Structures 29 (2021) 360–372

Contents lists available at ScienceDirect

Structures
journal homepage: [Link]/locate/structures

Non-linear stability and failure of laminated composite stiffened cylindrical

panels subjected to in-plane impulse loading
Vasanth Keshav *, Shuvendu Narayan Patel, Rajesh Kumar
Department of Civil Engineering, BITS Pilani, Pilani Campus, Rajasthan, India

A R T I C L E I N F O A B S T R A C T

Keywords: In this article, the non-linear dynamic buckling behavior and failure of laminated composite stiffened cylindrical
Impulsive in-plane loads (LCSC) panel is performed in the finite element framework when subjected to sinusoidal and rectangular in-plane
Dynamic buckling pulse loading. Balanced symmetric cross-ply laminates and balanced symmetric angle ply laminates are
Composite laminates
considered in this study. The first ply failure load (FPFL) of the panel is evaluated and checked whether it occurs
Stiffeners
Cylindrical panels
before the non-linear dynamic buckling phenomenon considering four different failure theories. Convergence
and validation studies are carried out using the present mathematical formulation and compared with the results
from the existing literatures. The effect of loading duration, loading function, aspect ratio of stiffener and the ply
orientation of the skin and stiffener on the non-linear dynamic buckling of LCSC panel is studied in detail and the
results are reported. It is observed that the non-linear dynamic buckling load (DBL) of balanced and symmetric
angle ply (45◦ /− 45◦ /− 45◦ /45◦ ) stiffened panel is lower than those of unstiffened composite cylindrical panel
upto aspect ratio of the stiffener (ds/bs) equal to 8 when subjected to rectangular pulse load. In case of balanced
and symmetric cross ply (0◦ /90◦ /90◦ /0◦ ) stiffened panel the DBL of stiffened panel is lower than those of
unstiffened panel when the aspect ratio of the stiffener is less than 4 with rectangular pulse load. Further, the free
vibration, static buckling and static post-buckling analyses of the panels are carried out as and when required.

1. Introduction article, the stability of LCSC panels subjected to in-plane impulse loads is
studied.
Cylindrical panels are a part of many complex structures such as The instability in stiffened isotropic and composite plates due to
aerospace structures, marine structures and automobiles. High strength static loads is studied by many authors. The earliest investigations were
and light-weight are important characteristics required in such struc­ carried out using analytical methods for isotropic stiffened plates. In this
tures and thus composite materials are used. Stiffeners are provided in regard, Barbré [9] reported the analytical solution for the buckling of
the cylindrical panels to increase the stiffness of the cylindrical panels. plates with one or two stiffeners in the transverse and the longitudinal
Analyzing as a whole structure is time consuming and costly, so, various direction at various locations in the plate subjected to uni-axial
parts of the structures are analyzed separately. Laminated composite compressive loads. Danielson [14] presented the analytical solution
stiffened cylindrical (LCSC) panels are part of complex structures. Dur­ for the calculation of static buckling load (SBL) of an isotropic plate with
ing the service life of LCSC panels, they may be subjected to dynamic T-stiffeners subjected to in-plane compression and lateral pressure for
loads in the axial direction from neighboring components [32,22]. Such the tripping mode for the stiffeners. With the introduction of composite
loads that act for a short duration could result in instability in the LCSC materials, analytical solutions for SBL of composite materials are also
panel rendering the entire structure unstable. The instability in structure reported by numerous researchers [51,54,26,10,49]. A few experi­
can be caused due to different types of dynamic loads such as, periodic mental investigations on the buckling analysis of composite stiffened
load which induces parametric instability and impulse loads which in­ plates by Starnes et al. [47], Romeo [40], Choi et al. [13] Nagendra et al.
duces pulse buckling/dynamic buckling [17]. The loss of stability in the [29], Karachalios et al. [18] and Priyadharshani et al. [36] are available
structure due to the action of in-plane impulsive load which acts for a where the authors have considered stiffened plates with various
short duration is known as dynamic buckling [17,20,21,59]. In this boundary conditions, lamination schemes and geometries. Furthermore,

* Corresponding author.
E-mail address: [Link]@[Link] (V. Keshav).

