You are on page 1of 12

International Journal of Mechanical Sciences 92 (2015) 194205

Contents lists available at ScienceDirect

International Journal of Mechanical Sciences


journal homepage: www.elsevier.com/locate/ijmecsci

Effect of reinforcement on buckling and ultimate strength


of perforated plates
Jeong-Hyeon Kim a, Jun-Hwan Jeon a, Joo-Shin Park b, Hyang-Duk Seo c, Hyung-Joon Ahn c,
Jae-Myung Lee a,n
a
Department of Naval Architecture and Ocean Engineering, Pusan National University, 30, Jangjeon-Dong, Geumjeong-Gu, Busan 609735, Republic of Korea
b
Samsung Heavy Industries, Co., Ltd, Geoje 656710, Republic of Korea
c
Sungdong Heavy Industries, Co., Ltd, Tongyeong 650827, Republic of Korea

art ic l e i nf o a b s t r a c t

Article history: Cutouts are widely used in ships and offshore structures. These cutouts are used mainly for inspection,
Received 14 March 2014 and they may be tted for various purposes, including passing pipes and weight reduction. In general,
Received in revised form plates with cutouts (perforated plates) are given high importance at the structural design stage because
30 October 2014
they can reduce the structural strength. In this regard, local reinforced perforated plates are used in
Accepted 19 December 2014
Available online 27 December 2014
shipyards to satisfy buckling and ultimate strength requirements, but quantitative evaluations of the
reinforced perforated plates have not yet been carried out. To mitigate the decrease in the strength of
Keywords: perforated plates, the Carling, face-plating, and doubling stiffening methods have been adopted with the
Reinforced perforated plate goal of increasing both the buckling and ultimate strengths. In particular, the Carling stiffener has been
Critical buckling strength
partially adopted at holes typically found in shipyards (e.g., access holes, lightening holes), but no
Ultimate buckling strength
standard methodologies or recommendations are available for the use of this stiffener. In the present
Carling
Face-plating study, a series of numerical studies were undertaken to analyze the buckling and ultimate strengths for
Doubling various stiffening methods (Carling, face-plating, and doubling) and loading conditions (axial compres-
sion and in-plane edge shear loading). An optimal reinforcement method was determined by comparing
stiffened weights and ultimate strengths of the three methods. Finally, a design formula for calculating
the ultimate strength of a perforated plate was developed on the basis of over 144 cases of nite element
analysis.
& 2014 Elsevier Ltd. All rights reserved.

1. Introduction [1] investigated the complex behavior of structural plate panels with
openings used in ships, to calculate buckling and ultimate strengths
The basic components of ships and offshore structures are steel using parametric studies employing linear and non-linear nite
plates, which are cut, shaped, bent, and manufactured to meet the element analysis (FEA) to investigate the inuence that dimensions
requirements of a desired design conguration. Stiffened plates of the plate panels, shape and size of openings had on shear and
including curved and perforated plates are the most commonly compressive stresses in longitudinal and transverse directions. Simple
adopted structural members in the shipbuilding industry. Hull deck design formulae were developed on the basis of these FEA results.
hatches, bottom girders, and diaphragms factor a certain extent of Kumar et al. [2] determined the effects of a rectangular central
perforation within their geometries since perforation decreases the opening on the ultimate strength of a square plate under axial
weight as well as natural frequency of a plate in comparison with a compression. The effects of the plate slenderness ratio and the
non-perforated plate. However, this also leads to a signicant reduc- opening area ratio on the ultimate strength were determined using
tion in both buckling and ultimate strength. To resolve the issue, the a nonlinear FEA. Shanmugam et al. [3] proposed a design formula to
dimension of the perforation hole is limited to a specic range at the predict the ultimate load capacity of perforated plates subjected to
design stage to limit the increasingly altered structural response that is compressive loading with different boundary conditions within the
associated with an increasing hole size. Thus, several studies have ABAQUS nite element package, to carry out an elastoplastic analysis
been reported in the literature to examine the effect of perforation on on plates under uniaxial and biaxial compressions. They found that
both buckling and ultimate strength of perforated plates. Wang et al. various parameters including plate slenderness ratio and opening size
affected post-buckling behavior and the ultimate load capacity of
perforated plates. Paik [4,5] investigated the ultimate strength char-
n
Corresponding author. Tel.: 82 51 510 2342; fax: 82 51 512 8836. acteristics of steel plates with a single circular hole under longitud-
E-mail address: jaemlee@pusan.ac.kr (J.-M. Lee). inal and axial compressive loads through a series of elastoplastic,

http://dx.doi.org/10.1016/j.ijmecsci.2014.12.016
0020-7403/& 2014 Elsevier Ltd. All rights reserved.
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 195

