Elsevier

Ocean Engineering

Volume 221, 1 February 2021, 108374
Ocean Engineering

Optimal design of longitudinal stiffeners of unsymmetric plate girders subjected to pure bending

https://doi.org/10.1016/j.oceaneng.2020.108374Get rights and content

Highlights

  • The optimum stiffener location of the plate with a single stiffener and two stiffeners is at around 0.42Dc and around 0.25Dc and 0.56Dc, respectively, regardless of the asymmetry of the girder section.

  • γrq obtained according to the AASHTO LRFD requirements is too conservative in cases ϕ> 1.0 but not conservative in cases of ϕ1.0.

  • The formulas for determining the minimum flexural rigidity requirement of a single and two stiffeners are suggested for practical design purposes.

Abstract

This paper aims at investigating the optimum positions and the required flexural rigidity (γrq) of a single and two longitudinal web stiffeners of unsymmetric plate girders subjected to pure bending by using gradient-based optimization algorithms. The optimization procedure is performed into two steps: The first step is to maximize the bend-buckling coefficient kb generated from eigenvalue buckling analyses, and then the second step aims to minimize the flexural rigidity of the stiffeners (γ). This procedure is implemented using the Abaqus2Matlab toolbox which allows for the transfer of data between Matlab and Abaqus and vice versa. Based on this research, the optimum locations of the stiffeners are recommended for the unsymmetric plate girders under pure bending. Additionally, equations are suggested to determine γrq of the longitudinal web stiffeners of the unsymmetric plate girders. These proposed equations are more realistic in the practical designs than those recommended by AASHTO LRFD specifications and several works in the literature in which γrq of the longitudinal web stiffeners is determined based on the simply supported plates but not the unsymmetric plate girders under bending.

Introduction

In deep plate girders having slender webs, longitudinal stiffeners are widely used since the presence of the longitudinal stiffeners can enhance the web strength against bend-buckling due to bending stresses by controlling the lateral deflection of the girder webs. Additionally, longitudinally stiffened web can provide improved restraint to the rotation of the compression flanges and consequently result in greatly increased bending strength (Ziemian, 2010; Myung et al., 2016). Research related to longitudinal stiffeners of the plate girders, especially their optimum location and γrq under pure bending, has been extensively conducted in the literature. Plate girders with stiffeners are a common structural component in various structures, e.g. bridges and ship hulls. In the latter case, stiffened panels and box girders are usually of critical importance for the ultimate strength of conventional cargo ships under longitudinal bending, the design of SWATH (Small Waterplane Area Twin Hull) ships against transverse bending due to their special sectional configuration (Liu et al., 2018), and the twin-hull inland catamarans which require special design against longitudinal bending due to their shallow draught (Xu et al., 2019).

For the steel plate girder with a single longitudinal stiffener, many researches have proven that the optimum position of the stiffener is at 0.2D from the compression flange for the symmetric girder assuming longitudinal edges of girder webs are simply supported (Cooper, 1967; Azhari and Bradford, 1993; Frank and Helwig, 1995; Maiorana et al., 2011; Vu et al., 2019a). Among these researches, Frank and Helwig (1995) conducted a series of eigenvalue buckling analyses for the optimum stiffener position of the unsymmetric girder, in which the longitudinal edges of the web were presumed to be simply supported. Based on their research, they suggested that the optimum stiffener position is at 0.4Dc (Dc is the web depth in compression in the elastic range), regardless of the asymmetry of the girder section. In addition, they also separately proposed the formulations computing kb of the stiffened webs depending on the stiffener location (ds/Dc0.4 and ds/Dc<0.4), which have been adopted by AASHTO LRFD (AASHTO, 2012). The AASHTO LRFD (AASHTO, 2012) is the national standard for the design and construction of bridges in the United States published by the American Association of State Highway and Transportation Officials (AASHTO). The AASHTO LRFD specifications (AASHTO, 2012) for kb of a longitudinally stiffened web plate girder are based on the research of Frank and Helwig (1995). The equations and the associated optimum stiffener position adopted by AASHTO LRFD (AASHTO, 2012) assume simply supported boundary conditions at the flanges.

