaspart (1998) comenawe | — T.stub and Extended end-plate moment Sees — LAGAMINE, a finite element research software on| ~ e =e] = ast y | neti cont RRS | - Besiamnementmccceteterssmaresere IS Sovtlooreey tee ee Connections FE research models Bahaari and Sherbourne (1999) — 30 model 4testsmens Connections FE research models + NK. Hassan (2004) | + 20 finite element modeling | | Connections FE research models + Emmett A. Summer (2003) + 12 tests, cycling load + 3D finite element modelling + Models of finger shims Connections FE research models + Maggi, Ribeiro (2004) ~ tests four bolts extended unstiffened end-plate ‘moment connections tests — ANSYS 30 modelsConnections FE research models + Gang Shi, Yougjiu Shi, Yuanging Wang, comenz| M.A.Bradford (2008) 3D finite element modeling 7 anes! « 6 experiments Ke Validation and verification procedures + Well developed in FEM theory + To check the physical accuracy «co=| * To check the proper use a cts | + Teche he asked design tv! al Definitions of Verification & Validation ..» |Validation Siz | comes te meena sckton wih te epimers coenss| Verification mm At | rs companaonctcometatnal tors wth hoy aca anal sas |ormamerea commen | erations sips 2 aot ence tat ruber mes ae ZI | prover iotementes and fut the rumencal sokdion is comect with respect ree = ezDefinitions of Verification & Validation = 5s ISODIS 16730 Fire safety engineering -Assessment, veification and validaton of caleulation methods. Geneva, 2008 4 General Aspects of Modeling Spectacular example of a software bug part oe $5120 millon F-22 Raptor Sommureatons| twas 2 computer glitch inthe millions of lines of code, ‘somebody made an error in a couple lines of the code and goes. everything fey nenartoneron yay ooFamous warning Sinking offshore platform Sieipner A What are the predictive capabi of our computer simulations? Computational Science and Engineering (CS&E) Predictive CapabilitiesHierarchy validation and verification == eee m= = \_ =i Aw : We a System response quantity SRQ + Validation is based on the comparison between computational results and experimental data + An experiment can provide much less information than the calculation | * Selection of the system response quantity (SRQ) is limited by the experiment output Validation domains Application domain defines the intended boundaries for the predictive capability of the computational model Validation domain characterizes the representation capabilities of the experiment.Validation metrics Evaluation of Mechanical Structural Response conmecensenn | * Local quantities wo stat) = Stresses cre | Internal forces swe"! _ Larger uncertainties especially > anata coneominns | + Global quantities saree | — Deflection cenedenamee| — ~ Whole (or a large part) of structure Sra | — — Boundery conditon se Verification - Mesh density study wesean [7 Baca OE hae OP THOT =| SE] «oom seonepece » Fl com, I | oss i an i i Jee | a eee.Verification and Validation of Computer Simulations FE model well replicates the experiment Verification and Validation of Computer Simulations —e Effect of calibration Databases of benchmark problems National Agency for Finite Element Methods. and Standards (NAFEMS) www.nafems.org + ~280 verification benchmarks ABAQUS Benchmarks Manual + 264 ($3-NAFEMS, 15-thermal analysis) Verificaton Manual, Example Problems Manual ANSYS® - around 250 problems + Fire engineering "| + Slandardised ISO/FDIS 16730 Fire safety engineering — Assessment, verification and validation cf calculation methods, Geneva, 2008, + DINEN 1991-1.2Validation of research model of T stub “===! 