large-deection FEAs carried out in ANSYS while varying hole size as and ultimate strengths of perforated plates in an effort to improve
well as the plate dimensions. El-Sawy and Nazmy [6] investigated the the current design practice, by analyzing a total of 90 unstiffened
effects of the aspect ratio, hole size, and hole location on the elastic and 9 stiffened plates and additionally investigating their collapse
buckling of uniaxially loaded rectangular perforated plates with characteristics under axial compression, using nonlinear FEA.
eccentric holes, as evidenced by a buckling coefcient, k. Komur and Their study resulted in a new formula by comparing numerical
Sonmez [7] also investigated the elastic buckling behavior of rectan- and experimental results for critical buckling strength of perfo-
gular perforated plates using FEA by modeling circular cutouts chosen rated plates. Alagusundaramoorthy et al. [12] studied the ultimate
at different locations along the principal axis of the plates subjected to strength of stiffened panels with cutouts under uniaxial compres-
linearly varying loads, in order to evaluate the effects of the cutout sion using an approximation method based on an approach of
location on the buckling behavior of the plates. They investigated the modeling struts that modeled the ultimate strength of simply
effects of the hole size and location on buckling load of rectangular supported stiffened panels with initial imperfections and cutouts.
plates with aspect ratios ranging from one to four. Narayanan and Their proposed method compared well with experimental results
Chow [8] investigated the ultimate capacity of uniaxially compressed for ideal stiffened panels with square cutouts in various sizes.
perforated plates with specic focus on examining the differences As many engineers have struggled to nd ways that increase
between perforated plates containing square holes and perforated the buckling and ultimate strengths through methods for local
plates containing circular holes, via experimental as well as theoretical reinforcement around holes, research on verifying these stiffening
methods. Dadrasi [9] investigated the effect of imperfections on the methods has not been conducted with sufcient scientic and
buckling load of perforated rectangular steel plates by performed methodological rigor. Further, in practice empirical designs are
numerical and experimental investigation of buckling behavior of commonly applied on the basis of experience. Furthermore, the
rectangular plates with circular and square cutouts under uniaxial, in- effect of local reinforcement methods has not yet been thoroughly
plane, compressive loading in the elastoplastic range, within various investigated.
loading bands. Therefore, this paper focuses on the elastic and critical buckling
As many studies have evidenced, the primary factor affecting strengths as well as ultimate buckling strength under three different
decrease in buckling strength of perforated plates is the perfora- methods for reinforcing perforated plates i.e. the carling method, the
tion ratio (dc/b, i.e., the ratio of diameter of a circle hole to breadth doubling method, and the face-plating method. Non-linear buckling
of the plate). Additionally, there are methods for increasing the analysis was conducted using a commercial FEA code, MSC NASTRAN,
buckling and ultimate strength including by altering the hole for a range of important factors including stiffener height, width, and
dimension, changing the hole locations, and increasing the plate's thickness. In addition, various loading conditions such as axial loading
thickness. Recently however, as weight and cost reduction of raw and in-plane edge shear loading condition as well as imperfections
materials has begun to receive great attention in the engineering caused by welding were considered. The results of this study have
community, the plates used in shipbuilding industries have potential application to the general problem of adopting perforated
become gradually thinner. In this regard, it is no longer a proper plates for structural design.
resolution to increase thickness and consuming lots of raw
materials in order to achieve greater strength. Therefore, research-
ers have studied the method of reinforcement (see Fig. 1) of 2. Finite element analysis
perforated plates and widely applied the inferences of such studies
in the shipbuilding industry to overcome reductions buckling and 2.1. Description of target modeling
ultimate strength resulting from perforation in an attempt to
reduce weight and cost. Cheng and Zhao [10] studied the buckling A circular, elliptical, or rectangular type of cutout may be used for
behaviors of uniaxially compressed perforated steel plates that ships and offshore structures, but a stadium-type opening is found
were strengthened by four stiffener types viz. ringed, at, long- most frequently in practice, so it is the only cutout considered in this
itudinal, and transverse stiffeners. Using the commercial code study. This assumption is based on the cutout shape was found to not
ANSYS, they performed a series of elastic and elastoplastic FEAs affect the buckling or ultimate strengths of the perforated plate.
for a range of plate slenderness ratios as well as hole-diameter Additionally, in this study, the cutout was located at the center of
ratios in an effort to investigate the most efcient cut-out- the plate so in order to present a conservative analysis. A schematic
strengthening methods for improving ultimate strengths. Kim for the modeled perforated plate geometry and specic dimension has
et al. [11] reported experimental results involving the buckling been provided in Fig. 2(a) and Table 1, respectively. In the present

Fig. 1. Perforated plate in the ship and offshore structures.


196 J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205

Fig. 2. (a) Schematic of the perforated plate and its reinforcement method around the cutout for different stiffeners: (b) carling stiffener (c), face-plating (d), and doubling
plate stiffener.