Performing numerical investigation of the entire steel plate girder under pure bending, Cho and Shin (2011) reported that the optimum stiffener location is achieved when it is located at around 0.425Dc from the compression flange. They suggested new equations to calculate kb for cases when ds/Dc0.425 and ds/Dc<0.425. More recently, Elbanna et al. (2014) and Kim et al. (2018) recommended that the optimum position of the stiffener is at around 0.42Dc regardless of the asymmetry of the girder section. It is noteworthy that the optimum stiffener location suggested by Cho and Shin (2011), Elbanna et al. (2014), and Kim et al. (2018) is slightly different from AASHTO LRFD (AASHTO, 2012) since they considered the effect of rotational restraints along the web edges due to the presence of the compression flanges.

Regarding the steel plate girder with multiple longitudinal stiffeners, researches related to its optimum locations and γrq are very limited. Rockey and Cook, 1965a, 1965b conducted an investigation on the optimum locations of multiple longitudinal stiffeners for the plate girders with doubly-symmetric section subjected to pure bending. In their model the longitudinal edges of the girder web were assumed to be either simply supported or clamped while the vertical edges of the girder web were presumed to be simply supported. It was also assumed that the longitudinal stiffeners were symmetrically located about the mid-plane of the girder web and their flexural rigidity was finite while torsional rigidity was negligible. Based on their results, the optimum positions of multiple stiffeners (up to six) were suggested. Rockey and Cook (1965b) also proposed equations to calculate γrq of the stiffeners of the girder webs with aspect ratio lower than 1.6.

The AASHTO LRFD (AASHTO, 2012) bridge design specification only mentions about kb, the optimum location and γrq of a single stiffener but not multiple stiffeners. It has been reported by Viet et al. (Vu et al., 2019a) that by using two stiffeners, kb can increase by as much as 180%, while the web thickness necessary for design decreases at least by 61.76% compared to the case in which only a single stiffener is used. However, engineers still use the AASHTO LRFD provision to determine kb of the web plates with multiple stiffeners. Therefore, in practical design, this provision leads to uneconomical design of plate girders with more than one stiffener.

As stated above, the optimum stiffener position has been completely investigated for the girders with a single stiffener but not for multiple stiffeners. However, the γrq of one stiffener for the steel plate girders has not been fully studied since the AASHTO LRFD (AASHTO, 2012) recommends γrq for the plate girders based on the results obtained from simply supported plates. Moreover, to the best of the authors’ knowledge, the issue of the optimum location and γrq of multiple stiffeners for the unsymmetric plate girders in which the influence of the flanges are taken into consideration has not yet been adequately addressed in the literature.

This research aims to examine the optimum locations and γrq of a single and two longitudinal stiffeners for the unsymmetric plate girders under pure bending using various optimization procedures. At first, an eigenvalue buckling analysis is performed using Abaqus (ABAQUS, 2014), after that the optimization algorithm is employed to firstly maximize kb until the optimum stiffener positions are attained, and then the same algorithm is used to minimize γ until the optimum stiffener dimensions are achieved. The Abaqus2Matlab toolbox (Papazafeiropoulos et al., 2017a), which integrates between Abaqus (ABAQUS, 2014) and Matlab (MathWorks and Inc.R, 2017) in a loop, is utilized for the optimization procedure. Based on this study, the optimum locations of a single and two longitudinal stiffeners are suggested for the unsymmetric plate girders subjected to pure bending. In addition, formulas to determine γrq of the longitudinal stiffeners are recommended for practical design purposes. The results achieved from these formulas are compared with AASHTO LRFD requirement (AASHTO, 2012) and previous works to demonstrate their efficiency.

Section snippets

Linear elastic buckling analysis

Research related to elastic buckling analysis of stiffened plates have been extensively implemented (Ranji, 2013; Song et al., 2019; Papazafeiropoulos et al., 2019; Yang et al., 2020). In this study, the procedure for the solution of the problem of linear elastic buckling of the stiffened plate girders that was followed includes two analysis components:

  • a)

    A standard linear perturbation static analysis is firstly carried out in which a (unit) reference load pattern F is applied to the model. The

