An investigation on stability of composite i girder bridge made of steel for bridge high performance structure (SBHS)

VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY NGUYEN XUAN BACH AN INVESTIGATION ON STABILITY OF COMPOSITE I GIRDER BRIDGE MADE OF STEEL FOR BRIDGE HIGH PERFORMANCE STRUCTURE (SBHS) MASTER’S THESIS Hanoi, 2020 VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY NGUYEN XUAN BACH AN INVESTIGATION ON STABILITY OF COMPOSITE I GIRDER BRIDGE MADE OF STEEL FOR BRIDGE HIGH PERFORMANCE STRUCTURE (SBHS) MAJOR: INFRASTRUCTURE ENGINEERING CODE: 8900201.04QTD RESEARCH SUPERVISOR: Dr DANG VIET DUC Ha Noi, 2020 ACKNOWLEDGEMENTS I would like to give sincere thanks to my supervisor, Dr Dang Viet Duc, for his guidance and supports throughout the time I study on this thesis By his advice and inspiration, Dr Duc has given me a passion for structural engineering and the motivation to excel in all my efforts I am honored to have conducted this research under his guidance I would also like to thank Prof Hironori Kato, Prof Nguyen Dinh Duc, Dr Phan Le Binh, Dr Nguyen Tien Dung, and the teachers of Master Infrastructure Engineering (MIE) Program Their instruction throughout my studying career in Viet Nam Japan University is greatly appreciated Also, their recommendation and question is really invaluable for me, which make my knowledge better I would like to thank all staff of Japan University, both past and present, for all their support and help throughout two years, I have studied in Viet Nam Japan University I will never forget my experiences here with them I am indebted to my parents for their endless support Without them, none of this would have been possible They have given me as much as possible They are my inspiration I would especially like to thank my girlfriend, Hoang Ngan, for being my best friend and my unending support I would like to extend my thanks to her family for giving me support and confidence throughout over the time ABSTRACT The steel-concrete composite I-girder is one of the most popular supperstructural types for highway and railway bridges They are design to a trend using larger twin girders and simple crossbeam instead of system more two girders and complex k-frame As a result, the slender ratio system girder becomes larger which makes the lateral buckling can occur in the system Current American bridge design specifications (AASHTO 2012) simplify the flexural design of I-girder sections by treating local and global inelastic stability phenomenon independently According to specifications, if a section is compact and sufficiently braced against lateral instability, the member will achieve or exceed its theoretical plastic moment capacity Treating local and global buckling has been proven by past experience to be suitable when formulating flexural design provisions for lower strength steels However, new researches are proving that these provisions have recently been inappropriate to apply to the design of High Performance Steel Besides, application of bridge high performance steels, SBHS500 girders is expected to be an economical solution for composite girder bridges Steels SBHS500 with yield strengths of 500, have been standardized in 2008 in Japanese Industrial Standards (JIS) They present the advantage of high yield strength, good weldability However, if compared to conventional (normal) steels they show different inelastic behavior, such as almost no yield plateau, smaller ductility, and a greater yield-to-tensile strength ratio Consequently, the main objective of the current study is to investigate the effect of crossbeam spacing on Ultimate Flexural Resistance of composite Igirder systems which are SM490Y and SBHS500 systems Finite element modes of SBHS500 and SM490Y