6 views

Uploaded by Viet Duc Dang

international conference paper

- Block Shear Failure Seminar Report
- Strength Consideration in Product Design, Shiva,Deepak
- Ultimate Strength
- Swedish Regulations for Steel BSK99
- Development of Coiled Tubing Stress Analysis[1]
- Fea for Backhoe
- CAR FRONTAL IMPACT
- purlin
- Uncofined Test
- HW1.1
- Yield Curvature for Seismic Design of Circular Reinforced Concrete Columns-Sheikh (2010)
- Tests on Rocks
- MEE06013_Khan_Shahzada_2007[1]
- EnsayosAlSikadur AnchorFix-4
- Is Std References List
- Griffiths Theory . Hw6
- Zhu 2017
- short delay times
- PM-PHI
- MDP 1 Lecture 1 Design Methods Revised

You are on page 1of 11

Journal:

Manuscript ID:

Theme:

Date Submitted by the Author:

Complete List of Authors:

Type of Structure:

Material and Equipment:

Other Aspects:

Seoul-0132-2012.R1

Structures and Materials - extending the limits

28-Feb-2012

Dang, Viet Duc; Saitama University,

Okui, Yoshiaki; Saitama University, Civil and Environmental Engineering

Hagiwara, Koichi; Saitama University, Civil and Environmental Engineering

Nagai, Masatsugu; Nagaoka University of Technology, Civil and

Environmental Engineering

Bridges

Steel

Page 1 of 10

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

Abstract for

18th IABSE Congress 2012, Seoul, Korea

Innovative Infrastructures Toward Human Urbanism

Date: 14/09/2011

Contact Author: Dang Viet, DUC, dangviet.duc@gmail.com

Local Buckling Strength and its Statistical Information of Normal and Bridge

High Performance Steel Plates

Dang Viet Duc, Yoshiaki Okui, Koichi Hagiwara, Masatsugu Nagai

The bridge high performance steels with advantages of high yield strength and good weldability have

been standardized as SBHS500 and SBHS700 in Japanese Industrial Standard (JIS) since 2008. These

SBHS steels exhibit different inelastic behavior from conventional steels, such as almost no yield

plateau and greater yield-to-tensile strength ratio. Hence, there is a need to examine the applicability of

SBHS to steel bridge structure and to develop design equations for SBHS.

In this paper, the compressive strengths of 4 edge simply supported normal and SBHS steel plates are

examined through parametric study employing elasto-plastic finite element analyses. In the parametric

study, 4 conventional steel grades SM400, SM490, SM490Y, SM570 and 2 SBHS steel grades

SBHS500 and SBHS700 were considered, and compressive strength of steel plates were evaluated at 10

slenderness parameter R values within the range from 0.4 to 1.4. The effect of different inelastic

behavior of different steel grades, the residual stress and the initial deflection, and the slenderness

parameter on the normalized compressive strength is discussed. It is found that the normalized

compressive strength of SBHS700 steel plates is about 6% greater than that of the SM400 steel plates

for R 0.7. The numerical analysis results with the conservative residual stress and initial deflection

levels show that the load-carrying capacity curve of Japan Specification for Highway Bridges (JSHB) is

unconservative for 0.45<R<0.75 and over conservative for R>0.8

In the second part of this paper, the mean value and standard deviation of the normalized compressive

strength is evaluated by using the response surface method and Monte Carlo simulation. To this end,

first, the response surface of the normalized compressive strength is identified based on 114 numerical

results for different residual stress and initial defection, where the response surface is assumed as a

simple algebraic function of the residual stress and initial deflection. Then, Monte Carlo simulation is

carried out to evaluate statistical distribution of the compressive strength. In this simulation, only the

residual stress and initial defection are considered as stochastic variables, and the probabilistic

distribution of residual stress and initial deflection are collected from previous experimental studies.

Comparing with the standard deviation obtained from experimental results, the standard deviation based

on this numerical simulation exhibits about half of that in a range of 0.6<R<1.2 and more reasonable

tendency.

