REF5
REF5
REF5
Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng
Research paper
A R T I C L E I N F O A B S T R A C T
Keywords: In actual marine structures, most stiffened plates are subjected to the combination of in-plane and out-of-plane
Stiffened plates loads. The influence of lateral loads on stiffened plates’ ultimate strengths and collapse modes under combined
Experiments load cases needs to be further investigated. For this purpose, the authors designed a combined load testing device
Combined loads
to apply variable combined loads to stiffened plates, and conducted experiments of three different load cases. In
Collapse modes
addition, Nonlinear Finite Element Analysis (NFEA) have been performed based on measured initial deflections,
Ultimate strength
Variable lateral loads which shows the ultimate axial loads and collapse modes match well with experimental results. Numerical
models with ISSC recommended initial deflections are also established to investigate the interaction between
axial and lateral loads. Results shows that lateral loads cause earlier yielding of the stiffeners. As a result, the out-
of-plane constraint to the stiffened plate is weakened, and the global displacement grows significantly as axial
load increases. The results of this research can be helpful for understanding lateral loads’ influence on the ul
timate strengths and collapse modes of stiffened plates under combined load cases, and can provide references
for structural designs.
derived from these researches, and the results matched well with cor
1. Introduction responding calculations and experiments. Fujikubo et al. (2005a,
2005b) analyzed the ultimate strength of continuous plates and stiffened
In actual ocean environments, ship structures bear bending moments plates subjected to combined lateral and transversal loads respectively.
and water pressure simultaneously. As the basic units of ship structures, They found that membrane stresses would restrain the occurrence of
stiffened plates are usually subjected to the combination of in-plane and buckling. Additionally, Paik and Seo (2009a, 2009b) employed NFEAs
out-of-plane loads. Such combined loads result from sectional bending to study the ultimate strength of stiffened plates under biaxial and
moments, water pressure and cargo weights etc. Moreover, in actual lateral loads and analyzed factors including loading types and di
combined load cases, the lateral loads vary within a certain range. In the mensions which might have significant influences on the ultimate
past decades, numerous studies focusing on buckling analysis and load-carrying capacity of stiffened plates. Classic formulae derived by
empirical formulae have been conducted to investigate the ultimate these investigations can briefly predict the ultimate strength of the
strengths of stiffened plates under combined loads. However, further stiffened plates under typical load cases, and multiple load components
researches investigating the impact of lateral loads on the collapse have been taken into account.
modes of stiffened plates are still needed. Additionally, the collapse In empirical formulae derived currently, more advanced influencing
mechanisms of stiffened plates under variable combined loads are of factors such as types of stiffeners, shearing loads, initial deflections and
great significance, but this has been seldomly reported yet. More influence of openings have been considered. Komoriyama and Yanagi
attention should also be paid to the interaction between axial hara (2018) derived a simplified formula quantifying the influence of
compression and lateral pressure loads. transversal compression. Kim et al. (2020) proposed an advanced
In the past decades, the ultimate strengths of stiffened plates are empirical formulation on axial ultimate strengths for T-bar stiffened
researched from various perspectives, mostly focusing on the prediction plates based on NFEAs. Wang et al. (2023) proposed an empirical
methods for ultimate strength of stiffened plates under the combination formulation to assess the ultimate strengths of side shells subjected to
of in-plane and put-of-plane loads. Numerous empirical formulae are combination of axial compression and shearing loads. The deteriorating
* Corresponding author. State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China.
E-mail address: jinjucui@sjtu.edu.cn (J. Cui).
https://doi.org/10.1016/j.oceaneng.2024.117189
Received 7 September 2023; Received in revised form 15 February 2024; Accepted 18 February 2024
Available online 7 March 2024
0029-8018/© 2024 Elsevier Ltd. All rights reserved.
G. Guo et al. Ocean Engineering 299 (2024) 117189
Table 1
Dimensions of experimental model.
a(mm) b(mm) tp (mm) hw (mm) tw (mm) bf (mm) tf (mm) β λ σY (MPa)
2
G. Guo et al. Ocean Engineering 299 (2024) 117189
cases with single loading component, such as axial or lateral load. Paik strengths and collapse modes have been researched in these experi
et al. (1997) designed a bulkhead air loading device and conducted ments. Wang et al. (2018) researched the influence of initial imperfec
lateral ultimate strength experiments on nine corrugated bulkhead tion on the buckling behaviors of the thin-walled cylinders
models, and the load case was chosen as lateral loading only. This work experimentally. Cui and Wang (2020) operated a uniaxial compression
verified the feasibility of compressed air loading. Gordo and Guedes experiment, and investigated the collapse mode of stiffened plates with
Soares (2008, 2012) designed two series of stiffened plates with different openings and corrosions. Xu and Guedes Soares (2021) designed seven
widths, and investigated the impact of materials, geometrical shapes and stiffened plates to access the influence of the unsupported lateral edges.
section properties of stiffeners on the ultimate strength of stiffened Guo et al. (2021) analyzed the ultimate strength of laser-welded Al–Li
plates. Xu and Guedes Soares (2011) analyzed the influence of various alloy stiffened plates with experimental and numerical methods, and
kinds of model designs in the research of experiment set-up. These derived a thermal–elastic–plastic model for evaluating welding defor
classic single load experiments provide useful reference for designs of mation and residual stresses. Woloszyk et al. (2023) analyzed the impact
experimental models and setups. Many extended single load experi of marine immersed corrosion degradation on the compressive strength
ments are designed based on these experiments. The influence of of the stiffened plates experimentally, where the lower degradation
corrosion, boundary conditions and manufacturing methods on ultimate levels were considered.