5 Modem history cme) 1997 =| + Busi. S., Jaspart J.P, Benchmarks for Finite Element = Modeling of Boted Steel Connections, Journal of = Constructional Stee! Research, 43 (1-3), 1997, 17-42 > eee | 1999 -——wacemm |. Virdi KS. et al, Numerical Simulation of Semi Rogid commen Connections by the Finite Element Method. Report of aaa Working Group 6 Numerical, Simulation COST C1 a Brussels Luxembourg, 1999. ¥ Hierarchy of benchmark studies for structural steel joints common| meee] = Uno jae ee | Fentiepiate — Genera loaded end pate + Slender plate in compression Tonguarhench Co) = Pat compression between bots + Hollow section joints. = CHS, RHS members + Column tases ~ Tsu in compression and in tension = Genera loaded base plate ss Verification and Benchmark cases Chapters sews | Description “seu | ¢ Component method 7 “al w,| CBFEM ere + Force-deformation curve {Global behaviour | (Verification of resistance a Global verification Initial etifnees, Resistance, Deformation capacity SLAPS | | Benchmark caseComponent based FEM * Joint analyses by FEM ~ Design material mode! + Component behaviour = Connectors + Botts = In shear + Welds + Anchor bolts soe |_— Slender plates Wie ~ Concrete block Material for FE design model + Bilinear ideal elastic-plastic model 2D elements - shells © Shells for design 8 degree of teedom elements 4 notes (degrades 103)Design model Slender plate in compression + Column web san | + Stiffeners = [IPIPIe = fe RB FE research model on T triangular stiffener + Von-Mises stress distribution for beam-to-column joint =| Intrame ata separate element Research FEM + Shell elements, true-stress true strain material model, mesh sensitivity + Geometrical and material nonlinear model with imperfections (GMNIA) + Imperfections based on 1* buckling mode wz | + Experiments ~ literature, own | + Code RFEM aa AWwe Experimental research + Material tests + Flanges — 3x fee edge, 3x partial stifener, 3x fully stiffening + Variation of — Stiffener thickness t — Haunch geometry h and w — Flange thickness t, and width b, Supports and loading + Hinges and compression + Strain gauges LY 11-6/120 2 XY11-6/120 + Deflectometers + Loading by steps till collapse Stiffener with free edge h= 400mm, w= 200 mm, t= 6mmStiffener with free edge 100 mm, t= 4 mm and t= 6 mm. Stiffener with partial stiffened edge h=w=400 mm, |, &= 6 mm, b= 6O mmStiffener with partial stiffened edge 20mm 2 mm, 400 mm, 4mm, f + Validation on force-deformation curve F-5 for resistance + Vertical deformation at centre + Vertical deformation at max. amplitude Validation of research model Free edgeValidation of research model Partial stiffened edge Validation of FE research model on T triangular stiffener tests + Validation of reserch FEM model thm | for resistance Design model of slender plate in compression «| + Buckling analysisDesign model of slender plate in compression + Buckling analysis Design model of slender plate in compression * Buckling analysis Design model Slender plate in compression To eliminate slender plates in joints Reduced stress method + According to EN 1993-1-5 Annex 8 i “| + Criveal buckling factor - Linear bucking analyses de: =="). Load amplifer- Material nonlinear analyses cath “Ziss| + Platesiondemess 4, = + Plate bucking reduction factor. p PB | + Betton esas PWS ~Design model Slender plate in compression owes = . F=57.6kN F 2=07 mm = 0.25% F = 105,8 kN 2=1,6mm c= 07% a= 4,123 aus = 1,659 om p= 0,603, 4001> 10K Design model Slender plate in compressionDesign model Slender plate in compression uss F=175,6KN > _ z= 64mm ‘2 = 5% Design model of bolted connections + Inshear + Intension wnestaettes| + Nonpreloaded conomainu| = Are preloaded > "Enis | + Preloaded bolts comessenreniow | Controlled slip Design model of bolt in tension © Force-displacement diagram SoutOne 7Bolted connection Deformation stiffness of bolt in tension References from literature © Experimental research List of own experiments 2 Two failure modes > Rupture of thread © Tearing down of nutValidation for rupture of thread Validation for tearing down of nut 8 wwsie| 2 Experiment — ~ Research FEM. ne Determation, mm Re ae Parametric study of the bolt tensile stiffness Srmess « NII ERG PETET ESE