Keywords: bridge high performance steels, compressive strength, residual stress, initial deflection, local

buckling

Page 2 of 10

Local Buckling Strength and its Statistical Information of Normal and Bridge

High Performance Steel Plates

Viet Duc DANG

PhD student, Dep. of Civil and

Environmental Eng., Saitama University,

Saitama Prefecture, Japan

Yoshiaki OKUI

Professor, Dep. of Civil and Environmental

Eng., Saitama University, Saitama

Prefecture, Japan

dangviet.duc@gmail.com

okui@mail.saitama-u.ac.jp

Koichi HAGIWARA

Master student, Dep. of Civil and

Environmental Eng., Saitama University,

Saitama Prefecture, Japan

Masatsugu NAGAI

Professor, Dep. of Civil and Environmental

Eng., Nagaoka University of Technology,

Nagaoka Prefecture, Japan

alexander.f.f.9@gmail.com

nagai@nagaokaut.ac.jp

1.

Introduction

Since 2008, the bridge high performance (SBHS) steels with advantages of high yield strength and

good weldability have been standardized as SBHS500 and SBHS700 in Japanese Industrial

Standard (JIS) [1]. However, SBHS steels exhibit different inelastic behaviour from conventional

steels, such as almost no yield plateau and greater yield-to-tensile strength ratio. Hence, it is

necessary to examine the applicability of the current compressive strength design equation of JSHB

[2] to steel plates with new steel grades.

The recent design specifications as Eurocode and AASHTO tend trend towards the partial safety

factor method. Regarding the predicting model on compressive strength of steel plates, in order to

determine the safety factors and the nominal compressive strength, statistical information, such as

the mean value and standard distribution of the compressive strength, is necessary. The current

study based on the Monte Carlo method and response surface to obtain probability distribution of

normalized compressive strength. The data of plate initial imperfections reported in [3] is

considered as the base of probabilistic characteristics of random inputs.

2.

1

z

a

rc

0.8

y

W0

u/ y

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

0.6

0.4

0.2

JSHB

SM400

SM490

SM490Y

SM570

SBHS500

SBHS700

0

x y

z(x,y) = W0 sin sin

a b

and sinusoidal initial deflection surface

0.3

0.6

0.9

1.2

1.5

with 6 steel grades in the case W0/b=1/400 and

rc/y=0.23

For the plate simulation, the steel material model applies assumptions of isotropic elasto-plastic

hardening, Von Mises yield surface and strain hardening hypothesis. The assumed uniaxial stressstrain relation of 6 steel grades is idealized from actual experiments. The four edges the of plate

model are assigned as simple supports. The residual stress and initial deflection are idealized as a

stress distribution and sinusoidal surface with surface function shown in Fig.1. The compression is

applied in the plate model by displacement control method. The ABAQUS S4R shell elements are

Page 3 of 10

applied for meshing the plate model with mesh size 30x30 elements.

The normalized compressive strength of steel plates (u/y) are compared between the mean value

of residual stress and the initial deflection from measurement data reported in [14] with rc/y=0.23,

W0/b=1/400. As shown in Fig.2 the compressive strengths of steel plates with six grades are quite

similar. The most significant difference occurs at R 0.7 and R 0.4, the normalized compressive

strength of SBHS700 steel plates (maximum value among six steel grades) is about 6% greater than

that of the SM400 steel plates (minimum value among 6 steel grades).For R > 0.4, the compressive

strength of SBHS steel plates with greater YR value is greater than that of the normal steel plate

with lower YR value. These results show the difference from conclusion in [4].

3.

plates is obtained by the application of a response

surface and a large number of random inputs of

0.8

residual stress and initial deflection. The

convergent mean and standard deviation values of

JSHB

0.6

compressive strength of steel plates are obtained

by processing 10000 random input couples of

M ean results

residual stress and initial deflection. In Fig.3, the

0.4

mean values of the normalized compressive

Fukumoto M curve

strength of steel plates with error bar equal to 2

0.2

Fukumoto M -2S curve

times of standard deviation values are plotted

along with the current JSHB design equation,

0

mean (M) curve and the mean minus 2 standard

0.0

0.2

0.4 0.6

0.8 1.0

1.2 1.4

deviation (M-2S) curve proposed in [3] .

R

As shown in Fig.3, the mean values of normalized

Fig. 3: Comparison between current study,

compressive strength of steel plates are similar to

JSHB [2], and Fukumoto el al [3] results

the mean curve reported in [3] which was

proposed from test results. Within the range 0.65R0.85, the mean values of the normalized

compressive strength of the current study are slightly greater than that of the mean curve reported in

[14]. The reason is that the current study doesnt consider steel plates with W0/b>1/150 and within

the mentioned range of R the influence of initial deflection on compressive strength is more

significant than other ranges. Also seen in Fig.3, the M-2S curve proposed in [14] is too

conservative if compared to corresponding results of the current study.