Table 1 Table 2
Dimensions of perforated plate. Dimensions of the stiffener around the cutout depending on the reinforcement
method.
a(mm) b(mm) t(mm) L(mm) dc(mm) a/b
Carling and Face-plating Doubling
3182 860 14 800 500 3.7
Height (mm) Thickness (mm) Width (mm) Thickness (mm)

50 5 50 19.5
10 26
investigation, as a collaborative project with a ship yard, the most
15 32.5
widely used width, thickness, and aspect ratio for perforated plate 20 39
have been considered. According to a previous study of 22 vessels 100 5 90 19.5
(container ship, LNG carrier, 10 26
pLPG
carrier, and tanker), the most widely
used slenderness ratio b=t Y =E and thickness ratio b=t are in the 15 32.5
20 39
range of 1.553.83 and 40.399.7, respectively [13]. These plates are 150 5 130 19.5
surely targets for elastic buckling and elastoplastic nonlinear analysis 10 26
considering local buckling in real designs. 15 32.5
There are three methods for reinforcing perforated plates viz. 20 39
200 5 180 19.5
carling stiffener method, doubling plate method, and face plating
10 26
method. The carling stiffener method, which is named such owing 15 32.5
to the use of a welded carling stiffener, is a form of stiffener type 20 39
around the cutout, as shown in Fig. 2(b). The doubling plate
method involves a form of attached duplicate plate around the
cutout, as shown in Fig. 2(c). The face plating stiffener method reinforcement methods. The weight tonnage of the stiffener is not
involves welding of a coaming which has a certain height around exactly the same but is similar, being in the range of 0.0080.11.
the cutout hole, as shown in Fig. 2(d). Generally, the carling
stiffener is widely adopted in the ship yard owing to its conve-
nience of use, whereas the doubling plate and face plating stiffener 2.2. Mesh convergence study
methods have seen more limited use owing to the difculties
involved in their production. A series of FEAs were carried out by In order to determine the optimum number of meshes, a mesh
changing the following variables affecting buckling and ultimate convergence study was performed for the at plate under long-
strength viz. local stiffening height hc , thickness of carling itudinal compression loading. Fig. 3 shows the comparative
stiffener t c , height of coaming hf , and thickness of face- investigation between the nonlinear FEA result considering the
plating t f , radial direction width wd , thickness of doubling number of elements and the exact buckling strength. As indicated
plate t d . in this gure, the optimum number of elements is determined
The size and dimension of both doubling and face-plating when the buckling strength begins to converge. A mesh number of
stiffeners were determined based on the carling stiffener weight. 20  60 was found to be reasonable to reduce the analytical time
This is done to appropriately compare the effects of the three for a at plate under an aspect ratio of 3.0. When the mesh
different reinforcements on the perforated plates under the number was 20  60, the buckling strength did not differ from
various modeled loading conditions. The stiffener weight is that in the analyses of plates having higher mesh numbers.
calculated based on the density, mass, and volume. Table 2 shows Therefore, the mesh of the perforated plate was set to 20  60
the dimensions of these variables for each of the three studied in the present research. At this time, the error between the FEA
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 197

where W e is the external work and W i is the internal work.


Therefore, the virtual work can be rewritten by Eq. (2):
 
K  W fRg fQ g 2

where K is stiffness matrix, fWg is nodal displacement incre-


ment, is load scale parameter, fRg is applied nodal force vector,
and fQ g is unbalance force to be covered. By applying Eq. (2), the
arc-length methods are represented by Eqs. (3)(7). At step i 1 ,
the incremental form is represented by Eq. (3) as follows:
n o    
KW i  1  W 0i Q i  1 0i RW i  1 3

where R is the reference arc-length radius, W is the increment of


the deection component, and 0i is the average stress at step i.
The arc length must satisfy the following relation:
 2
W2 0i r 2i 4

where r i is the arc length o is also the rst increment of


n at step i and
the arc length at step i W 0i ; 0i ; these two parameters are in
Fig. 3. Mesh convergence study of at plate.
the same direction. Therefore, they can be linearized as follows:
( )T
X l n o X l
W i  1
m
W 0i mi  1 i r i
0 2
5
m0 m0
n Because
o Eqs.
 (4)
 and (5) can be solved as a linear system,
W 0i and 0i can be calculated, as shown in Eq. (6):
i 2 3
KW i  1  RW i  1 ( ) ( )
6 X X 7 W 0i Qi1
i
6 l l 7 6
4 W mT i1 mi1
5 0 r 2i
i
i
m0 m0

During the repeat and convergence calculation of the n -steps,


the stiffness matrix equation can be represented as follows:
" # ( )
i X
n   X
n 1 n o
KW i  1 W m
i W ni  ni RW i  1 W m n1
i Qi
m0 m0

7
wi n o
where Q ni  1
denotes the nonparallel force, as shown in Eq. (7).
wi
The arc-length r i is required to satisfy the same length, and can be
i
wi represented by Eq. (8):
wi   !2
 n1 
X 2 X
n 1
 W m
W n  m
n
r 2i 8
0  i i i i
wi w m 0  m0

Fig. 4. Schematic of the arc-length incremental method [14]. Eq. (5) can be linearized, as shown in Eq. (9):
( )T
and the exact solution was 1% or less. Moreover, this mesh size is X
n 1   nX1
W i m
W ni m
i i 0
n
9
widely used by many designers, making it a reasonable selection. m0 m0