3
G. Guo et al. Ocean Engineering 299 (2024) 117189
2. Experimental setup
Stiffened plates are commonly used in the decks, sides and double
bottoms of ship structures, such as oil tankers and container ships. Ac
cording to the ISSC’s recommendations (2012), both 1/2 + 1+1/2 spans
and 1 span designs of stiffened plates are acceptable for experiments or
simulations. The purpose of this experiment is to study the collapse
modes of the stiffened plate subjected to the combination of axial
compression and lateral pressure loads. So, in this research, the ‘one
span and one bay’ model is adopted, in order to facilitate the combined
Fig. 3. Structural plan of Model 3. loading experiment under the current laboratory condition. The di
mensions of the models are also determined with corresponding scales.
To get closer to the actual load cases, researchers designed combined Compressed air is used to apply the lateral loads, and the lateral loads
loading experiments based on single load experiments. Lateral pressure are kept constant during the axial loading process by using the air pump
loads and in-plane shearing loads were investigated in these (Fig. 6). Two stiffened plates are symmetrically arranged in the vertical
direction, and thick plates are arranged in longitudinal direction to
4
G. Guo et al. Ocean Engineering 299 (2024) 117189
apply lateral load and transversal constraints to the stiffened plates. stiffened plates also vary within a specific range. The pressure loads of
Thick plates are also arranged on the longitudinal endings of the stiff water are calculated by the equation P = ρgh, where ρ is the sea water
ened plates to ensure the uniform distribution of the longitudinal density, 1025kg/m3 i.e., the lateral loads are corresponding to the drafts
compression load in the section. These structures constitute an enclosed of 0m, 10m and 15m. On the other hand, the axial load is applied in the
chamber for applying lateral loads using compressed air, and longitu form of axial compression by the hydraulic cylinder. The summary of
dinal compression loads can be applied on the longitudinal endings. load cases is shown in Table 2.
The similarity between the actual structure and the test model is The global and local deformations of stiffened plates are measured by
considered in the model design. In detail, the slenderness ratios of the displacement gauges and strain gauges, respectively. The experimental
stiffened plates are calculated by Eq. (1), explanations of related vari models’ dimensions and the positions of the strain/displacement gauges
ables can be found in the nomenclature. Zhang and Khan (2009) are illustrated in Figs.1~3. Each model is equipped with 10 strain
concluded in actual ship structures, the slenderness ratios of stiffened gauges (S1–S10) and 4 displacement gauges (D1-D4). The strain gauge
plates ranged from 0.25 to 1, and the slenderness ratios of plates ranged rosettes can measure strains in three dimensions, and the angle between
from 1.0 to 4.5. According to Eq. (1), the slenderness ratio of the plate is each local gauge is 45∘ . Additionally, the local gauges point to 0∘ and 90∘
0.69, and the slenderness ratio of the stiffened plate is 3.51, i.e., the in the local coordinate system measure normal strains. In each figure,
experimental design is within the ranges of actual ship structural design red legends represent local gauges point to 0∘ ; blue legends represent
from the viewpoint of slenderness ratios. local gauges point to 45∘ ; green legends represent local gauges point to
√̅̅̅̅̅ 90∘ . Strain gauges “S1”- “S6” are set on the plate, and strain gauges “S7” -
b σY
β= “S10” are set on the stiffener. Additionally, four displacement gauges are
tp E
set on the plate and stiffener to measure the vertical global displace
√̅̅̅̅̅
a σY ments of plate and stiffeners.
λ= (1)
πr E
√̅̅̅
I 2.2. Layout and testing
r=
A
According to the experimental design, the layout is shown in Fig. 4.