1

u/ y

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

4.

Conclusions

The mean values obtained in the current study are similar to the mean equation proposed in

Fukumoto et al (1984) [3] but slightly greater within the range 0.7<R<0.9.

The standard deviation values of compressive strength of the current study exhibits about half of

which was reported in [3] within a range of 0.6<R<1.2 and has a more reasonable tendency.

For R>0.4, for the same levels of R and initial imperfections, the compressive of steel plates is

similar, however; compressive strength of steel plates with SBHS grades with greater YR quantity

is slightly greater than that of normal grades, show the difference from conclusion in [4].

References

[1]

[2]

[3]

[4]

Japanese Industrial Standard, JIS G 3140, Higher yield strength steel plates for bridges,

2008

Japan Road Association, Specifications for Highway Bridges part II. Steel Bridges, 2002

FUKUMOTO. Y., and ITOH, Y., Basic compressive strength of steel plates from test data,

Proc. of JSCE No.334/I-1, 1984.

FUKUMOTO. Y., New constructional steels and structural stability, Engineering

structures, Vol.18, 1996, pp. 786-791

Page 4 of 10

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

Local Buckling Strength and its Statistical Information of Normal and Bridge

High Performance Steel Plates

Viet Duc DANG

PhD student, Dep. of Civil and

Environmental Eng., Saitama University,

Saitama, Japan

Yoshiaki OKUI

Professor, Dep. of Civil and Environmental

Eng., Saitama University, Saitama, Japan

okui@mail.saitama-u.ac.jp

dangviet.duc@gmail.com

Koichi HAGIWARA

Master student, Dep. of Civil and

Environmental Eng., Saitama University,

Saitama, Japan

Masatsugu NAGAI

Professor, Dep. of Civil and Environmental

Eng., Nagaoka University of Technology,

Nagaoka, Japan

alexander.f.f.9@gmail.com

nagai@nagaokaut.ac.jp

Summary

The current Japanese design equation for load-carrying capacity of compressive steel plates is

examined for four conventional steel grades and new bridge high performance steel grades

SBHS500 and SBHS700. The analysis results of FEM plate models with initial imperfections show

that the current Japanese design specification is un-conservative within the range 0.5<R<0.75 and

over-conservative in the range R>0.8, where R is the slenderness parameter. Statistical distribution

of the normalized compressive strength is obtained by means of Monte Carlo simulation in

combination with the response surface which is obtained from sufficient number of FEM plate

model analyses. Both initial out-of-plane displacement and residual stress are considered as sources

of variability. The mean values of normalized compressive strength in this study are similar to those

obtained from experimental tests [14]. The standard deviation of the current study exhibits about

half of the experimental results [14] within the practical range 0.6<R<1.2.

Keywords: bridge high performance steels, compressive strength, residual stress, initial deflection,

local buckling.

1.

Introduction

Box columns and box plate girders consisting of unstiffened steel plates are widely used in bridge

structures. The local buckling strength of the steel plates frequently governs the load-carrying

capacity of these structural elements.

The current compressive strength design equation for unstiffened plates in Japanese Specifications

for Highway Bridge (JSHB) version 2002 [2] has been originally proposed in 1980 [3]. This

equation was based on experimental data for normal steel with yield strengths mainly less than 450

MPa. The bridge high performance steels, which poses high yield strength and good weldability,

have been standardized since 2008 as SBHS500 and SBHS700 in Japanese Industrial Standard (JIS)

[4]. However, SBHS steels exhibit different inelastic behaviour from conventional steels, such as

almost no yield plateau and greater yield-to-tensile strength ratio. Hence, it is necessary to examine

the applicability of the current compressive strength design equation of JSHB to steel plates with

new steel grades.

Regarding the compressive strength design equation of JSHB, Usami and Fukumoto [5], Usami [6],

and Kitada et al. [7] show that the design equation is un-conservative within the range 0.5<R<0.75

(intermediate range) and over-conservative in the range R>0.8 (slender range), in which R is the

slenderness parameter. However, studies [5], [6] consider only the normal (SM490Y) steel plates

and they employed the perfectly elstoplastic assumption for modeling the inelastic behavior of steel

material.