As the system of linearization can generate Eqs. (4) and (6),


W ni and ni can be calculated, as shown in Eq. (10):
2 3
2.3. Numerical method X
n 1 X
n 1
6 KW i  1 Wm
i RW i  1 Wm
i 7( ) ( )
6 7 W ni Q ni  1
6 m0 m0 7
In order to calculate the nonlinear buckling analysis, the arc 6 n1 7
6 X X
n 1 7 ni 0
length method was adopted as the numerical method (Fig. 4). 4 W mT m 5
i1 i1
Because of diverging solutions, the NewtonRaphson method was m0 m0

not suitable for predicting buckling behavior. On the other hand, 10


the arc length method tracks complex buckling behavior, such as
In this calculation, Eq. (10) is repeated until the nonparallel
secondary buckling or snapback phenomena of the plate with a
force Q i is equal to zero.
laminate thickness, to obtain the static solution of nonlinear
buckling.
The arc length method can be deducted by the principle of 2.4. Loading and boundary conditions
virtual work applied in an incremental form. The principle of
virtual work is expressed by Eq. (1): The nite element method (FEM) formulation of a four-node
shell element was used in our elastoplastic large deformation
analyses. As indicated in the previous section, MSC. PATRAN/
W e W i 1 NASTRAN (SOL600)a commercially available FEA codewas used
198 J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205

Table 3
Material properties of the perforated plate.

Material Elastic modulus Yield stress Shear yield Poisson (ton/


(MPa) (MPa) stress (MPa) ratio mm3)

AH 32 206,000 315 181.9 0.3 7.85E-


09

to determine the buckling and ultimate strengths of the modeled


reinforced perforated plates.
The material properties for the adopted material viz. high-
tensile steel which is typically used in ship yards are listed in
Table 3. In this table, shear yield stress is calculated using Eq. (11):
p
Y Y = 3 11

where, Y is the yield stress and Y is the shear yield stress. The
behavior of the material was assumed to be elastic-perfectly-
plastic i.e., linearly elastic before yielding and perfectly plastic
behavior after yielding with no strain hardening effects.
Perforated plates in ships and offshore structures are typically
supported by adjacent plates and thus are in a constrained state
that can be best modeled as a combination of a rotational
constraint and a xed constraint for out-of-plane direction. There-
fore, the studied perforated plates were considered to be simply
Fig. 5. Initial imperfection shapes (a) obtained by eigenvalue analysis and
supported on all edges. In terms of structural design, when (b) produced by welding.
evaluating the buckling and ultimate strength for a perforated
plate, adopting a simply supported condition gives more conser-
vative results compared to the clamped condition. Moreover, for where, m and n represent the number of half-waves in x - and y
the safe design of ships and offshore structures, this design -directions, respectively. The shape of initial deection in local
approach of using the simply supported condition provides a panel is assumed to be the same in the buckling model. Magni-
proper safety factor. This means that the lateral deection stiffness tudes of Aomn are determined from the condition that the max-
on the boundaries is innite and the rotational stiffness on the imum of initial deection as determined by the buckling model is
edges is innitesimally small. Previous research [15] has proven equal to the maximum initial imperfection determined by Eq. (13):
that this approximation consistently provides conservative results, w0 0:05 t
2
13
making it useful for practical engineering.
As listed in Table 2, three different loading conditions (long- where t denotes the thickness of the plate and represents the
itudinal compression, transverse compression, and in-plane edge slenderness ratio of plate. Fig. 5(a) shows the initial imperfection
shear loading) were considered to estimate buckling and ultimate shape produced by eigenvalue analysis under a longitudinal
strengths for the adopted reinforcement methods. The three differ- compressive load and Fig. 5(b) illustrates an example of an initial
ent loads and various plate heights, widths, and thicknesses were imperfection shape caused by welding, as measured on various
adopted as the main parameters in over 144 cases of the linear/ types of plates. In the present study, the perforated plate was
nonlinear buckling analysis carried out on a perforated plate. assumed to go through the buckling mode from initial imperfec-
tions that occurred by welding.

2.5. Initial imperfection


3. Results and discussion
To fabricate the perforated stiffened plates in ship and offshore
structures, llet welding is widely adopted to combine the plate 3.1. Effect of plate parameters
and stiffener. Thus, initial imperfections such as initial deection
as well as residual stress caused by welding are naturally created One of the recent issues in shipbuilding industries is to reduce
both in the plating and stiffeners. Further, in advanced ship stru- the thickness of the steel plate to reduce the weight of the ship
ctural design, load-carrying capacity calculations of ship plating and to save raw materials. This can be realized by developing
should be performed considering the inuences of post-weld structural analysis technologies as well as structural optimization
initial imperfections. However, owing to the fact that the char- methodologies in the ship designing process. It is then essential to
acteristics of initial imperfection have not been thoroughly inves- perform buckling analysis to verify the integrity of the target
tigated, it is difcult to dene the actual shape and magnitude of structure. In addition, the reduction of the structural strength
such initial welding imperfections. The general form for an initial caused by using a relatively thinned plate can be compensated by
imperfection wave, wo, is represented by Eq. (12) [16,17]. This attaching stiffened panels to satisfy the safety requirements of the
formulation was derived by idealizing the plate-to-beam connec- classication society. In the case of a perforated plate, however,
tion and assuming a boundary condition that provides simple the structural strength can only be satised by increasing the plate
supports. thickness to satisfy the requirements of the classication society in
X
1 X the absence of guidelines and design formulations. Therefore, the
1
m x n y
wo Aomn sin sin 12 present study performed a series of numerical analyses to inves-
a b
m1n1 tigate the buckling and ultimate strength for various stiffening
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 199