[Link]
Received 14 July 2020; Received in revised form 25 October 2020; Accepted 12 November 2020
Available online 5 December 2020
2352-0124/© 2020 Institution of Structural Engineers. Published by Elsevier Ltd. All rights reserved.
V. Keshav et al. Structures 29 (2021) 360–372

Bradford [12] and Ovesy and Assaee [30] calculated the static buckling of LCSC panels are examined in detail.
of stiffened plates using finite strip method. With the introductions of
fast computers, finite element method (FEM) became a popular method 2. Theory and formulation
of analysis due to its capability to handle complex geometries. Many
investigations are available on the calculation of SBL of isotropic and LCSC panel is modeled by FEM. The governing equations, equations
composite stiffened plates using FEM [7,27,37,41,16,53,31] for the failure analysis and FEM modelling is presented in this section.
Stiffened cylindrical panels are also important components of com­
plex structures. Several studies are available on the stability of stiffened 2.1. Governing equations
cylindrical panels made of isotropic and composite materials subjected
to static loads. The analytical solution for buckling of LCSC panel has The dynamic responsive equation for the LCSC panel can be written
been presented by Block et al. [11] and Arciniega et al. [6]. Some as [8,20,21,33,59]
experimental studies to investigate the buckling behavior of the panels
[M]{ü} + [C]{u̇} + [K({u} ) ]{u} = [F(t) ] (2.1)
have also been undertaken [42,28,4,58,5,3,25]. Abramovich and Weller
[2] reported that the J-type stiffener geometry although have a higher
where,
collapse load, they do not improve the local SBL. The buckling of stiff­ ∫
ened isotropic stiffened panel has been studied in the FEM framework
[M] = N T ρNdxdydz
[43,44,50]. SudhirSastry et al. [48] reported the results of the buckling
analysis of LCSC panel with stiffeners of various geometries using FEM. ∫
It is observed that the static buckling analysis of isotropic and composite [K] = BT DBdxdydz
stiffened plates and cylindrical panels is investigated using various
methods extensively. ∫ ∫ ∫
The instability due to the action of periodic loads along the bound­ {F} = N T bdV + N T τdS + N T pdA
aries is of considerable importance for the case of stiffened plates and
cylindrical panels made of isotropic or composite materials which is where, ‘[M]’, ‘[C]’ and ‘[K]’ and are the mass, damping and the stiffness
studied by a few authors. Liao and Cheng [24] investigated the para­ matrices. The stiffness matrix is a function of displacements. This is due
metric instability of composite stiffened panel in the finite element to incorporation of geometric non-linearity in the panel. ‘[F(t) ]’ is the
framework using method of multiple scales for identifying the dynamic load vector. The nodal acceleration vector, velocity vector and the
instability region. Srivastava et al. [45,46] investigated the parametric displacement vector are ‘{ü}’, ‘{u̇}’ and ‘{u}’ respectively. ‘τ’, ‘ρ’, ‘p’
instability of stiffened plates with cutout subjected to uniform and non-
and, ‘b’ are the boundary traction over the surface, mass density, the
uniform in-plane loads. Kumar et al. [38] studied the parametric insta­
pressure acting on the surface and body force per unit volume respec­
bility of composite shells with central cutout. Patel et al. [33–35] pre­
tively. ‘B’ is the matrix connecting displacements and strains and ‘N’ is
sented the results of parametric instability of stiffened shell panels with
the shape function matrix. The general equation of motion for the
various geometries subjected to uniform loads and investigated the in­
stiffened cylindrical panel is given in Eq. (2.1). Here, global stiffness and
fluence of stiffener orientation and aspect ratio on the parametric
mass matrices for combined skin and stiffener are taken from Abaqus.
instability of stiffened shells. It is seen that the study on parametric
The Eq. (2.1) is a general equation of motion. After some idealization,
instability of stiffened plates and shell panels is taken up by a few in­
the equations for static buckling, natural frequency, post-buckling and
vestigators. However, there is considerable scope still existing in this
dynamic buckling problem can be taken as presented below:
area.
The instability in structures due to in-plane impulse loads is another
2.1.1. Static buckling
vital area which has been investigated by some authors. Yaffe and
The governing equation for static buckling problem is given in Eq.
Abramovich [56] analyzed stiffened isotropic cylinders subjected to si­
(2.2) [33] where ‘Pcr’ is the critical buckling pressure and ‘[KG ]’ is the
nusoidal loading function and calculated the dynamic buckling load
geometric stiffness matrix.
(DBL) both experimentally and finite element method. The authors used
Hutchinson-Budiansky criterion for dynamic buckling. Zhang et al. [57] [[K] − Pcr [KG ] ]{u} = 0 (2.2)
calculated the DBL of stiffened plates using analytical method with the
help of Budiansky-Roth criterion. The authors discussed the influence of 2.1.2. Natural frequency
duration of loading, imperfection and stiffener dimensions on dynamic The governing equation for calculating the natural frequency is given
buckling of stiffened plates. The authors also reported that when the in Eq. (2.3) [33], where, the ‘ωn’ is the frequency of vibration.
initial imperfection of the plate was large, Budiansky-Roth criterion was [ ]
not suitable. Xu et al. [55] analyzed cylindrical shells subjected to axial [K] − ω2n [M] {u} = 0 (2.3)
loads using Hamiltonian system. Less and Abramovich [23] presented
the results of the dynamic buckling of stiffened cylindrical panel using 2.1.3. Post buckling
Hutchinson-Budiansky criterion and reported that the loading duration The non-linear static displacement is calculated using post buckling
is critical in dynamic buckling analysis of stiffened cylindrical panels. It analysis (Eq. (2.4)), [21,8] which is solved using static Riks approach.
is evident that a very few works are available in the literature about the [K{u} ]{u} = F (2.4)
instability of structures due to impulse load and the dynamic buckling of
LCSC panels along with failure remains unexplored.
2.1.4. Dynamic buckling
In the present investigation, the non-linear dynamic buckling and
The equation used to solve dynamic buckling problem is given in Eq.
first ply failure of LCSC panels is investigated using finite element
(2.5) [8,20,21,59]. The damping in the panel is not considered in the
software, Abaqus. The first ply failure is investigated to ascertain the
current study.
precedence of DBL and the first ply failure load (FPFL) in the LCSC
panels. The accuracy of the present finite element method is examined [M]{ü} + [K({u} ) ]{u} = [F(t) ] (2.5)
by comparing the present results with those available in the literature.
Several new results considering the influence of loading duration, 2.2. Failure studies
loading function, aspect ratio of stiffeners (ds/bs) and lamination scheme
of panel skin and stiffeners on the dynamic buckling and first ply failure The FPFL of the panel is calculated and compared with the DBL. The