In this research, three combined loading experiments on stiffened
The uniform loading of axial compression is guaranteed by the pushing
plates are performed to study the load-bearing capacity and the collapse
blocks. Additionally, slide rails fixed on the transversal sides of the outer
mechanism. The load cases are designed as: axial load only, 0.1 MPa
frames (Fig. 4(a) and (b)) can restrain the transversal and vertical
lateral load combined with axial load, and 0.15 MPa lateral load com
movements of the block. The pushing block 6 shown in Fig. 5(a) and (c)
bined with axial load, corresponding to Model 1, Model 2 and Model 3
ensures the transversal position fixing of the model. The loading side of
respectively. According to the actual measurements of the model, the
the experimental model and blocks are connected by strong bolts, while
scantlings of the plates are 600mm × 450mm × 1.7 mm. The scantlings
the non-loading end is longitudinally restricted by the outer frame. As
of stiffener section is 30mm × 20mm × 1.7 mm, i.e. the web height of
shown in Fig. 5(b), the testing device is vertically supported by the
stiffener (hw ) is 30 mm, and the web thickness of stiffener (tw ) is 1.7 mm,
blocks. As the axial load increases, the model is compacted by the block
and the flange plate width of stiffener (bf ) is 20 mm, and the flange
and the outer frame (Fig. 5).
thickness (tf ) of stiffener is 1.7 mm. The dimension of longitudinal girder
In this experiment, lateral loads are applied and maintained by the
is 60 mm × 600 mm × 3.66 mm. The summary of the experimental
air pump and controlling valve. The air pump is capable of storing
model’s geometrical property is shown in Table 1.
compressed air to maintain a constant lateral load for the experiment,
Various load cases occur during the life cycle of ships, including the
while the controlling valve is used for adjusting the values of lateral
drafts changes of ships. Consequently, the lateral loads applied on
loads in various load cases.
5
G. Guo et al. Ocean Engineering 299 (2024) 117189
(Fig. 7(b)). In each experiment, the lateral load is applied firstly, and the
axial compression load is applied by the hydraulic cylinder after the
pressure of the compressed air stabilizes at the design value.
The details of the strain gauge rosettes are shown in Fig. 8, where the
local gauge ε0 points to the X direction of the local coordinate system,
and the local gauge ε90 points to the Y direction of the local coordinate
system. The measured strains can be converted to principal strains, and
the relationship between the values measured by the strain gauge and
the plane strain components is shown in Eq. (2). Positive strain signifies
that tension, while negative strain stands for compression. Based on this,
the deformation behaviors of the structure in the loading process such as
buckling, torsion, and tripping can be judged. In this experiment, the
local gauges in the X direction all point to the fixed ending, the
arrangement of the strain gauges and the detailed orientations of local
gauges can be found in Figs. 1–3.
εx = ε0
εy = ε90 (2)
γxy = 2ε45 − (ε0 + ε90 )
As shown in Figs. 5(b) and Fig. 7, the structure of the pushing block In the past decades, the nonlinear finite element method has been
permits the penetration of pipe without affecting the uniform distribu developed into a general method for ultimate strength analysis. The
tion of the axial compression load. A valve is welded on the central point computer code ABAQUS 6.14 is used for finite element analysis in this
of the loading side to help apply lateral loads to the stiffened plates
6
G. Guo et al. Ocean Engineering 299 (2024) 117189
7
G. Guo et al. Ocean Engineering 299 (2024) 117189
8
G. Guo et al. Ocean Engineering 299 (2024) 117189
block’s loading ending, serving as the controlling point of the rigid body. actual experiments, the transversal and vertical displacements and the
Another reference point RP − 2 is set at the geometric center of each rotation around the longitudinal and vertical axis of the loading end are
numerical model’s non-loading ending, and the nodes in the non- restrained, while the rotations, vertical displacement and axial
loading end are coupled with RP − 2. Based on the constraints in displacement of the non-loading end are restrained. The calculation of
9
G. Guo et al. Ocean Engineering 299 (2024) 117189
10
G. Guo et al. Ocean Engineering 299 (2024) 117189
explicit and implicit methods are employed in NFEAs, and the results of of the stiffeners in each model are different, indicating the occurrence of
the two methods match well with each other. The displacements in the local deformations or global buckling under different loading cases.
experimental and numerical results denote the displacement of the After the lateral load is applied, strains are captured near the endings of
cylinder and the end-shortening of the model, respectively. However, the stiffeners. These strains increase rapidly as the compression load
limited by the actual conditions of authors’ laboratory, the non-loading grows. In combined load cases, early yielding occurs around the endings
end of the model cannot be ideally fixed. Therefore, the displacements of the stiffeners, which changes the collapse mode of the whole stiffened
measured in the experiments include the rigid body displacement of the plate.
entire structure results from the movement of the connecting structure, Detailed differences regarding to above three experimental models
and similar phenomena have been reported in previous experiments are discussed more extensively in the following sections.