Fig. 6. Buckling strength of perforated plate under longitudinal compression loading for aspect ratios of (a) 2.2 and (b) 3.7.

methods (carling, face-plating, and doubling). In addition, design JohnsonOstenfeld formula is given by Eq. (14):
formulas for the ultimate strength of a perforated plate were 0 1
E 
E r 0:5 F
developed for use in ships and offshore industries. Because the cr @ 1  F A 14
present study mainly aims to investigate the buckling and ultimate F 4 E E 4 0:5 F
strength depending on the stiffening method, the effect of plate
parameters on perforated plate was not considered. However, it is where cr is the critical buckling strength, E is the elastic buckling
necessary to analyze the plate behavior before investigating the strength, and F is the reference yield stress. The yield stress
perforated plate. Therefore, the effect of the plate parameters on depends on the loading type, which ispestimated
as F Y for
the perforated plate is discussed in brief. axial compression loading and F Y = 3 for edge shear loading.
Fig. 6 shows the result of the ultimate buckling strength analysis The horizontal axis, , in Fig. 7, which is plotted against the
under longitudinal compression loading to investigate the effect of stress ratio on the vertical axis, is a function of a dimensionless
plate parameters such as the aspect ratio, slenderness ratio, and cutout value that includes the carling height, face-plating height, or
diameter and plate breadth (dc/b). The investigated plate slenderness doubling width, as given by Eq. (15):
r
ratio and cutout diameter and plate breadth ratio are in the range of b Y
0.96083.3629 and 0.17440.8139, respectively. In addition, the cur- 15
f hc ; hp ; wd E
rent plate slenderness ratio of 3.7 is compared with a ratio of 2.2. As
shown in this gure, the ultimate buckling strength decreases with The critical buckling strength were found to exhibit a large
increasing plate slenderness ratio as well as cutout diameter and plate variation between the cases of transverse compression and in-
breadth ratio. However, the inuence of aspect ratio on the ultimate plane edge shear loading and was highly dependent upon the
buckling strength of the perforated plate is relatively small compared reinforcement method. However, this was not the case for long-
to that of the other plate parameters. itudinal compression loading wherein the buckling strength was
independent of the reinforcement method. Generally, if the thick-
3.2. Buckling strength with respect to reinforcement method ness of a perforated plate increases, the buckling and ultimate
strengths also increase. Such a typical trend was not seen in our
Fig. 7 depicts the results of the critical buckling strength of the elastic buckling analysis. The estimation of the elastic buckling
reinforced perforated plate for a 50 mm caring stiffener height, strength of a reinforced perforated plate is limited to the simula-
face-plate height, and doubling width, for a variety of analytical tion of the buckling phenomenon owing to difculties in modeling
scenarios. For all cases, each point represents the dimensionless the same. However, elastoplastic analysis, which inherently con-
critical buckling strength for the lowest possible value (50 mm) of siders geometrical and material nonlinearities, can effectively
the carling stiffener height, face-plate height or doubling width. capture the fundamental buckling and yielding mechanisms
Since the difference in critical buckling strength is not large when around the cutout of the perforated plate. The ultimate buckling
plotting for the overall scenario, the smallest thickness is selected strength determined by non-linear structural analysis techniques,
as a representative case. which in-turn considers material non-linearity, was evaluated and
In result of calculating the critical buckling strength of a the optimal reinforcement method is presented in the next
perforated plate using the JohnsonOstenfeld plastic correction section.
formula are somewhat erroneous. For example, the load-bearing
capacity is overestimated when compared to the actual strength 3.3. Ultimate strength with respect to stiffening method
for the large cutout case. However, because the aim of this
investigation is a simple comparison of the magnitude of The ultimate buckling strength of the carling stiffener method,
the buckling strength, the application of the JohnsonOstenfeld doubling method, and face-plating method, under loading condi-
plasticity correction formula is deemed reasonable. The tions of longitudinal compression, transverse compression and
200 J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205

Fig. 8. von-Mises stress distributions and deformations under longitudinal com-


pression for: (a) carling stiffener method, (b) face-plating method, and (c) doubling
plate method.