For theses reasons, the compressive strength design equation of JSHB need to be examined and

Page 5 of 10

developed for SBHS steel grades. However all the referred studies were based on the deterministic

method. The recent design specifications trend towards the partial safety factor method [9], in

which a safety factor separates into individual causes, such as the variability on material property

and the confidence of strength prediction method. The partial safety factor method with probabilitybased partial factors has been employed in Eurocode [10], [11] and AASHTO [12] as well. In order

to determine the safety factors and the nominal compressive strength, statistical information, such

as the mean value and standard distribution of the compressive strength, is necessary. Komatsu and

Nara [13] carried out a number of FE analyses of steel plates with considering initial deflection

based on collected measurement data from actual steel bridges to obtain the mean value and

standard deviation of compressive strength. However, this study considered only residual stress

rc/y=0.3 as a deterministic quantity, where rc is the compressive residual stress. Fukumoto and

Itoh [14] proposed the mean (M) and mean minus twice standard deviation (M-2S) curves of the

compressive strength on the basis of a database on single plate and box column compression test

data. The M and M-2S curves also show that the current JSHB design equation for steel plates

under compression is un-conservative within intermediate range and over-conservative in slender

range. Fukumoto and Itoh [14] also reported the statistical distribution of residual stress and initial

deflection. However, this study also considered the steel plates with initial deflection W0/b > 1/150,

where W0 and b are the maximum initial deflection and plate width, respectively.

This paper intends to examine the current JSHB design equation of steel plate compressive strength

for normal and SBHS steels, and to evaluate the mean and standard deviation of compressive

strength.

2.

2.1

Plate properties

Four normal steel grades SM400, SM490, SM490Y,

SM570 and two bridge high performance steel

SBHS500, SBHS700 are considered in the current study.

Fig. 1 shows yield strengths and slenderness parameters

R considered in the FE analyses, where R is defined by

800

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

600

400

R=

200

SM400

SM490

SM490Y

SM570

SBHS500

SBHS700

0

0.3

0.5

0.7

0.9

1.1

1.3

1.5

b y 12(1 2 )

.

t E

2k

(1)

thickness, yield strength, elastic modulus, poison ratio,

and buckling coefficient, respectively. The aspect ratio

of all steel plates is assigned to 1.

yield strength in FE analyses

2.2

FEM model

Nonlinear FE analysis considering both material and geometric nonlinearity is conducted. PrandtlReuss equation is employed to model the steel plasticity. The idealized uniaxial stress-strain

relationships used in the numerical analyses are shown in Fig.1.

Regarding the boundary condition, all four edges of a plate model are assigned as simple supports.

Figure 3 shows the distributions of residual stress and initial deflection assumed in the FE analysis.

The probabilistic distributions of residual stress and initial deflection are based on measurement

data reported in [14].

The displacement control method is used to apply the compressive stress. ABAQUS S4R shell

elements are used for plate FE model with mesh size of 30x30 elements.

Page 6 of 10

800

z

a

1 1 (MPa)

600

rc

400

SBHS700

SBHS500

SM570

SM490Y

SM490

SM400

0.05

0.1

0.15

W0

0

0

200

x y

sin

a b

z(x,y) = W0 sin

0.2

1 1

relations of steel grades considered

in current study

distribution and sinusoidal initial

deflection surface

2.3

Comparison of FE results with experimental results

The normalized compressive strengths obtained from the FE analyses as well as past experimental

results [1, 15] are plotted in Fig. 4 as a function of R. In these analyses, an initial deflection W0/b =

1/150 and a residual stress rc/y = 0.4 are considered as a conservative assumption. As shown in

Fig.4, the FEM results lay the lower bound of experimental results, which corresponds to the

conservative assumption.

1.2

1

1

0.8

u/ y

0.8

u / y

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

0.6

JSHB

Dorman Long

Chin

SM400

SM490Y

SBHS500

0.4

0.2

0.6

0.4

Rasmussen

Harrison

Moxham

SM490

SM570

SBHS700

0.2

JSHB

SM400

SM490

SM490Y

SM570

SBHS500

SBHS700

0

0.0

0.5

1.0

results to previous test results

1.5

0.3

0.6

0.9

1.2

1.5

of plate with 6 steel grades in the case

W0/b=1/400 and rc/y=0.23

2.4

Compressive strength of different steel grades

In this section, the normalized compressive strengths of different steel grades are compared in the

case of mean values of the residual stress and the initial deflection. The mean values of normalized

residual stress and initial deflection are obtained as rc/y=0.23 and W0/b=1/400, respectively from

the measurement data reported in [14]. In the evaluation of the mean value of the normalized initial

deflection, the measurement data for W0/b>150 are excluded owing to an allowable fabrication

upper limit in JSHB.