systems, employing nonlinear analysis, are used to determine crossbeam spacing and Ultimate Flexural Resistance, precisely After indicating that limit of crossbeam spacing in conventional design standards are over conservative for composite SBHS500 steel girders TABLE OF CONTENTS CHAPTER INTRODUCTION 1.1 Introduction of composite girder bridge .1 1.2 Issue of design for composite twin bridges 1.2.1 Thicker steel plates and new steel grades 1.2.2 Assumption of lateral-torsional buckling 1.3 SECTION CLASSIFICATION 1.3.1 Behavior of bending beam 1.3.2 Requirement Unbraced Length of compact section .9 1.3.3 Requirement of Wed Plate 11 1.4 Objectives 12 1.5 Overview of Thesis Organization 12 CHAPTER LITERATURE REVIEW 13 CHAPTER METHODOLOGY AND MODELING 16 3.1 Methodology 16 3.1.1 Nonlinear finite element analysis 16 3.1.1 Material modeling 17 3.1.2 Initial Geometrical Imperfection 18 3.2 Finite element model 20 3.2.1 The simplified twin I girder composite structure 20 3.2.2 Steel and concrete material model 23 3.2.3 Meshing of composite girder model 25 i 3.2.4 Initial geometrical imperfection 25 3.2.5 Phases structural analysis 26 CHAPTER RESULT AND DISCUSSION 28 4.1 Effect of crossbeam spacing on the ultimate flexural resistance 28 4.2 Effect of stiffener spacing on the ultimate flexural resistance 30 CHAPTER CONCLUSIONS AND RECOMMENDATIONS 32 5.1 Conclusion 32 5.2 Recommendations for future research 33 REFERENCE 34 APPENDIX 36 Appendix Properties of composite section 36 Appendix Checking dimension of section by provision of AASHTO 2010 41 Appendix Results and bending moment and curvature plots 42 ii LIST OF FIGURES Fig 1.1 The composite twin I-girder bridges in France (Le Viaduc de l'Hyrôme,2011) Fig 1.2 Global buckling mode in I-girder under loading of wet concrete and simplified outstanding plate Fig 1.3 The actual test stress-strain relation of conventional and high strength SBHS steel grades (Dang Viet Duc, 2013) Fig 1.4 Nagata Bridge made of SBHS 500 steel grade with truss system (Source: https://www.pinterest.com/pin/297659856597529870) Fig 1.5 Tokyo Gate made of SBHS 500 steel grade with full-welded truss girder (Source https://photos.com/featured/tokyo-gate-bridge-i-kadek-wismalana) Fig 1.6 Twin I-girder system (Yoseph Yura, 2008) .7 Fig 1.7 Beam Behavior (Yura, Galambos, and Ravindra 1978) Fig 1.8 Classification of beding section referred by AASHTO-2010 Fig 1.9 Basic form of all I-section flexural resistance equations follow unbraced length referred by AASHTO 2010 10 Fig 1.10 Assumption of tress distribution in composite steel section referred by AASHTO 2010 11 Fig 2.1 Finite element model of Twin-I girder bridge 13 Fig 2.2 Steel material comparison for different girder depths 14 Fig 2.3 Stability comparison for different stiffener spacing 14 Fig 3.1 Constitutive Law-True Stress versus True Strain (abaqus-docs.mit.edu) 17 Fig 3.2 Dominant buckling modes of I-girder systems 19 Fig 3.3 Vertical view of the model 20 Fig 3.4 Plan view and condition boundary 20 Fig 3.5 Horizontal view of the model 21 Fig 3.6 Bending moment produced by displacement control method 22 Fig 3.7 Idealized stress-strain relations for steel material model (Dang Viet Duc, 2013) 23 Fig 3.8 Idealized stress-strain relations for concrete material model (Material Laboratory of Ha Noi University of Transport and Communication) 23 Fig 3.9 Cross-section and type of elements 25 iii Fig 3.10 Initial deflection for the beam supported simply as the composite girder models on plan view (a) and 3D model (b) 26 Fig 3.11 Different construction phases 27 Fig 4.1 Curvature relationship between bending moment and rotation of SM 490Y structure with varying crossbeam spacing 28 Fig 4.2 Curvature relationship between bending moment and rotation of