Fig. 7. Critical buckling strength of reinforced perforated plates under Fig. 9. von-Mises stress distributions and deformations under transverse compres-
(a) longitudinal compression, (b) transverse compression, and (c) edge shear sion for: (a) carling stiffener method, (b) face-plating method, and (c) doubling
loading. plate method.

edge shear loading, was investigated. Representative cases of the face-plating was modeled as 100 mm and its thickness was
stress distributions and deformation shapes at the ultimate considered to be 5 mm. In the doubling method, the width of
strength limit are illustrated in Figs. 810. The carling method the doubling was considered as 90 mm and its thickness was
was modeled with a height of the carling equal to 100 mm and a 19.5 mm. As shown in Fig. 8, the buckling half wave number of the
thickness of 5 mm. In the face-plating method, the height of the perforated plate under longitudinal compression load was 3 or
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 201

Fig. 10. von-Mises stress distributions and deformations under edge shear loading
for: (a) carling stiffener method, (b) face-plating method, and (c) doubling plate
method.

5 depending on the employed method of reinforcement. Also, the


collapse of the carling stiffener and face plate were conrmed
graphically. Fig. 9 illustrates the von-Mises stress distributions and
deformation characteristics under transverse compression. As
shown in this gure, the buckling half-wave numbers for all three
reinforcement conditions were found to be equal. In case of the
carling stiffener, the phase of the mode changes based on the
carling ends owing to reinforcement effects. Additionally, failure
phenomena for in-plane edge shear loading was found to occur
because of yielding effects around the cutout, as shown in Fig. 10.
The most effective stiffening method was found to be the face-
plating method since here the stiffener prevents the plate from
distorting at the end of the cutout.
After analyzing the collapse patterns, the relationship between
the weight of the stiffener and the ultimate strength was found to
be an important factor to select the optimal reinforcement method
(see Fig. 11). When longitudinal compression loads were modeled
on the reinforced perforated plate, the doubling plate reinforcement
method was found to overestimate the ultimate strength owing to
the considerable increase in stiffness around the cutout. In the
carling reinforcement method, plate weight around the cutout was
found to share a progressively linear relationship with buckling
strengths as well as ultimate strength. This study also evidences
that for a reinforced perforated plate subject to transverse compres-
sion loading, the most effective stiffening method is the carling
strengthening process. A carling stiffener located in the transverse
direction of the perforated plate was found to make a signicant
difference to strength of the plate. The benets of the face-plate
strengthening process were found to be insufcient under the
transverse compression conditions and the doubling plate method
resulted in the highest ultimate strength of over 0.075 ton. How-
ever, it is unreasonable to compare this value with the actual Fig. 11. Variation of ultimate strength with respect to weight for three different
efciency of reinforcement because the concept of an inserted plate reinforcement methods around the opening in a plate viz. (a) longitudinal axial
is modeled around the cutout, which is unrealistic. Under in-plane compression, (b) transverse axial compression, and (c) edge shear loading.
202 J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205

formula. The ultimate strength obtained using the carling stiffener


method under longitudinal compression is given as follows:


xu tc
C1 0:43628 16
Y hc
where,
C1 0:0145hc 0:6205
The ultimate strength obtained using the carling stiffener
method under transverse compression is given by:


yu tc
C1 C2 17
Y hc
where,
)
C 1  0:0019hc 0:5049
for 50 r hc r125
C 2 0:0025hc 0:0749
)
C 1 0:0068hc 0:6057
for 125 o hc r 200
C 2 0:3783
Fig. 12. Stressstrain relationship of stiffened perforated plates at the same weight
(0.2 ton) under different longitudinal compression conditions.
where, yu is the ultimate strength under transverse compression.
The ultimate strength obtained using the carling stiffener method
edge shear loading conditions, the results of the variation in under edge shear is:
ultimate strength with respect to the reinforcement method are

u tc
illustrated in Fig. 11 (c). Both the face-plate and the doubling plate C1 C2 18
Y hc
methods showed large variation in the shear strength with small
increases in the reinforcement cross-sectional area. The carling where,
height or thickness was shown to not signicantly affect the )
C 1 0:0055hc 0:111
strength. The face-plating method was found to be better than the
C 2 0:0002hc 0:396
doubling method within a certain weight range (0.030.07 ton);
however, in other cases, the ultimate buckling strength of the In the face-plating method, height of the coaming and its
doubling plate method was remarkably superior to that of the thickness are employed to develop a design formula. The ultimate
face-plating method. strength obtained using the face-plating method under longitu-
Fig. 12 demonstrates the longitudinal compressive stress versus dinal compression is given by:
strain under identical weight conditions (i.e. 0.2 ton) as modeled

xu tp
in the illustrations of Fig. 11. Both the doubling plate method and C1 C2 19
the face-plate method led to large increments of buckling and
Y hp
ultimate strengths per unit increase in plate thickness. However, where,
the carling stiffener exhibited a different tendency owing to the )
C 1 0:0306hp  1:1248
fact that it did not signicantly contribute to the buckling strength. at t p r 10
In other words, this means that the effect of reinforcement on C 2  3  10  5 hp 0:3998
plate strength did not increase with increasing moment of inertia. )
To obtain the same effect, the stiffener weight should be secured C 1  0:0004hp 0:6028
for 50 r hp r 100 at t p 4 10
above approximately 0.1 ton. C 2 0:0034hp 0:1974
As noted in Figs. 1112, the stiffening effect for each loading
)
condition indicates the existence of a different optimal reinforcement C 1  0:0061hp 1:1513
method of the perforated plate under each loading condition. In for 100 o hp r 200 at t p 4 10
C 2 0:0011hp 1:1513
summary, the doubling plate method is the optimal method under
longitudinal compression, the carling stiffener method is the optimal The ultimate strength obtained using the face-plating method
method under transverse compression, and the face plate is the under transverse compression is:
optimal method under edge shear conditions.