Verification of FE model

To demonstrate the accuracy of developed finite element (FE) models, results obtained from these models are compared with previous works. The results taken from Model 1 are compared with those given in the paper of Elbanna et al. (2014), while the results of Model 2 are used to compare with the results of Elbanna et al. (2014) and Cho and Shin (2011). For the Model 1, it can be seen from Table 1 that the results obtained from present work agree well with those obtained from Elbanna et al. (2014)

Proposed optimization algorithms

In this section, the efficiency of three optimization algorithms, namely IPA, SQP, and AS are investigated on the Model 1 and Model 2 with ϕ = 0.52, 0.76, 1, and 1.6 to choose the algorithm with the best performance to be used for all subsequent optimization analyses. Table 4, Table 5 report the comparison of three algorithms about running time, optimum location and γrq of a stiffener for the Model 1 and Model 2, respectively. It can be seen from these tables that all algorithms yield the same

Steel plate girders reinforced with a single longitudinal stiffener

In this section, the optimum position and γrq of a single stiffener are examined for the Model 1 and Model 2 using the optimization procedure presented in Section 2.3. Various web thicknesses of 9, 10, 12, and 15 mm corresponding to web slenderness ratios, λw of 333, 300, 250 and 200 respectively are considered to investigate the effect of λw, while different top flange thicknesses of 54, 45, 36, and 25 mm corresponding to top flange slenderness ratios, λfc of 11.11, 13.33, 16.67 and 24

Steel plate girders reinforced with two longitudinal stiffeners

This section investigates the optimum locations and γrq of two stiffeners for the Model 1 and Model 2. Since the AASHTO LRFD (AASHTO, 2012) mentions that for a single longitudinal stiffener λw 300, for the girder reinforced by two stiffeners, the values of 300 and 333 of λw (λw 300) are taken into consideration to study the effect of λw on γrq of the stiffeners. Different tfc of 54, 45, and 25 mm corresponding to λfc equal to 11.11, 13.33, and 24 respectively are taken into account to examine

Conclusions

In this study, a comprehensive work related to the optimum location and γrq of longitudinal stiffeners placed at one side of the web of steel plate girders subjected to pure bending is performed using various gradient-based optimization algorithms. The optimization procedure is implemented into two steps: The first step is to maximize kb and the second step is to minimize γ. The Abaqus2Matlab toolbox which integrates between Abaqus and Matlab is used for this optimization procedure. The results

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research is funded by Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number 107.01-2019.322.

References (46)

  • V.H. Truong et al.

    Effect of multiple longitudinal stiffeners on ultimate strength of steel plate girders

    Structure

    (2019)
  • Q.V. Vu et al.

    Effect of intermediate diaphragms on the load-carrying capacity of steel-concrete composite box girder bridges

    Thin-Walled Struct.

    (2018)
  • Q. Vu et al.

    Optimum linear buckling analysis of longitudinally multi-stiffened steel plates subjected to combined bending and shear

    Thin-Walled Struct.

    (2019)
  • Q. Vu et al.

    Bending-buckling strength of steel plates with multiple longitudinal stiffeners

    J. Constr. Steel Res.

    (2019)
  • S. Xu et al.

    Experimental and numerical analysis of ultimate strength of inland catamaran subjected to vertical bending moment

    Ocean Eng.

    (2019)
  • B. Yang et al.

    In-plane impact-induced elastic buckling behaviour of rectangular transversally isotropic stiffened plates with all edges elastically restrained against rotation

    Int. J. Mech. Sci.

    (2020)
  • AASHTO

    AASHTO LRFD, Bridge Design Specifications, Sixth

    (2012)
  • ABAQUS

    Analysis User's Manual Version 6.14

    (2014)
  • C.G. Broyden

    The convergence of a class of double-rank minization algorithms

    J. Math. Its Appl.

    (1970)
  • R.H. Byrd et al.

    An interior point Algorithm for large-scale nonlinear programming

    SIAM J. Appl. Math.

    (1999)
  • R.H. Byrd et al.

    A trust region method based on interior point techniques for nonlinear programming

    Math. Program. Ser. B.

    (2000)
  • E.Y. Cho et al.

    Elastic web bend-buckling analysis of longitudinally stiffened I-section girders

    Int. J. Steel Struct.

    (2011)
  • P.B. Cooper

    Strength of longitudinally stiffened plate girders

    J. Struct. Div.

    (1967)
  • Cited by (0)

    View full text