(Cui and Wang, 2020). As a result, the displacement of the cylinder is not
equal to the end-shortening of the experimental model. Whereas, the 4.1.1. Results of model 1
ultimate strength and collapse modes, as the main concerns of this study, The displacements by gauges at measured points in Model 1 are
can be guaranteed by the quasi-static loading process in experiments and shown in Fig. 16. Measured points 1 and 2 are located on the flange of
numerical simulations, e.g., the asserted loads are recorded after the test the stiffener, while measured points 3 and 4 are located on the plate. The
model structures stabilize in each loading step. specific positions of the measured points can be found in Fig. 1. Vertical
By analyzing the results of displacement gauges and strain gauges, it displacements are observed as the axial load increases. The gradients of
can be concluded that the displacements of all the models increases with the curves remain constant until the axial load exceeds 300 kN. How
the increase of axial compressive load. However, the existence of lateral ever, when the axial load is between 300 kN and 390 kN, the gradients of
loads causes variations in out-of-plane displacements among different curves for gauges 1, 2, and 3 begin to increase. Once the axial load
models. Additionally, the strains measured at the endings and mid-spans reaches 390 kN, the measured displacements grow rapidly. In the linear
11
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 16. Displacement at strain gauges vs. applied axial force (Model 1). 4.1.2. Results of model 2
The displacements at measured points in Model 2 are shown in
elastic stage, the vertical displacements of the four measured points are Fig. 20. Measured points 1 and 2 are located on the plate, while
all below 1 mm. The measured displacements indicate that the global measured points 3 and 4 are located on the flange of the stiffener, and
deformation of the stiffened plate is not significant under axial loading
cases.
The axial strains measured at the plate are shown in Fig. 17 (a).
Initially, as the axial load grows, the strain at each measured point in
creases approximately linearly. Due to the distribution of the initial
deflection, tensile and compressive strains are captured at different
measured points. As the load continues to increase, the compressive
strains measured at points 2, 1, and 3 gradually decrease and turn into
tensile strains. This phenomenon signifies the buckling of the plates
around these measured points. After the compressive strain measured at
point 2 begins to decrease (at an axial load of 120 kN), the compressive
strains measured at point 4 accelerates to increase, indicating simulta
neous buckling of the plates around measured points 2 and 4.
The transversal strains measured at the plate are shown in Fig. 17 (b).
As the axial compressive load increases, the tensile strain measured at
point 1 sustains to increase. During the loading process, the transversal
strain measured at point 1 is larger than the axial strain. The ultimate
axial and transversal strains measured at point 2 are similar. When the Fig. 18. Local buckling of the plate.
Fig. 17. Measured strain distribution and developments of plate in the collapse process (Model 1).
12
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 19. Measured strain distribution and developments of stiffener in the collapse process (Model 1).
13
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 21. Measured strain distribution and developments of plate in the collapse process (Model 2).
14
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 22. Measured strain distribution and developments of stiffener in the collapse process (Model 2).
15
G. Guo et al. Ocean Engineering 299 (2024) 117189
measured at the ending of the flange are shown in Fig. 26(c). After the
lateral loads is applied, axial tension and vertical compression strains
are captured at the mid span of the web, while axial and vertical
compression strains are captured around the ending of the web. Addi
tionally, vertical compression strain is captured at the ending of the
flange. After the axial load exceeds 60 kN, the transversal tensile strain,
longitudinal compressive strain and vertical compressive strain
measured around the ending of the web increase rapidly. The values of
such strains near the ultimate state exceed 0.01.
After the axial load exceeds 70 kN, the transverse tensile strain
measured at the ending of the flange increases rapidly due to the Pois
son’s effect. And the measured final transversal strain is significantly
larger than that in load case 2. It can be concluded that the coupling
between axial compression and transversal torsion deformation occurs
at the endings of the stiffeners and get severer as the axial load increases.
As the stiffeners lose vertical restraint and axial bearing capacity earlier,
the global deformation of the stiffened plate occurs earlier than that in
load case 2. Meanwhile, as the stiffeners collapse earlier, the growth of
the global out-of-plane deformation get milder than that in load case 2.
Fig. 24. Displacement at strain gauges vs. applied axial force (Model 3).
4.2. Numerical simulation results and comparative analysis
measured point increase linearly with the axial load. However, the
growth of displacements accelerates after the axial load exceeds 120 kN. 4.2.1. NFEA results and comparison with experiments
Under the combined load with a larger lateral load, the stiffeners yield By taking the measured initial deflections into account for the nu
earlier compared to load case 2, and the out-of-plane displacements merical models, the ultimate strengths of the relevant numerical models
develop more gradually. are investigated, and the results are shown in Fig. 15. The ultimate
The axial strains and transversal strains measured at the plate are strength values of the three models calculated by NFEAs are 456 kN,
shown in Fig. 25. After the lateral load is applied, axial compression 383 kN and 335 kN, respectively. The errors between the finite element
strain is measured at point 1, while the axial tensile strains are measured calculation results and the experimental results are 0.8%, 4.9% and
at point 3, 5, and 6. Transversal tensile strains are measured at all the 6.3%, respectively. As the results match well, the numerical methods by
points. Initially, the strain measured at each point increases linearly as using NFEAs are reasonable and can be employed as an alternative for
the axial load increases. Under the combined load, the axial compressive ultimate strength evaluations.
strain measured at point 1 continues to increase. When the axial load The final collapse modes of the three models are shown in Fig. 27.
exceeds 100 kN, the axial compressive strain measured at points 3 and 4 The shapes and distributions of the deformations in the finite element
accelerate to increase, indicating buckling in the plate around those results are consistent with the experimental results. In axial loading
points. After the axial load exceeded 150 kN, inflections appear in the case, the primary deformation behavior of the stiffened plate is the local
axial and transversal strain curves of measured point 2, indicating buckling of the plate. However, in combined loading cases, obvious
buckling in the plate around that point. global out-of-plane deformations of the stiffened plates are observed
Fig. 26 shows the strains measured at the stiffener. The axial strains besides the local buckling of the plate. These phenomena show that the
measured at the web are shown in Fig. 26(a), the vertical strains global out-of-plane deformations of the stiffened plate get significant
measured at the web are shown in Fig. 26(b), and the transversal strains under combined loads.