yu tp
C1 C2 20
Y hp
3.4. Development of design formula
where,
In the present study, we develop the design formula for )
C 1 0:0037hp  0:1235
calculating the ultimate strength of a perforated plate under three C 2  0:00006 hp 0:2376
different loading conditions on the basis of the nonlinear analysis
results from FEA. The design formula is developed by changing The ultimate strength obtained using the face-plating method
variables such as height, width and thickness variable while under edge shear loading is:
employing the carling, face-plating, and doubling stiffening meth-

yu tc
ods, respectively. Although the dimension and thickness of the C1 C2 21
Y hc
perforated plate are limited to remain within the specic range,
the presented formulations can be easily applied to ship and where,
offshore structures since the analyzed dimensions are widely )
C 1 0:0457hp  1:5904
adopted in ship yard. In carling stiffener reinforced method, local for 50 r hp r 100 at t p r 10
C 2  0:0007hp 0:418
stiffener height and thickness is changed to develop the design
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 203

)
C 1 0:0457hp 1:5904
for 100 ohp r200 at t p r10
C 2 0:00006 hp 0:3572

)
C 1 0:011hp 0:1152
for 50 r hp r 100 at t p 4 10
C 2 0:0028hp 0:2617

)
C 1 0:0016hp 1:0202
for 100 o hp r 200 at t p 4 10
C 2 0:001 hp 0:4462

In the doubling plate method, the width in the radial direction


and its thickness is employed to develop a design formula. The
ultimate strength of the doubling plate method under longitudinal
compression is:


xu t
C1 d C2 22
Y wd
where,
)
C 1 0:0073wd 0:7478
for 50 r wd r 90 at t d r 26
C 2 0:0046wd 0:2176
)
C 1 0:0008wd 0:0303
for 90 o wd r 180 at t d r26
C 2 0:0002wd 0:6194
)
C 1 0:0006wd  0:0074
at t d 4 26
C 2 0:0001wd 0:6356

The ultimate strength obtained using the doubling plate


method under transverse compression is:


yu t
C1 d C2 23
Y wd
where,
)
C 1 0:0046wd  0:0740
for 50 r wd r 90
C 2 0:0004wd 0:1736
)
C 1 0:0158wd  1:0755
for 90 o wd r 180
C 2 0:0007wd 0:2511

The ultimate strength obtained using the doubling plate


method under edge shear is:


u t
C1 d C2 24
Y wd
where,
)
C 1 0:0133wd  0:2134
for 50 r wd r 90
C 2 0:0002wd 0:3171
)
C 1 0:0340wd  2:1739
for 90 o wd r 180
C 2 0:0017wd 0:4383

Fig. 13 shows the correlations and standard deviations from the


empirical formula of the perforated plate subjected to longitudinal
compression (LC), transverse compression (TC), and edge-shear
loading (ES), when compared against the FEA results. The normal-
ized ultimate strength as determined by the FEA results are
plotted against the normalized ultimate strength from the empiri-
cal formula for the perforated plate. The results from both
methods were in good agreement.

3.5. Verication

In the present study, empirical formulas for stiffened perfo- Fig. 13. Correlations between results from FEM and the empirical formula of
rated plates have been developed depending on the stiffening ultimate strength, using the (a) carling stiffener, (b) face-plating, and (c) doubling
method such as carling, face-plating, and doubling to investigate plate method under single and combined loads.
204 J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205

Fig. 14. Experimental specimens of stiffened perforated plates, and (b) comparative study of the ultimate strength between a previous study and the current investigation.