i Shank length (mm aVerification of design model of bolt in tension Thread length 17,6mm Thread length 28,2 mm Bolts M24 grade 8.8 e Modelling of T stub behaviour + Research model ~ Validation + Design model ~ Verification Bolts research model + Unequal stresses distribution in the threaded part of the bolt al “T stub behaviour Research FE model ccomiatnan| 7 Mesh sensitivity T stub behaviour Research FE model + Bending of bolt + MIDAS FEM HEB300 HEB400Validation — global deformation 400 comuctontnn| Z te entees| $950 Comet & i] 00 250 wettat| 200 conta Experiment > Sims | 100 —Sois elements bots commer] 50 =| , 00 ~«20~«NSS*«S*« MSC ‘terme fe HEB300 Validation — local deformation aeeee = Experiment, tain gauge tt me) — Research FEM ; Deformation lai} JE328 | rate detormaton i yl ine News . Validation — local deformation = Ssperiment stain gauge 8 — Reseach FEM Deformation [uen/e} Plate deformation in yeld ineExperiment with T stub = Experiment stain gauge 1 = Research FEM Deformation, mm T stub design model VerificationT stub design model Verification commen | mange componermenos coven sore a oe me ia 2 oe 1 oo} ow 1 2m 1 m4 me = om 2 wm 1 2 ms emma ee oom 2 se fr © 3 tw? ow 3 km T stub design model Verification “iaetee|| FE ‘ammee| oat - > component method >a ~~ COFEM i —— ie = rene : = we eS : es T stub design model Verification ml weewen| 3, cea] § as mastinan| coment) ES ata eee * Plate thickness: ee * bot eze ment + Bot mate wowecnee “ * Bolt distence * Taub mekness ser /Connection behaviour bolted connections of open section + Generally ~ Shear = Tension snares — Compression “oe'mn| + Research model Sieses| = Validation + Design model = Verification Verification bolts in tension and in shear comectonseun | Beam splices joint Sgrsmmemer'| End plate connection Teton | Stet $255 sonny] Pate 12mm RAB | swewens NOS i Verification bolts in tension and in shear 10 20 20 a 30 ‘abo ead) Moment - rotational diagram predicted by CBFEM and CMGlobal behaviour bolts in and in shear — Me28 kN 746 mrad 7,5 MNem/rad Moment, kNm Rotation, mrad Elastic stresses Global behaviour bolts in tension and in shear 5 MNm/rad metas ‘Moment, kNm 328 Rotation, mrad rg Plastification round the lower bolts Global behaviour bolts in tension and in shear Rotation, mrad Plastification round the upper boltsGlobal behaviour bolts in tension and in shear M=38kNe Fict,714 mrad Sie22,2 MNm/rad Rotation, mrad Plastification round the flanges Global behaviour bolts in tension and in shear Si=20,3 MNm/rad ‘Moment, kNm Rotation, mrad Plastification in the web of HEA ws Global behaviour bolts in tension and in shear Rotation, mrad Plastification in the web of HEA ieGlobal behaviour bolts in tension and in shear M244 kNm_ 138 mrad 0 MNm/rad Rotation, mrad Plastification in the web of HEA Global behaviour bolts in tension and in shear Meas kNm 383 mead | Sis13,3 MNm/rad Rotation, mrad: Plastification in the web of HEA Global behaviour bolts in tension and in shear Meso kNm Fis4,626 mrad 8 MNm/rad Moment, kNm Rotation, mrad Full plastification in the web of HEA. 19Verification bolts in tension End plate thickness Verification bolts in tension Mot rc Geometrical and material parameter Verification bolts in shear err ae ‘eee | - aisVerification bolts in shear {EM Design shearresitance (i) Verification deformation capacity cae 10% | we eam Home| Scenic ue RARE | mre coumcoamnnnerme 33 04s Prediction of Deformation capacity © Material c For resistence fog 5% © For deformation capacity &25= 15% © Actual yield strength —_— EN 1998-1-8 cl. 6.2 = Overstrenght factor gy = 1,25 all Iymox S48 goofygoecentnnir Generaly loaded connections Experiments — beam splices =a kx 34} Verification to component model Welded portal frame eaves moment connection ‘Beam IPE3S0 to column HES260 connection Verification to component model Welded portal frame eaves moment connectionVerification to component