As shown in Fig.5, the compressive strengths of steel plates with 6 steel grades are quite similar in

the whole range of R. The largest difference occurs at R 0.7 and R 0.4, and the normalized

compressive strength of SBHS700 steel plates (maximum value among 6 steel grades) is about 6%

Page 7 of 10

greater than that of the SM400 steel plates (minimum value among 6 steel grades). For R > 0.4, the

compressive strength of SBHS steel plates with larger YR value is greater than that of normal steel

plate with lower YR value.

3.

3.1

Response surface

numerical results

1.1

1

u / y

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

0.9

0.8

0.7

0

rc / y

0.5

0.005

W0/b

0.01

0.015

compressive strength, Monte Carlo method is

employed. However, to obtain a convergent result

in the Monte Carlo simuration, it is essential to

implement a large number of deterministic

analyses, and accordingly it would take long time.

Hence, to overcome this problem, a response

surface of the normalized compressive strength,

which is an approximate algebraic function of the

initial deflection and residual stress, is instead of

the deterministic analyses. The current study

employs ten response surfaces corresponding to

the 10 considered R values.

presented along with FEM results for the

case R = 0.8

The response surface is expressed as a simple algebraic function,

u = p 00 + p10 r + p 01W0 + p 20 r + p11 r W0 + p 02W0 + p 21 r W0 + p12 r W0 + p 30 r + p 03W0

2

(2)

where u = u / y , W0 = W0 / b and r = rc / y . The constants pij in Eq.(2) were determined from a set

of 114 deterministic FE results for each steel grade and R value by using the least square method.

Fig.6 shows an obtained response surface along with FE numerical results for R=0.8. All constants

of ten response surfaces are presented in Table 1. The obtained response surfaces show good fit for

numerical results in the cases R0.7 with the coefficient of determination (R-square) > 95%. For

R<0.7, the R-square values become slightly lower due to the influence of the hardening of high

strength steels (SM570, SBHS500, SBHS700).

Table 1: Constant values of 10 Response surfaces

R value

0.40

0.50

0.60

0.70

0.80

0.92

1.04

1.16

1.28

1.40

p00

1.098

1.034

1.012

1.037

1.047

0.963

0.850

0.757

0.697

0.650

p01

-40.22

-15.15

-1.72

-34.54

-65.98

-40.44

-23.31

-12.66

-10.08

-7.44

p02

5442.0

1683.0

-1894.0

3169.0

8382.0

3081.0

1188.0

377.6

367.4

229.3

p03

-248100

-90660

100500

-135000

-389100

-112100

-34700

-11430

-12430

-8387

p10

0.007

-0.012

-0.087

-0.234

-0.584

-0.937

-0.788

-0.567

-0.435

-0.343

p11

-2.320

-8.400

-21.150

-32.100

-0.520

57.210

47.270

25.720

23.330

18.580

p12

-48.4

-42.4

507.9

1465.0

1309.0

-891.1

-769.2

46.8

-269.2

-208.5

p20

-0.007

0.069

0.284

0.568

1.006

1.368

1.169

0.856

0.617

0.462

p21

2.25

6.72

11.01

9.74

-15.13

-35.91

-29.12

-20.27

-15.11

-11.63

p30

0.000

-0.057

-0.198

-0.342

-0.492

-0.632

-0.556

-0.419

-0.288

-0.211

Page 8 of 10

3.2

Stochastic inputs of initial imperfections

In the Monte Carlo simulation, the probabilistic distribution of the initial deflection is assumed as

the Weibull distribution, and the residual stress as the Lognormal distribution as shown in Fig.7 and

8 [14]. In this figure, the generated random variables in the Monte Carlo simulation are also plotted

as the histogram chart. The generated normalized initial defections more than W0/b>1/150 are

excluded in the simulation due to an allowable fabrication upper limit in JSHB.

0.10

0.25

0.08

0.07

0.06

Relative Frequancy

Weibull dist.

=0.00280

=0

m=1.37

=0.00233

=0.00156

0.09

Relative Frequancy

0.05

0.04

0.03

0.02

Lognormal dist.