Fig. 25. Measured strain distribution and developments of plate in the collapse process (Model 3).
16
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 26. Measured strain distribution and developments of stiffener in the collapse process (Model 3).
17
G. Guo et al. Ocean Engineering 299 (2024) 117189
18
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 28. Lateral load-vertical displacement curve. Fig. 29. the ultimate state curve and scatter points.
4.2.2. Transformation of the collapse modes mode will be similar to that under the axial load case. The amplitude of
To research the interaction between lateral loads and axial
compressive loads, we also established a stiffened plate model with
buckling mode initial deformation recommended by ISSC (2012).
Firstly, the axial and lateral ultimate strengths are calculated. Fig. 28
shows the displacement of the central point under lateral loading. Ac
cording to the curve, the lateral ultimate strength of the stiffened plate is
determined as 0.16 MPa while the axial ultimate strength of the stiffened
plate is 197.5 kN. In addition, a series of combine load cases with var
iable lateral loads are set to investigate the ultimate strength and the
deformation behaviors of the stiffened plates, and the lateral pressures
selected in such load cases are shown in Table 5.
Based on the calculations of axial, lateral and combined loading
cases, the dimensionless ultimate state curve of the model is derived.
Fig. 29 shows the curve and the scatter points of each loading case, and
the equation of the ultimate state curve is expressed by Eq. (9), where Au
denotes the axial ultimate strength of the stiffened plate, Au− s denotes
the model’s axial ultimate load under a combined load, Lu denotes the
lateral ultimate strength of the stiffened plate, and L denotes the cor
responding lateral load in a combined load case. From the curve we can
find that when the applied lateral load is 60% of the ultimate lateral
load, the axial ultimate load still remains 72.5% of the ultimate axial Fig. 30. Deformations of the central line.
load. After the lateral load exceeds 80% of the ultimate load, the axial
loading capacity decreases rapidly. the global deformation mode surpasses that of the local buckling mode
( )1.385 ( )2.017 after the lateral load exceeds 0.032 MPa, indicating that the global
L Au− s
+ =1 (9) deformation become the dominant collapse mode for the combined load
Lu Au
case. At the same time, the gradient of the ultimate state curve accel
Fig. 30 shows the out-of-plane deformations at the ultimate states of erates to increase.
the plate near the longitudinal central line. As the lateral load increases,
mglobal = Agloabal sin(π x / a) (10)
the deformation behavior of the stiffened plate changes from local
buckling to the combination of global buckling and local buckling. The mlocal = Alocal sin(mπ x / a) (11)
global buckling of the stiffened plate gets nonnegligible after the lateral
load exceeds 0.04 MPa. Additionally, the deformations are fitted by base According to the analysis of the experimental results, it can be
functions represent the global buckling mode (Eq. (10)) and local concluded that under the combination of axial and lateral loads, the
buckling mode (Eq. (11)). The amplitudes of the buckling modes are earlier yielding occurs at the endings of stiffeners, leading to the global
shown in Fig. 31. When the lateral load reaches 0.032 MPa, the ampli deformation. Therefore, the stress distributions in the stiffeners before
tudes of the two modes are equal. If the amplitude of the local buckling is the ultimate state in different load cases (with similar time increments)
significantly larger than that of the global deformation, the collapse are shown in Fig. 32. It can be concluded that the yielding regions
Table 5
Lateral loads selected in combined load cases.
Number of load case 1 2 3 4 5 6 7 8 9 10 11
Lateral load (MPa) 0.02 0.03 0.04 0.05 0.06 0.078 0.097 0.12 0.14 0.145 0.155
19
G. Guo et al. Ocean Engineering 299 (2024) 117189
around the endings expand downwards along the web as the lateral load
increases. After the lateral load exceeds 0.03 MPa, most region around
the endings yield before the ultimate state, resulting in noticeable ver
tical bending of stiffeners.
The stress distributions of the stiffened plates at the ultimate states
are shown in Fig. 33. As shown in Fig. 33(a), (b), If the lateral load is
below 0.03 MPa, the dominant collapse modes are the local buckling of
plate and the transversal tripping of the stiffeners. As the lateral load
increases, the global buckling of the stiffened plates gets obvious, while
the transversal tripping of stiffeners decreases. After the lateral load
exceeds 0.03 MPa, the vertical bending of the stiffeners and the global
deflection of the stiffened plate get significant. Additionally, the trans
versal tripping of stiffeners is replaced by the vertical bending at the mid
spans and the torsion at the endings. It is because the stiffeners yield
near the endings before the ultimate state, leading to the vertical
deformation of the mid span of the stiffeners. Based on the analysis
above, the critical lateral load causing the transformation in collapse
mode is confirmed to be 0.032 MPa.