model Welded portal frame eaves moment connection ‘Size of IPE profile ofthe beam Verification to component model Welded portal frame eaves moment connection © Thickness of the column web 190 Verification to component model Welded portal frame eaves moment connection 34Joint between open and hollow section + Verification ~ Welds ~ FEMmeshing + Benchmark case — HEA 240 and hollow section RHS 180x100x10 — Loaded by bending moment and shear force 2 Welds for FEM design model + Model of weld - 2,8 -stifness, resistence, deformation capacity =i - stifess, resistence, deformation capacity -8 neglected Joint between open and hollow section sie mae Elastic stageJoint between open and hollow section Plastification of the hollow section RHS upper flange Joint between open and hollow section Initial plastification in the open section web be Joint between open and hollow section sep5inn F256 mead Initial plastification in the open section flangeJoint between open and hollow section esas mrad Sa Mra Full plastitication through the open section web Joint between open and hollow section Meson ‘Sats wad Initial plastification in the hollow section RHS web Joint between open and hollow section | essanm Sie ihr ‘uninay | Plestification Jas of second fange of open section HEA wsJoint between open and hollow section The open section web Teaches design strain 5% Joint between open and hollow section Vetification of resistence CBFEM to CIDEC mod: c+ 4 4 6 6 Resistance, COFEM,AN 143 Summary + Results shows the good accuracy of CBFEM verified to CM + For higher stiffness / resistance / deformation capacity CBFEM compare to CM Verification by research FEM validated to experiments, 149Summary »Verification deals with mathematics; validation deals with physics” Roache P.. (1998) Verifcaton and validation in computational selence ard engineering, Hermosa Publishers Abuquerque, NM, Summary cousondetn | © CM in tables and tools limits poor design hiesces | — by Incompetent amateurs Yasser | 0 CBFEM allows properly analysed/checked commeiuea| —-~ COMBlex design solutions seman | ~ COmplicatedly loaded joints wees) — By well-trained experts > som us Summary Benchmark cases and correct use of V&V limits the improper use of model comeing) © The high-quality education coed) the background ‘sana ) of design of pretty structural connectionsHierarchy of benchmark studies for structural steel joints seuinten > Welded joints comnanesenp | oheoe fe | = inbending rmeden | — Long ‘rect | _— enble late ‘esnee | + Bolted connections - % ~ stub in tension =a swasticce | — Spies in shear svecs| Seve add ed plate Seweeatt | «Slender plate in compression seieome | Wiangularhaunch wets) Stifeer ofcoiumn web Comatenttwer| _— Flatain compression between bots SSeS | + Hollow section joints ~ CHS, RHSmenders > tamer | ___— gow ae open Sections es + Column bases a — "tun eompreccion andi tension Ws = Gaal aod basa sit Background references Component based FEM Imreacten| 4 Wald, F, Gédrich, L, Sabatka L. Kaboigé, J. Navratil J. conecionsein| Component Based Finite Element Model of Structural et| Connections. in Steel, Space end Composite Structeres. Crees Singapore, 2014, 337-344, ISBN 978-981-09-0077-9, vesime| + Godrich L, Wald F, Sokol Z, Advanced Modelling of End meen] Plate. In Eurosteo! 2014. Brussels, ECCS, 2014, 287.288, sevemen) ISBN 978.92.9147-121-8 Sorverenvis| « Godnich L., Kurejkova M., Wald F, Sokol, Z., The Bolts and. “sss | Compressed Plates Modelling. In Steel, Space and. Composite Structures, Singapore, 2014, 215.224, ISBN 978- ‘omeconrnrion'| 981.09.0077-9 suwwses| « Waid F, Sabatka L., Kaboléé J., Kolaja D., Pospisil M > momen | Structural Analysis and Design of Stee! Connections using 3 ‘Component Based Finite Element Model (CBFEM), Journal (of Civil Engineering and Architecture, 10:2015 URL: www.ocel-drevo.fsv.cvut.cz ee Frantitek Wald Czech Technical University in Prague