=0.23

=0.145

0.20

0.15

0.10

0.05

0.01

0

0

W0/b

0

0

7

x 10

-3

0.2

0.4

/

rc

deflection

0.6

0.8

1.0

residual stress

3.3

Results of Monte Carlo simulation

The probabilistic distribution of compressive strengths is obtained by application of the response

surface and a large number of random inputs of residual stress and initial deflection. The converged

mean and standard deviation are obtained by processing 10000 random input couples of residual

stress and initial deflection. The histogram of compressive strength with 10000 random input

couples in the case of R=0.8 is shown in Fig.9. Results of 10 convergent mean and standard

deviation of compressive strengths are presented in Table 2.

0.1

Relative Frequancy

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

0.08

0.06

0.04

0.02

0

0.5

0.6

0.7

0.8

0.9

/

u

1.0

1.1

1.2

R value

0.4

0.5

0.6

0.7

0.8

0.92

1.04

1.16

1.28

1.4

Page 9 of 10

M value

S value

1.039

1.006

0.982

0.938

0.862

0.766

0.701

0.653

0.617

0.586

0.0277

0.0157

0.0212

0.0413

0.0515

0.0399

0.0294

0.0216

0.0171

0.0145

In Fig.9, the mean values of normalized compressive strength with error bar equal to two times the

standard deviation are plotted along with the current JSHB design equation, mean (M) curve and

mean minus 2 standard deviation (M-2S) curve proposed in [14] .

0.20

0.20

1

Standard deviation

Current study

0.15

u/ y

0.8

u/ y

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

JSHB

0.6

M ean results

0.4

Fukumoto M curve

0.10

0.05

0.2

0.15

0.10

0.05

0.00

0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.00

0.3

0.6

0.9

1.2

1.5

study, JSHB [2], and Fukumoto el al.

[14] results

As shown in Fig.9, the mean values of normalized compressive strength are similar to the mean

curve reported in [14], which was proposed from test results. Within 0.65R0.85, the mean values

of the current study are slightly greater than that reported in [14]. One of the possible reasons is that

the current study does not consider steel plates with W0/b>1/150 and the influence of initial

deflection on compressive strength is more significant within the mentioned range of R than other

ranges. Also seen in Fig.9, the M-2S curve proposed in [14] is too conservative if compared to

corresponding results of the current study.

A comparison of standard deviation of compressive strengths obtained in the current study and

previous study [14] is presented in Fig.10. Within the practical range 0.6<R<1.2, the standard

deviations of the current study exhibit about half of reported values in [14]. As seen in the figure,

the standard deviations obtained by the current study have a clearer tendency, and attained to the

maximum value at R0.8. Within the range 0.7<R<0.9, the ultimate state of compressive plates is

elastoplastic buckling, and accordingly the compressive strength is significantly influenced by both

residual stress and initial deflection. For R>0.9, the standard deviation decreases, and the elastic

buckling becomes dominant. The residual stress has almost no effect on the compressive strength.

For R < 0.7, the compressive strength tends to attain the yield strength and is mainly influenced by

initial deflection. In particular for R<0.5, the hardening behavior of high strength steel (SM570,

SBHS500 and SBHS700) starts to have a significant effect on the compressive strength.

4.

Conclusions

The mean values obtained in the current study are similar to those proposed in [14], but slightly

greater within the range 0.7<R<0.9. The standard deviation of compressive strength obtained in this

study is about half of values reported in [14] within a range of 0.6<R<1.2. The M-2S curve

proposed in [14] is too conservative if compared to corresponding values obtained in the current

study. The results of M-2S of the current study show that the current load-carrying-capacity JSHB

design equation of steel plates is un-conservative within the range 0.5 < R < 0.8 and overconservative for R > 0.85.

0.3

0.6

Fig. 10: Co

deviation v

study and r

Page 10 of 10

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

Within the range of slenderness parameter 0.4R1.4, the normalized compressive strengths of

normal steel and SBHS steel are similar for the same levels of R and initial imperfections. However,

compressive strength of steel plates with SBHS grades with greater YR is slightly greater than that

of normal steel grades.

References

[1]

DWIGHT J. B., MOXHAM K. E., Welded steel plates in compression, The Structural

Engineer, Vol. 47, No. 2, 1969, pp. 49-66.

[2]

[3]

Japan Road Association, Specifications for Highway Bridges part II. Steel Bridges, 2002

Japan Road Association, Specifications for Highway Bridges part II. Steel Bridges, 1980

[4]

Japanese Industrial Standard, JIS G 3140, Higher yield strength steel plates for bridges,

2008

[5]

USAMI. T., and FUKUMOTO. Y., Deformation analysis of locally buckled steel

compression members, Journal of Constructional Steel Research, Vol.13, 1989, pp. 111135.