SBHS500 structure with varying crossbeam spacing 29 Fig 4.3 Curvature relationship between bending moment and rotation of SM 490Y structure with varying stiffener spacing 30 Fig 4.4 Curvature relationship between bending moment and rotation of SBHS500 structure with varying stiffener spacing 31 iv LIST OF TABLES Table 1.1 Definition and web slenderness limits in AASHTO 2010 12 Table 2.1 Cross section dimension based on same section area 13 Table 2.2 Stress and displacement reponse for different cross-beam spacing 15 Table 2.3 Stress and displacement response for different cross-beam depth 15 Table 3.1 Steel girder systems investigated in the study 21 Table 3.2 Characteristic of steel in the study 24 Table 3.3 Characteristic of concrete in the study 24 Table A.1 Result of plastic moment 39 Table A.3 Result of internal moment of sections 42 v LIST OF ABBREVIATION UFR Ultimate Flexural Resistance Lb Spacing between crossbeams LTB Lateral-torsional buckling GLB Global lateral buckling Mp Moment plastic of section Fy Yield strength of steel vi bending moment in process of construction affects insignificantly on stability of structure when crossbeam spacing is varied from 6.3 to 12.6 m 4.2 Effect of stiffener spacing on the ultimate flexural resistance In current design, most of girders are braced with stiffeners spacing to 1÷2D (D as depth of girder) However, larger distance (2÷4D) is considered in the study to prove current design over conservative According to the line graph in Fig 4.3, the SM 490Y structure still keeps global stability in spacing of 3.15 and 4.2 However, the structure is not stable with the spacing of 5.04m Largest stiffener spacing of SM 490Y structure at 4.2m is less than that in Haiying Ma and Xuefei Shi’s study at 4.375m Fig 4.3 Curvature relationship between bending moment and rotation of SM 490Y structure with varying stiffener spacing The result of investigating SBHS500 structure is presented in Fig 5.3, the largest stiffeners spacing enables to achieve to 5.04m without global buckling The stiffener spacing is two times larger than that of current design and provision of 30 AASHTO 2010 And so, the limit pacing of SBHS500 structure is 30% greater than that of SM 490Y Fig 4.4 Curvature relationship between bending moment and rotation of SBHS500 structure with varying stiffener spacing 31 CHAPTER CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusion In an effort to investigate the effect of crossbeam spacing on ultimate flexural resistance, twin I girder composite bridge is model by FEM with two types of steel: SM 490Y and SBHS500 In addition, the change of load and structure regarding unshored construction method is considered in the study Main conclusions of this research are as follows: The largest crossbeam spacing of SBHS500 twin I girder composite structure enable reach to 10.04m without global buckling The SBHS500 section achieves moment plastic and still has elsto-plastic part in order to distribute rotation Therefore, it is recommended that larger spacing is more effective Applying SBHS500 steel to twin girders can extend significantly the limit spacing (Lp) for reducing the number of crossbeams and stiffeners In comparison with the investigation of SM 490Y structure, the UFR of SBHS500 structure is greater The SBHS500 structure is not only larger spacing (about 20%) but also distribution of rotation after the section achieves plastic moment The elasto-plastic part of SBHS500 structure is higher than that of SM 490Y According to crossbeam spacing investigated, the effect of initial bending moment caused by loads in construction is insignificant Because the structure still keeps stable without any extra supports (un-shored construction method) The FE model study intends the