Boundary conditions are also important influential factors in ulti
mate strength analysis. For comparison, the collapse modes and ultimate
Fig. 31. Amplitudes of the buckling modes. strengths of stiffened plates with clamped boundary conditions (rotation
about the edges are restricted) has been studied. The collapse modes of
20
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 33. Stress distributions at the ultimate states of different load cases.
the stiffened plates are shown in Fig. 34, while the ultimate strengths are models decreases as the lateral loads increase, and the global deforma
shown in Fig. 35. The axial ultimate strengths of the models with tion of the whole stiffened plate occurred. These phenomena result from
clamped boundary conditions get larger, and the axial ultimate strengths the yielding around the endings of the stiffeners. Additionally, the
decrease as the lateral load increases. The impact of lateral loads on axial following conclusions can be drawn.
ultimate strengths is similar to that discussed before. Moreover, as the
lateral load increases, there is an increase in global buckling and (1) The out-of-plane displacements of the stiffened plate are linearly
yielding of the stiffeners. The global buckling of the stiffened plate gets related to the axial load at the initial stage of axial loading. In this
significant after the lateral load exceeds 0.03 MPa. Comparing with the case, the out-of-plane displacements are less than 3 mm before
stiffened plates with simply supported boundary condition mentioned the axial load reaches the ultimate axial load. As the out-of-plate
earlier, the collapse modes under the same load case are consistent, and displacements of plate are restrained by stiffeners, the local
the critical lateral load causing the increase of global buckling is also buckling of the plate is the principal collapse mode.
similar. (2) In the combined load cases, the out-of-plane displacements at the
measured points increase significantly in the loading process.
5. Conclusions Severe out-of-plane deformation of plate and stiffeners are
observed. Above phenomena show that the global deformation of
This paper focuses on the ultimate strength of stiffened plates sub the model increases significantly in the combined load cases.
jected to combined loads. In the research, both experiments and NFEAs (3) In the combined load cases, the strains measured near the endings
are carried out. The results show that both the ultimate strength values of the stiffeners increase rapidly with the increased axial load. A
and collapse modes match well between the experiments and NFEAs coupled deformation of compression and torsion occurred around
(maximum difference: 6.3%). Using the designed experimental device, the endings of the stiffeners before the ultimate state, resulting in
variable lateral loads can be easily applied, and the collapse phenome the stiffeners’ earlier yielding.
non can be observed intuitively. This research can provide references for (4) Larger lateral loads cause the stiffeners deform severer and yield
designs of actual structures and similar experiments. earlier around the endings than that with smaller lateral loads.
In combined load cases, the axial ultimate strength of experimental Due to the enhancement of global buckling and decreased
21
G. Guo et al. Ocean Engineering 299 (2024) 117189
Fig. 34. Stress distributions at the ultimate states of different load cases (clamped boundary conditions).
under the conditions of the authors’ laboratory, and the scale factors
used in this study is relatively large. Experiments of larger models (with
small scale factors) which can better reflect the mechanical behaviors of
original ship structures are expected to be carried out in the future.
Data availability
Fig. 35. the ultimate strengths of the model under different load cases (clam
ped boundary conditions). Data will be made available on request.
ultimate bearing capacity of the stiffeners, the ultimate axial load Acknowledgements
capacity of the stiffened plate decreases in combined load case
with increased lateral loads. This paper is co-funded by NSFC (No. 52371328, No. 51809167, No.
(5) The deformations of the plate are composed of the local buckling 51979163 and No. U2241266), Hainan Deep-Sea Technology Innova
in plating and the global buckling in stiffened plate. As the lateral tion Center (DSTIC-CYCJ-2022007), and the Fundamental Research
load in the combined load case increases, the amplitude of global Funds for the Central Universities. The authors’ thanks also go to Prof.
deformation will gradually increase and exceed the amplitude of Zhe Li, Dr. Zhonghua Cai, Mr. Weilong Zhang and Mr. Rongpeng Cui etc.
local buckling. The critical lateral load can be determined by for their kind help in the whole process of model experiments.
equaling amplitudes of the two buckling modes at the ultimate
state (0.032 MPa in this study).
Appendix A. Supplementary data
In this research, the sizes of the experimental test model are designed
Supplementary data to this article can be found online at https://doi.
22
G. Guo et al. Ocean Engineering 299 (2024) 117189
org/10.1016/j.oceaneng.2024.117189. Ma, H.Y., Kawamura, Y., Okada, T., Wang, D.Y., Hayakawa, G., 2023. A general scaled
model design method of stiffened plate subjected to combined longitudinal
compression and lateral pressure considering the ultimate strength and collapse
References modes. Mar. Struct. 90, 103405.