the buckling characteristics as well as the ultimate strength under analysis that considers two nonlinear factors. Analysis results
various loading conditions. The developed empirical formulas for showed some differences depending on three different stiffen-
determining ultimate strength were compared with the nite ing methods.
element analysis results to verify the effectiveness of the present The doubling plate method around the cutout is most effective
investigation. However, from an application viewpoint, it is neces- stiffening method under the longitudinal compression loading.
sary to perform a buckling experiment targeting a stiffened In addition, the carling stiffener method is effective for trans-
perforated plate. As a literature investigation, many buckling- verse compression loading owing to a large section modulus in
related studies have been reported for steel plates and stiffened the transverse direction.
panels without an opening. However, relatively few studies have In-plane edge shear effectively restrains buckling and yielding
focused on perforated steel plates. Furthermore, it is not easy to occurrences around the cutout. For this, the face-plating
nd experimental investigations of the buckling behavior of method is more effective than the other methods with the
stiffened perforated plates depending on the stiffening methods same weight for increasing the strength of the perforated plate.
such as carling, doubling, and face-plating to adopt a steel The developed design formulas were developed from a trend
structural member. Kim et al. reported the buckling and ultimate line using regression analysis to be used in the ship yard.
strength of perforated plate panels subjected to axial compression Depending on each reinforcement method, design equations
by conducting both experimental and numerical investigations that can be applied to the three types of loading were
with design formulations [13]. A series of buckling collapse tests developed.
for steel plates and stiffened panels subjected to axial compressive Limited verication studies have been conducted on the carling
loads has been conducted. A total of 90 perforated plates and stiffener method. The predicted ultimate strength of a stiffened
9 stiffened panels with an opening are tested up to beyond perforated plate was close to the experimental investigation
ultimate strength. Therefore, we verify the developed empirical result compared to the previously developed formula.
formula through an experimental investigation. Fig. 14(a) shows
the experimental stiffened perforated plate specimens and
Fig. 14(b), a comparative study of the ultimate strength between
a previous study and the current investigation. In Fig. 14(b), the Acknowledgments
blue bar graphs show the comparative investigation in the
previous research outcomes. FEA and the formula results are not
This research was supported by Basic Science Research Program
identical but are close to the experimental results. The reddish bar
through the National Research Foundation of Korea(NRF) funded
graph shows the result obtained using the present empirical
by the Ministry of Education(No. 2013R1A1A2A10011206). This
formula for the carling stiffness method. This graph suggests that
work was nancially supported by a National Research Foundation
the predicted ultimate strength of a stiffened perforated plate was
of Korea (NRF) grant funded by the Korean government (MEST)
close to the experimental investigation result compared to the
through GCRC-SOP (Grant no. 2011-0030667).
previously developed formula.

References
4. Concluding remarks
[1] Wang G, Sun H, Peng H, Uemori R. Buckling and ultimate strength of plates
The objective of this study was to investigate the optimal with opening. Ships Offshore Struct 2009;4:4353.
[2] Kumar MS, Alagusundaramoorthy P, Sundaravadivelu R. Ultimate strength of
reinforcement method that maximizes the buckling and ultimate square plate with rectangular opening under axial compression. J Nav Archit
strengths of a perforated plate used frequently in ships and Mar Eng 2007;4:1526.
offshore structures. Three types of reinforcement methods were [3] Shanmugam NE, Thevendran V, Tan YH. Design formula for axially compressed
perforated plates. Thin-Walled Struct 1999;34:120.
classied, and the optimal reinforcement method for each loading
[4] Paik JK. Ultimate strength of perforated steel plates under edge shear loading.
condition was found in this study. The principal conclusions are as Thin-Walled Struct 2007;45:3016.
follows. [5] Paik JK. Ultimate strength of steel plates with a single circular hole under axial
compressive loading along short edges. Ships Offshore Struct 2007;2:35560.
[6] El-sawy KM, Nazmy AS. Effect of aspect ratio on the elastic buckling of
It is strongly recommended that the buckling and ultimate uniaxially loaded plates with eccentric holes. Thin-Walled Struct
strengths of the perforated plate be estimated via elastoplastic 2001;39:98398.
J.-H. Kim et al. / International Journal of Mechanical Sciences 92 (2015) 194205 205

[7] Komur MA, Sonmez M. Elastic buckling of rectangular plates under linearly [13] Kim UN, Choe IH, Kwon JC, Paik JK. A study on the buckling strength of plate
varying in-plane normal load with a circular cutout. Mech Res Commun panels with opening. J Soc Nav Archit Korea 2010;47:21024 (in Korean).
2008;35:36171. [14] Kim JH, Park JS, Lee KH, Kim JH, Kim MH, Lee JM. Computational analysis and
[8] Narayanan R, Chow FY. Ultimate capacity of uniaxially compressed perforated design formula development for the design of curved plates for ships and
plates. Thin-Walled Struct 1984;2:24164. offshore structures. Struct Eng Mech 2014;49:70526.
[9] Dadrasi A. On the effect of imperfection on buckling load of perforated [15] Paik JK, Ham JH, Kim EN. A new plate buckling design formula. J Soc Nav Archit
rectangular steel plates. Res J Rec Sci 2013;2:3643. Jpn 1992;6:26774.
[10] Cheng B, Zhao J. Strengthening of perforated plates under uniaxial compres- [16] Smith CS, Davidson PC, Chapman JC, Dowling PJ. Strength and stiffness of
sion: Buckling analysis. Thin-Walled Struct 2010;48:90514. ships plating under in-plane compression and tension. R Inst Nav Archit Trans
[11] Kim UN, Choe IH, Paik JK. Buckling and ultimate strength of perforated plate 1988;130:22796.
panels subject to compression: experimental and numerical investigations [17] Ueda Y, Yao T. The inuence of complex initial deection modes on the
with design formulation. Ships Offshore Struct 2009;4:33761.
behavior and ultimate strength of rectangular plates in compression. J Constr
[12] Alagusundaramoorthy P, Sundaravadivelu R, Ganapathy C. Ultimate strength
Steel Res 1985;5:265302.
of stiffened panels with cutouts under uniaxial compression. Mar Struct
1995;8:279308.

You might also like