knowledge based on the ultimate flexural behavior of composite I girder With compact sections suffering to two-point loads in which the failure of system is caused by crushing in concrete slap The obtained ultimate flexural resistance indicates that the prediction of crossbeam spacing AASHTO 2010 provisions is restricting the application of SBHS500 Stability of structure investigated by a nonlinear model, the result of stiffener and crossbeam spacing is smaller than linear model 32 5.2 Recommendations for future research Future research is necessary to investigate the ultimate flexural resistance of composite girders with changing dimension of section In addition, effect of crossbeam position on depth direction of girder also needs to be considered With this further research, a more conclusive analysis may be done on the UFR of girders for developing section classification limits Moreover, the FE model should be extended to model the behavior of continuous girders As a result, the behavior of twin composite girders in negative bending should be investigated 33 REFERENCE [1].American Association of State Highway and Transportation Officials (AASHTO) (2010) LRFD bridge design specifications–2012 interim revisions, Washington, D.C [2].American Association of State Highway and Transportation Officials National Steel Bridge Alliance, (2014), Guidelines for Steel Girder Bridge Analysis 2nd Edition, Bud Wright, Washington DC [3] Dang Viet Duc, 2013, Flexural Capacity of Composite Girders: Design Equation Accounting For Bridge High Performance Steel, Department of Civil and Environmental Engineering Graduate School of Science and Engineering Saitama University [4].Ansourian, P (2002) Plastic rotation of composite beams, Journal of Structural Division, ASCE, Vol 108, No 3, pp 643-659 [5].Barth, K.E and White, D.W (2000) Finite element evaluation of momentrotation characteristics in continuous-span steel I Girders, Engineering Structures, Elsevier, Vol 20, No 8, pp 761-778 [6].Basker, K., Shanmugam, N.E., and Thevendran, V (2002): Finite-element analysis of steel-concrete composite plate girder, Journal of Structural Engineering, ASCE, Vol 128, No 9, pp 1158-1168 [7].Bradford, M.A., Uy, B and Pi, Y.L (2001) Behavior of unpropped composite girders curved in plan under construction loading, Engineering Structures, Elsevier, Vol 23, Issue 7, pp 779-789 [8].Vivek Kumar Gupta, (2006), Development of section classification criterion and ultimate flexural equation for composite I-Girder., Department of Civil and Environmental Engineering Graduate School of Science and Engineering Saitama University 34 [9].Theodore V Gakanbos, Andrea E.Surovek, (2008), Structural Stability of Steel: Concepts and Application for Structural Engineers, Published by John Wiley & Sons, Inc., Hoboken, New Jersey Published simultaneously in Canada [10].T.V Galambos, (1963), Inelastic lateral buckling of beams, Proc ASCE, Vol.89 (ST5), Publication No.236, Laboratory Reports by an authorized administrator of Lehigh Preserve [11].Joseph Yura, M.ASCE; Todd Helwig, M.ASCE; Reagan Herman, A.M.ASCE; and Chong Zhou, A.M.ASCE4, (2008), Global Lateral Buckling of IShaped Girder Systems, Journal of Structural Engineering ASCE [12].Haiying Ma and Xuefei Shi, (2016), Parametric Study on Behaviour Twin-I girder Bridge Systems with Cross-beams, he Worl Congress on Structures, Department of Bridge Engineering, Tonji University, Shanghai, Chinao [13].Masahide Takagi, Koji Homma, (2015), Creation of high quality and costeffective bridges using newly developed steels for bridge high performance structures (SBHS), Nippon Steel & Sumitomo Metal Corporation, Tokyo, Japan [14].Michael J Garlich, (2015), Girder Stability in Erection & Demolition, S.E, P.E Collins