Ozguc, O., 2021. Nonlinear buckling simulations of stiffened panel Exposed to
Chen, B., Guedes Soares, C., 2018. A simplified model for the effect of weld-induced combination of transverse compression, shear force and lateral pressure loadings.
residual stresses on the axial ultimate strength of stiffened plates. J. Mar. Sci. App. J. Inst. Eng. India Ser. C. 102, 397–408.
17, 57–67. Paik, J.K., Seo, J.K., 2009a. Nonlinear finite element method models for ultimate
Chen, X.W., Yuan, H.X., 2023. Local buckling behaviour of longitudinally stiffened strength analysis of steel stiffened plate structures under combined biaxial
stainless steel plate girders under combined bending and shear. Thin-Walled Struct. compression and lateral pressure actions—Part I: plate elements. Thin-Walled Struct.
184, 110541. 47 (8–9), 1008–1017.
Cui, J.J., Wang, D.Y., 2020. An experimental and numerical investigation on ultimate Paik, J.K., Seo, J.K., 2009b. Nonlinear finite element method models for ultimate
strength of stiffened plates with opening and perforation corrosion. Ocean Eng 205, strength analysis of steel stiffened-plate structures under combined biaxial
107282. compression and lateral pressure actions—Part II: stiffened panels. Thin-Walled
Cui, J.J., Wang, D.Y., Ma, N., 2016. A Study of Ultimate Strength for Container Ship Struct. 47 (8–9), 998–1007.
Bottom Structures under Bi-axial Loads Considering Corrosion Effects, the 26th Paik, J.K., Thayamballi, A.K., Chun, M.S., 1997. Theoretical and experimental study on
International Ocean and Polar Engineering Conference. International Society of the ultimate strength of bulkheads. J. Ship Res. 41, 301–317.
Offshore and Polar Engineers. Paik, J.K., Thayamballi, A.K., Lee, J.M., 2004. Effect of initial deflection shape on the
Cui, J.J., Wang, D.Y., Ma, N., 2019. Case studies on the probabilistic characteristics of ultimate strength behavior of welded steel plates under biaxial compressive loads.
ultimate strength of stiffened panels with uniform and non-uniform localized J. Ship Res. 48 (1), 45–60.
corrosion subjected to uniaxial and biaxial thrust. Int. J. Nav. Archit. Ocean Eng. 11 Peng, H., Wang, Y., Jin, L.Z., Ma, S.H., Wang, B., 2023. An isogeometric design-analysis-
(1), 97–118. optimization workflow of stiffened thin-walled structures via multilevel NURBS-
Fujikubo, M., Harada, M., Yao, T., Khedmati, M.R., Yanagihara, D., 2005a. Estimation of based free-form deformations (MNFFD). Comput. Methods Appl. Mech. Engrg. 408,
ultimate strength of continuous stiffened panel under combined transverse thrust 115936.
and lateral pressure Part 1: continuous plate. Mar. Struct. 18 (5–6), 383–410. Putranto, T., Kõrgesaar, M., Jelovica, J., 2022. Ultimate strength assessment of stiffened
Fujikubo, M., Harada, M., Yao, T., Khedmati, M.R., Yanagihara, D., 2005b. Estimation of panels using Equivalent Single Layer approach under combined in-plane
ultimate strength of continuous stiffened panel under combined transverse thrust compression and shear. Thin-Walled Struct. 180, 109943.
and lateral pressure Part 2: continuous stiffened panel. Mar. Struct. 18 (5–6), Ringsberg, J.W., Darie, I., Nahshon, K., Shilling, G., Vaz, M.A., Benson, S.,
411–427. Yanagihara, D., 2021. The ISSC 2022 committee III. 1-Ultimate strength benchmark
Gordo, J.M., Guedes Soares, C., 2008. Compressive tests on short continuous panels. Mar. study on the ultimate limit state analysis of a stiffened plate structure subjected to
Struct. 21, 113–137. uniaxial compressive loads. Mar. Struct. 79, 103026.
Gordo, J.M., Guedes Soares, C., 2012. Compressive tests on Long continuous stiffened Shanmugam, N.E., Zhu, D.Q., Choo, Y.S., Arockiaswamy, M., 2014. Experimental studies
panels. J. Offshore Mech. Arctic Eng. 134, 021403-1. on stiffened plates under in-plane load and lateral pressure. Thin-Walled Struct. 80,
Guo, Z.F., Bai, R.X., Lei, Z.K., Jiang, H., Zou, J.C., Cheng, Y., 2021. Experimental and 22–31.
numerical investigation on ultimate strength of laser-welded stiffened plates Wang, X.T., Amdahl, J., 2023. How loads interact? A numerical investigation of
considering welding deformation and residual stresses. Ocean Eng 234, 109239. aluminum stiffened panels under stiffened panels under bi-axial and lateral loads. In:
Han, P., Pak, C.J., Yun, C., Ri, K., Choe, T.L., 2022. Empirical formulations for estimation ASME 2023 42nd International Conference on Ocean, Offshore and Arctic
of the ultimate strength of the perforated stiffened panels under the combined lateral Engineering. American Society of Mechanical Engineers.
pressure and in-plane compression. Ocean Eng 266, 112620. Wang, B., Zhu, S.Y., Hao, P., Bi, X.J., Du, K.F., Chen, B.Q., Ma, X.T., Chao, Y.J., 2018.