[6]

USAMI. T., Effective width of locally buckled plates in compression and bending, Journal

of Structural Engineering, ASCE, Vol. 119, No. 5, 1993, pp. 1358-1373.

KITADA. T., YAMAGUCHI. T., MATSUMURA. M., OKADA. J., ONO. K. and OCHI, N.,

New technology of steel bridge in Japan, Journal of Constructional Steel Research, Vol.58,

2002, pp. 21-70.

FUKUMOTO Y., New constructional steels and structural stability, Engineering

structures, Vol.18, 1996, pp. 786-791.

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

International Standard ISO 2394, General principles on reliability for structures, Third

edition, 1998

CEN, Eurocode3, Design of Steel Structures, Part 1-5, plated structural elements, 2004

CEN, Eurocode4, Design of Composite Steel and Concrete Structures, Part 2, General

rules and rules for bridges, 1994

American Association of State Highway and Transportation Officials, AASHTO LRFD

bridge design specification Fourth edition, 2007

KOMATSU S., and NARA. S., Statistical study on steel plate members, Journal of

Structural Engineering, ASCE, Vol. 109, No. 4, 1983.

FUKUMOTO Y., and ITOH, Y., Basic compressive strength of steel plates from test data,

Proc. of JSCE No.334/I-1, 1984.

RASMUSSEN K. J. R., and HANCOCK. J. G., Plate slenderness limits for high strength

steel sections, J. Construct. Steel Research. Vol.23, 1992, pp. 73-96

- Block Shear Failure Seminar ReportUploaded bypraxie
- Strength Consideration in Product Design, Shiva,DeepakUploaded byShivaprasad.P
- Ultimate StrengthUploaded byJim Gray
- Swedish Regulations for Steel BSK99Uploaded byAmatfr
- Development of Coiled Tubing Stress Analysis[1]Uploaded bytofa01
- Fea for BackhoeUploaded bymcrajpura
- CAR FRONTAL IMPACTUploaded byapi-3762972
- purlinUploaded bymukasa1971
- Uncofined TestUploaded bywanfaiz123
- HW1.1Uploaded byWilliam Mata
- Yield Curvature for Seismic Design of Circular Reinforced Concrete Columns-Sheikh (2010)Uploaded byAlberto Ezequiel León Tamayo
- Tests on RocksUploaded byPartkunan Niroshkanth
- MEE06013_Khan_Shahzada_2007[1]Uploaded bys_murugan02
- EnsayosAlSikadur AnchorFix-4Uploaded bydadva
- Is Std References ListUploaded byManjari Pujar
- Griffiths Theory . Hw6Uploaded byRawandAlbarody
- Zhu 2017Uploaded byMặc Thế Nhân
- short delay timesUploaded byenrike zamora
- PM-PHIUploaded byMrinal Agarwal
- MDP 1 Lecture 1 Design Methods RevisedUploaded byYahya Faiez Waqqad
- XiaoChen Published Energies 07 02274Uploaded byneuchenxiao
- Kisssoft Tut 003 E KeyUploaded byreinaldosidejunior
- 1. Csr for Double Hull Oil Tankers Xls523Uploaded bymvjayaram007
- STM MODEL 3Uploaded byJester Abucay
- 35prUploaded bybtcong93
- 10c-Beam Flanged T Multi Span 01.PngUploaded byParthi Ban
- Deepbeam Si AashtoUploaded byahm2011
- TEXT.pdfUploaded byDurga Babu
- NCR 36-02Uploaded byyrdgcgf5664
- j1360Uploaded byRavi Khandelwal