361
V. Keshav et al. Structures 29 (2021) 360–372

failure criteria used in the current study are Tsai -Wu criterion, Tsai Hill
criterion, Azzi-Tsai-Hill criterion and Maximum stress criterion. The
equations for calculating the failure index using different theories are
presented in the subsequent sections [39,1]. In these equations (Eqs.
(2.6)–(2.9)), ‘σ11’ is the stress in normal direction, ‘σ 22’ is the stress in
the transverse direction, ‘σ12’ is the stress in the 1–2 plane. ‘YC’ and ‘YT’
are the compressive and tensile strengths in transverse direction
respectively. ‘XC’ and ‘XT’ are the compressive and tensile strengths in
normal direction respectively. ‘S’ is the shear strength.

2.2.1. Azzi-Tsai-Hill failure theory

The Azzi-Tsai-Hill failure criterion is given by Eq. (2.6).

σ 211 |σ11 σ 22 | σ222 σ 212

FailureIndex = − + 2+ 2 <1 (2.6)
X2 X2 Y S

2.2.2. Maximum stress failure theory

If σ22 > 0, Y = YT; otherwise, Y = YC. If σ11 > 0, X = XT; otherwise, X
= XC. The failure index is given by Eq. (2.7). Fig. 1. Ply orientation in skin and stiffener in laminated composite stiff­
ened panel.
(σ σ ⃒σ ⃒ )
22 ⃒ 12 ⃒
(2.7)
11
FailureIndex = max , , ⃒ ⃒ <1
X Y S
element size in the panel. Moreover, all the values in the manuscript are
non-dimensionalized. The lowest distance between two nodes is 5.955
2.2.3. Tsai-Hill failure theory
× 10− 3 unit and the highest distance between two nodes is 0.02 unit in a
If σ22 > 0, Y = YT; otherwise, Y = YC. If σ11 > 0, X = XT; otherwise, X
maximum of unit dimension of the panel. Non-linear kinematic re­
= XC. The Tsai-Hill failure criterion is given by Eq. (2.8).
lationships due to large deformations are considered. The calculations
σ 211 σ11 σ 22 σ222 σ 212 are done at local coordinates and the results are utilized in the subse­
FailureIndex = − + + <1 (2.8)
X2 X2 Y2 S2 quent steps [1]. In Abaqus/Explicit, the time integration is done using
central difference method.
2.2.4. Tsai-Wu failure theory
The Tsai-Wu failure criterion is given by Eq. (2.9). 3. Results and discussion
FailureIndex = F1 σ 11 + F2 σ 22 + F11 σ211 + F22 σ222 +F66 σ 212 + 2F12 σ 11 σ22 < 1
In this part, the results of convergence study and the validation
(2.9)
studies are presented first. Then the results and discussions of the cur­
The coefficients are defined as: rent study are furnished.
1 1 1 1 − 1
F1 = + F2 = + F11 =
XT XC YT YC XT X C 3.1. Convergence study and the validation study
− 1 1 √̅̅̅̅̅̅̅̅̅̅̅̅̅
F22 = F66 = F12 = f F11 F22 For convergence and validation study, stiffened plate and stiffened
YT YC S2
cylindrical panel are considered. The validation study of shock spectrum
In this study, ‘f’ = − 0.5 corresponding to the material considered of an isotropic plate and the validation of dynamic buckling of an
[15]. orthotropic plate are also presented in this section.