Engineers [15].Sean Justin Coffelt, (2010), Stability Analysis of Single and Double Steel Girders during Construction, University of Tennessee, Knoxville [16].ABAQUS 2009 Finite element program users’ manual, version 8.0, Houston [17].Yura, J A., and Widianto, (2005), “Lateral buckling and bracing of beams-A re-evaluation after the Marcy bridge collapse.” Proc., Structural Stability Research Council, Montreal, 277–294 35 APPENDIX Appendix Properties of composite section In calculating yield moment, My and plastic moment, Mp for concrete-steel composite girders homogeneous section was conducted in accordance with the AASHTO 2010 specification and study of Dang Viet Duc,2013 In all of the under figures, the positive (+) and negative (-) sign in stress diagrams shows the stress distribution of girders in tension and compression, respectively A.1 Yield moment (Dang Viet Duc,2013) Fig A.1 Composite section elastic stress distribution under positive bending As shown in Chapter III, yield moment My is calculated with assumption of elastic stress distribution in the section once either of extreme steel section firber start yielding In the Fig A.1 The depth of elastic neutral axis from bottom, ypB for composite girders under positive bending (Fig A.1) is calculated as given below y pB  Ac  y  Auf yuf  Aw yw  Alf ylf Ai yi  n  c    Apc  Ac     Auf  Aw  Alf  n  36 Where Ac, Auf, Aw and Alf are the areas and yc, yuf, yw and ylf are the centroidal depth of each component from bottom of the steel girder and Apc is the total area of composite girder Ratio of the moduli of steel to concrete n: n  Es Ec Where Es and Ec are Young’s modulus of steel and concrete, respectively The moment of inertia of composite girder, Ipc is 3 tuf bc tc3 buf tuf bwtw3 blf tlf Ac  tc  b I pc       y pT    Auf ( y pT  tc  )  Aw ( y pB  tlf  w ) 12n 12 12 12 n  2 2 tlf  Auf ( y pB  ) 2 The section modulus, Smin is given by Smin  I pc y p ,max ( y pT , y pB ) Therefore, the moment of resistance of composite girder, My can be determined as following My=fySmin Where fy is the yield stress of steel material A.2 Plastic neutral axis and plastic moment capacity of section The general dimensions and plastic forces at ultimate limit state are represented as shown in Fig A.4.The idealized rectangular stress distribution has been adopted in calculation because it simplifies considerably flexural strength calculations For the ultimate strength design calculations, it is assumed that the concrete slab is not reinforced in the longitudinal direction and there is full interaction between concrete slab and steel girder 37 Fig A.2 Stress block for a composite steel section at the ultimate limit state For plastic neutral axis lies in the web of the steel girder because Pw+Plf>Pc+Puf The plastic forces of different components of composite girders are calculated as follows: The homogeneous steel section a Concrete slap: Pc=0.85fcbctc b Compression flange Puf=fybuftuf c Web Pw=fybwtw Pw=fyDcptw Pw=fy(bw - Dcp) tw d Tension flange Plf=fyblftlf Force equilibrium equation for assumed plastic stress distribution in the homogeneous composite sections Pc+Puf+Pw,com=Pw,ten+Plf The distance from top of the web to plastic neutral axis is given by 38 D cp  Plf  Pc  Puf f y tw  bw Where, Dcp is the depth of the plastic neutral axis from the top of the web The nominal plastic moment capacity of the composite section, Mp is calculated next by taking moment of the compressive and tensile forces about plastic neutral axis Dcp  tuf   bw  Dcp   tlf  t  M p  Pc  c  tuf  Dcp   Puf   Dcp   Pw,com  Pw,ten    Plf   bw  Dcp  2      2  Table A.1 Result of plastic moment No Name Sign Unit Value Depth of girder hd mm 2200 Thickness of concrete slap bc mm 200 Thickness of top flange tuf mm 40 