- 135640720 FB MultiPier Help ManualUploaded byViet Duc Dang
- Vb2011 Angol Kovacs-szalai Page58-65Uploaded byViet Duc Dang
- Anh Duc Paper 13.02Uploaded byViet Duc Dang
- FFFEABEDd01Uploaded byViet Duc Dang
- SoilWorks Verification SummaryUploaded byViet Duc Dang
- Kovesdi Balazs PhdUploaded byViet Duc Dang
- ari_45Uploaded byViet Duc Dang
- Non-Linear Time History Analysis of Cable Stayed BridgesUploaded bypraxie
- MODERN PRESTRESSED BRIDGESUploaded bycivengbridge64
- FFT Algorithms.pdfUploaded byIqbal Novramadani
- Structural RetrofitUploaded byJr Remotigue
- Winter 2010 PCI Journal RosignoliUploaded byDave Thompson
- JGC15 Standard Specifications Design 1.0Uploaded byViet Duc Dang
- Repairs, Modifications, And Strengthening With Post-Tensioning, PTI Journal, July 2006Uploaded byViet Duc Dang
- 1-s2.0-S0167473004000505-mainUploaded byViet Duc Dang
- 1-s2.0-S0167473002000073-mainUploaded byViet Duc Dang
- 15 Load and Resistance Factor Rating LRFR.pdfUploaded byViet Duc Dang
- Calibration of Load Factor for LRFR Load EvaluationUploaded byViet Duc Dang
- 1-s2.0-S0045794904000811-mainUploaded byViet Duc Dang
- 1-s2.0-S004579490400080X-mainUploaded byViet Duc Dang
- CAU BTCTUploaded byViet Duc Dang
- Dang Viet Duc S09DE052 Research Report of DepartmentUploaded byViet Duc Dang
- KSSC13-3(0)13-064Uploaded byViet Duc Dang
- CauTreoDayVong MIDASUploaded byqhuyhn1982
- Prestressed Concrete Bridges - Nigel HewsonUploaded byNguyễn Xuân Nguyên
- 2010_Bridges_AnalysisandModelling_LDavaineUploaded bymamad66
- Bridges and Civil Engineering StructuresUploaded byapi-3770279
- UCG-ES-00163Uploaded byViet Duc Dang
- Plate Buckling eUploaded byViet Duc Dang

- ContentsUploaded byAfia S Hameed
- concrete coring test 2Uploaded byChoon Ping Yeoh
- Design of Rcc DrainUploaded byvishnumani3011
- Reactive Powder Concrete Properties With Cement Replacement Using Waste MaterialUploaded byAvish Gunnuck
- Volume (11) Issue (5) 684 - 693.pdfUploaded byAmit Pathak
- D4219Uploaded bysmanoj354
- LExan 9034 - Sabic.pdfUploaded bytesttest
- 2Uploaded byKartickPai
- Baru Brick Test Final Lab ReportUploaded byamar aiman
- Durability Properties of Banana Fibre Reinforced Fly Ash ConcreteUploaded byIRJET Journal
- S1Uploaded byAHMED SOHEL
- Is 516 Method of Test for Strength of ConcreteUploaded byselva_651950
- Ground ImprovementUploaded byMohamed
- Experimental Investigation of Fiber Reinforced Concrete Beams.pdfUploaded byK.sumithira
- Partial Replacement of Fine Aggregate with Iron Ore Tailings and Glass PowderUploaded byIJSTE
- A study on strength characteristics of glass fibre reinforced high performance-concreteUploaded byAnonymous CUPykm6DZ
- ASR - Adobe Construction (Turkey)Uploaded byJohn Johnson
- Experimental Study on the Mechanical Behaviour of Sisal Fibre Reinforced Self-Compacting ConcreteUploaded byAnonymous kw8Yrp0R5r
- Issues in Evaluating Capacity of Rock Socket FoundationsUploaded bypaduco
- USE OF FRP COMPOSITES FOR STRENGTHENING OF RCC COLUMNSUploaded byIJIERT-International Journal of Innovations in Engineering Research and Technology
- Utilization of Demolished Concrete Waste for New Construction and Evaluation of Its StrengthUploaded byIRJET Journal
- Lab Manual Ce324 Sm IIUploaded byHafiz Ur Rehman
- Parametric Study for Assessing the Effects of Coarseness Factor and Workability Factor on Concrete Compressive StrengthUploaded byJaga Jay
- New Rich Text DocumentUploaded byRubina Fatima
- MergeUploaded byOga Meo
- Point Load Strength Index Test of Rock (is-8764)Uploaded bygorkem
- 2010 - Ultra high strength fiber reinforced concrete using aramid fiber.pdfUploaded byMonica Natalia Gavilanez Rivera
- Experimental Study on Use of Quarry Dust and Fly Ash with Partial Replacement of Fine Aggregates and Cement in ConcreteUploaded byGRD Journals
- 11_chapter 5.pdfUploaded byBpdplanning Madhucon
- 592448.Condition_assessment_and_repair_of_antenna_towers_concrete_foundations.pdfUploaded byM.Rezvani