3.1.1. Buckling of stiffened plate

2.3. Finite element model The results of convergence and validation studies on the SBL of
stiffened plate is presented in this section. For which, a square, simply
For modeling the LCSC panel, both the stiffeners and the skin are supported stiffened plate is considered. The geometric properties are: bs/
modeled as shells. The geometry of the cylindrical panel along with the h = 2, ds/bs = 2, a/h = 100 and a/b = 1. The material properties E = 2.11
stiffener is drawn and extruded to the required length. The mesh density × 1011 N/m2 and υ = 0.3. The width of the stiffener is denoted by ‘bs’, ‘ds’
near the junction of the stiffener and the skin is kept more than that at is the depth of the stiffener and ‘h’ is the depth of the skin. Two stiffener
the edges of the skin. This would help in determining if the failure occurs orientations are considered for this case, one along the direction of
at the junction of the skin and the stiffener. The ply orientation in the loading and one in the perpendicular direction. Fig. 2 shows the ge­
skin and the stiffeners are perpendicular to each other as shown in Fig. 1. ometry. Fig. 3(a) shows the pre-buckling boundary conditions. Fig. 3(b)
In Fig. 1, ‘n’ represents the normal direction and the piles are oriented in shows the buckling boundary conditions. The mesh size for the stiffener
counterclockwise about ‘n’. This means, 0◦ aligns with ‘1’ direction and is 100 × 4 and the mesh size for the skin is varied. This results in higher
90◦ aligns with ‘2’ direction. mesh density the junction of the skin and the stiffener. Fig. 4 shows the
In Abaqus, both S4R and S3 shell elements are used to model the non-dimensional buckling loads with different total number of elements
LCSC panel. A portion of the panel consisting of both S3 and S4R ele­ of the plate with one Y-directional stiffener. Similar results are plotted in
ments is shown in Fig. 16. S4R is a four-node element with four corner Fig. 5 for the stiffened plate with one X-directional stiffener. In both
nodes and one integration point at the center. S3 is a three-node trian­ Figs. 4 and 5, the total number of elements in the plate are shown.
gular element with three corner nodes and one integration point at the Table 1 shows the results of present study and the FEM results of Patel
center. The S4R is a general-purpose element with finite membrane et al. [33]. The buckling load is non-dimensionalized using Eq. (3.1).
strains. In the present investigation, the edges of the panel are seeded
and then Abaqus is allowed to mesh the panel automatically. So, the NXX b2
N XX = (3.1)
edges of the skin are divided into 50 divisions and the edges of the D
stiffener are divided into 100 divisions. This results in an irregular The results converge at total number of elements 3076. This

362
V. Keshav et al. Structures 29 (2021) 360–372

Fig. 4. Non-dimensional buckling load vs total number of elements for a plate

with one Y-directional stiffener.
Fig. 2. Stiffened plate geometry.

Fig. 5. Non-dimensional buckling load vs total number of elements for a plate

with one X-directional stiffener.