Thickness of bottom flange tlf mm 60 Thickness of web tw mm 300 Depth of total system h mm 2500 Width of top flange buf mm 800 Width of bottom flange blf mm 900 Width of cantilever concrete btm mm 3000 10 Width of internal concrete bpm mm 3600 11 Effective width of girder Beff mm 2613 12 Modulus of concrete Es Mpa 200000 SM490 Section Yield strength fy Mpa 355 Ultimate strength of concrete fc Mpa 35 Compressive force Puf N 1.14E+7 Tensile force Plf N 1.92E+7 39 Force in web Pw N 1.86E+7 Force in concrete slap Pc N 1.83E+7 Depth of plastic neutral axis Dcp mm Yield moment My N.mm 4.35E+10 Plastic moment Mp N.mm 6.17E+10 459.52 SBHS500 Section Yield strength fy Mpa 500 Ultimate strength of concrete fc Mpa 35 Compressive force Puf N 1.60E+7 Tensile force Plf N 2.70E+7 Force in web Pw N 2.63E+7 Force in concrete slap Pc N 1.83E+7 Depth of plastic neutral axis Dcp mm Yield moment My N.mm 6.13E+10 Plastic moment Mp N.mm 7.68E+10 40 459.52 Appendix Checking dimension of section by provision of AASHTO 2010 A.2-1 Provision required for section - The flanges satisfy the following ration: Iyc Iy Iyc/Iy 1706666667 5354401041 0.319 Ok - The section satisfies the web slenderness limit: √ 36.76 89.24 Ok - Flange proportions bf 2.t f b f  hw / 6;  12.0; t f  1.1tw The section satisfies all above provision - Unbraced length required Lb≤Lp Where, Lb and Lp are unbraced length (mm) and limit unbraced length to achieve the ultimate flexural resistance (Mp), respectively Lp  1.0rt E = 5850 (mm), where Rt  f yc 41 b fc  Dctw  12 1   b t  fc fc    201.1 Appendix Results and bending moment and curvature plots The results of numerical analyses in Chapter V are tabulated below The relationship between bending moment and curvature of composite sections for different amount of initial bending moment are shown Table A.3 Result of internal moment of sections Girder ID Grade steel Lst(m) Lcb(m) My Mp Mu A.210.6300 SM 490Y 2.10 6.30 4.35E+10 6.17E+10 6.17E+10 A.210.8400 SM 490Y 2.10 8.40 4.35E+10 6.17E+10 6.17E+10 A.210.1008 SM 490Y 2.10 10.08 4.35E+10 6.17E+10 6.13E+10 A.315.6300 SM 490Y 3.15 6.30 4.35E+10 6.17E+10 6.17E+10 A.420.8400 SM 490Y 4.20 8.40 4.35E+10 6.17E+10 6.17E+10 A.504.1008 SM 490Y 5.04 8.40 4.35E+10 6.17E+10 6.16E+10 B.210.630 SBHS500 2.10 6.30 6.13E+10 7.68E+10 7.68E+10 B.210.840 SBHS500 2.10 8.40 6.13E+10 7.68E+10 7.68E+10 B.210.1008 SBHS500 2.10 10.08 6.13E+10 7.68E+10 7.68E+10 B.210.12600 SBHS500 2.10 12.60 6.13E+10 7.68E+10 7.65E+10 B.315.6300 SBHS500 3.15 6.30 6.13E+10 7.68E+10 7.69E+10 B.420.840 SBHS500 4.20 8.40 6.13E+10 7.68E+10 7.68E+10 B.504.1008 SBHS500 5.04 10.08 6.13E+10 7.68E+10 7.68E+10 B.840.8400 SBHS500 8.40 8.40 6.13E+10 7.68E+10 7.68E+10 * Note: Lst and Lcb is spacing between stiffeners and crossbeams, respectively 42 (a) B.210.6300 (b) B.210.8400 (c) B.210.1008 (d) B.210.12600 (e) B.315.6300 (f) B.420.8400 (g) B.504.1008 (h) B.840.840 Fig A.3 Curvature relationship between bending moment and rotation of SBHS500 structure 43 (a) A.210.630 (b) A.210.840 (c) A.210.6300 (d) A.315.6300 (e) A.420.8400 (f) A.504.1008 Fig A.4 Curvature relationship between bending moment and rotation of SM 490Y structure 44 ...VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY NGUYEN XUAN BACH AN INVESTIGATION ON STABILITY OF COMPOSITE I GIRDER BRIDGE MADE OF STEEL FOR BRIDGE HIGH PERFORMANCE STRUCTURE (SBHS). .. Transportation Officials (AASHTO) (2010) LRFD bridge design specifications–2012 interim revisions, Washington, D.C [2].American Association of State Highway and Transportation Officials National Steel Bridge. .. investigate the ultimate flexural resistance of composite girders with changing dimension of section In addition, effect of crossbeam position on depth direction of girder also needs to be considered