Hanif, M.I., Adiputra, R., Prabowo, A.R., Yamada, Y., Firdaus, N., 2023. Assessment of Buckling of quasi-perfect cylindrical shell under axial compression: a combined
the ultimate strength of stiffened panels of ships considering uncertainties in experimental and numerical investigation. Int. J. Solids Struct. 130–131, 232–247.
geometrical aspects: finite element approach and simplified formula. Ocean Eng 286, Wang, Z., Kong, X.S., Wu, W.G., 2023. Empirical formula to assess ultimate strength of
115532. side shell on passenger ship under combined axial and shear load based on
ISSC, 2012. Report of specialist committee III.1 ultimate strength. Rostock, Germany. In: experimental and numerical analysis. Ocean Eng 288, 116419.
Proceedings of The18th International Ship and Offshore Structures Congress, Woloszyk, K., Garbatov, Y., Kowalski, J., 2023. Experimental ultimate strength
pp. 289–352. assessment of stiffened plates subjected to marine immersed corrosion. Appl. Ocean
Kim, D.K., Yu, S.Y., Lim, H.L., Cho, N., 2020. Ultimate compressive strength of stiffened Res. 138, 103679.
panel:an empirical formulation for Flat-bar type. J. Mar. Sci. Eng. 8, 605. Xu, M.C., Guedes Soares, C., 2011. Numerical study of the effect of geometry and
Komoriyama, Y., Yanagihara, D., 2018. A Simple design formula to estimate the ultimate boundary conditions on the collapse behavior of short stiffened panels. Adv. Mar.
strength of stiffened panels under Bi-axial compression mainly in transverse Struct. 229–237.
direction. In: Proceedings of the 37th International Conference on Ocean, Offshore Xu, M.C., Guedes Soares, C., 2013a. Experimental study on the collapse strength of
and Arctic Engineering. American Society of Mechanical Engineers. narrow stiffened panels. J. Offshore Mech. Arctic Eng. 135 (2).
Li, D.Y., Chen, Z., 2023. Progressive collapse analysis and ultimate strength estimation of Xu, M.C., Guedes Soares, C., 2013b. Comparisons of calculations with experiments on the
continuous stiffened panel under longitudinal extreme cyclic load and lateral ultimate strength of wide stiffened panels. Mar. Struct. 31, 82–101.
pressure. Ocean Eng 285, 115340. Xu, M.C., Guedes Soares, C., 2021. Experimental evaluation of the ultimate strength of
Li, S., Georgiadis, D.G., Samuelides, M.S., 2022. A comparison of geometric imperfection stiffened panels under longitudinal compression. Ocean Eng 220, 108496.
models for collapse analysis of ship-type stiffened plated grillages. Eng. Struct. 250, Xu, M.C., Zhao, J.S., Wang, T., Zhang, W.Z., Pan, J., 2022. Empirical formulae
113480. assessment of ultimate strength for perforated stiffened panels under longitudinal
Li, D.Y., Chen, Z., Chen, X.C., 2023. Numerical investigation on the ultimate strength compression. Ocean Eng 264, 112455.
behavior and assessment of continuous hull plate under combined biaxial cyclic Yang, H., Feng, S.J., Peng, H., Ma, X.G., Wang, B., Xu, W.X., Gao, Q., 2022. Uncertainty
loads and lateral pressure. Mar. Struct. 89, 103408. quantification for initial geometric imperfections of cylindrical shells: a novel bi-
Liu, B., Gao, L.J., Ao, L., Wu, W.G., 2021. Experimental and numerical analysis of stage random field parameter estimation method. Aerosp. Sci. Techonl. 124, 107554.
ultimate compressive strength of stiffened panel with openings. Ocean Eng 220, Yang, B., Fu, K.K., Yan, L., 2021. Modelling damage evolution and CAI strength of
108453. composite laminates under biaxial compression. J. Compos. Mater. 55 (21),
Ma, H.Y., Xiong, Q.F., Wang, D.Y., 2021. Experimental and numerical study on the 2887–2898.
ultimate strength of stiffened plates subjected to combined biaxial compression and Zhang, S.M., Khan, I., 2009. Buckling and ultimate capability of plates and stiffened
lateral loads. Ocean Eng 228, 108928. panels in axial compression. Mar. Struct. 22, 791–808.
Ma, H.Y., Mei, H.Y., Wang, D.Y., 2022. Scale model design for the stiffened plate Zhang, Q., Yang, H.Q., Wu, S., Cheng, W.B., Liang, Y.H., Huang, Yi, 2023. A study on the
subjected to combined longitudinal compression and lateral pressure considering ultimate strength and Failure mode of stiffened panels. J. Mar. Sci. Eng. 8 (11), 605,
collapse modes. Ocean Eng 243, 110289. 1214.
23