Table 1
Validation of buckling load for stiffened plate.
Fig. 3a. Pre buckling conditions for stiffened plate. Analysis Nondimensional Static Buckling load

Stiffener along Y-axis Stiffener along X-axis

Present 68.4891 120.6261

Patel et al. [33] 67.036 119.168

bs/h = 2, ds/bs = 2, a/h = 100, a/b = 1 and material properties E = 2.11

× 1011 N/m2 and υ = 0.3. Fig. 6 shows the panel geometry. Fig. 7(a)
shows the pre-buckling boundary conditions. Fig. 7(b) shows the
buckling boundary conditions. The discretization is done in a similar
way as the previous section. The results are non- dimensionalized using

Fig. 3b. Buckling boundary conditions for stiffened plate.

corresponds to 50 × 50 divisions for the edges of the skin of the panel

and 100 × 4 divisions for the stiffener. The total number of elements are
for S4R and S3 elements together in the LCSC panel. From Table 1, it is
observed that the present study results match well with the FEM results
of Patel et al. [33].

3.1.2. Static buckling of cylindrical panel

In this part, the results of the SBL of a stiffened cylindrical panel are
validated. The geometric properties of the cylindrical panel are R/a = 2,
Fig. 6. Geometry of the stiffened cylindrical panel.

363
V. Keshav et al. Structures 29 (2021) 360–372

sections. The results of the current study are presented along with the
analytical results of Azarboni et al. [8] in Fig. 8.
It is observed from the Fig. 8 that the results of the present study
agree with the analytical results of Azarboni et al. [8]. The peak response
is observed in the neighborhood of T/Tb = 1.

3.1.4. Validation of DBL

In this section, the DBL of an orthotropic plate is computed and
validated. A simply supported orthotropic plate is considered. It is
subjected to rectangular impulse load. The geometric properties and the
material properties are presented in Kowal-Michalska and Mania [20].
Imperfection in the plate is in the form of the first buckling mode shape.
As done in Section 3.1.3, the boundary conditions in pre-buckling and
buckling stages are same (Fig. 3b). Fig. 9 depicts the plot of Ndyn/Nst vs.
w/h for plate subjected to rectangular pulse loading. The results are
compared with the FEM results of Kowal-Michalska and Mania [20]. The
results of the present study match well.
Fig. 7a. Pre buckling boundary conditions for stiffened cylindrical panel.
3.2. Present study

After the validation study, the non-linear dynamic buckling of a

LCSC panel is investigated and the effect of duration of loading, stiffener
aspect ratio, loading function and ply orientation of skin and stiffener
are also studied. The FPFL is calculated in order to check if the dynamic
buckling is occurring before first ply failure. Fig. 10(a) shows the ge­
ometry. The boundary conditions are in Fig. 10(b). The boundary con­
ditions are applied on the edges of the skin as well as on the vertical
edges of the stiffener. The DBL is calculated using Vol’mir criterion: the
in-plane dynamic load at which the maximum displacement is equal to
the thickness [52]. The results obtained from this criterion and the re­
sults from Budiansky-Hutchinson criterion vary by 10% [21]. Moreover,
Budiansky-Hutchinson criterion is not suitable for stiffened cylindrical
panels [32]. In the current study, the ratio v/h = 1 is critical and the
magnitude of load at which this ratio is reached, is the DBL. The material
properties are taken from the book of world-wide failure exercise [15]
conducted between 1998 and 2004. Table 3 shows the material
properties.
Fig. 7b. Buckling boundary conditions for stiffened cylindrical panel. The stacking sequence of the stiffener is perpendicular to the stack­
ing sequence of the skin as shown in Fig. 1 and both the stacking se­
Eq. (3.1). The results are compared with the FEM results from Patel et al. quences are perpendicular to their normal direction in the local level.
[33] in Table 2. Two types of ply orientations are considered for the study: cross-ply
From Table 2 it is observed that the results match well. The total laminates (0◦ /90◦ /90◦ /0◦ ) and angle ply laminates (45◦ /− 45◦ /− 45◦ /
number of elements in the stiffened panel is 5489. This corresponds to 45◦ ) in both skin and the stiffener of the LCSC panel. The geometry of the
50 × 50 divisions for the edges of the skin of the panel and 100 × 4 panel is not changed; a/b = 1, a = 1 m, bs/h = 2 and a/h = 100 is the
divisions for the stiffener. The total number of elements are calculated same for all cases. Also, ρ = 1700 kg/m3 [19]. Two types of loading
taking both S4R and S3 elements together in the LCSC panel. It is also functions are considered in the study rectangular impulse load (Fig. 11a)
observed that the number of divisions along the stiffener depth is not and sinusoidal impulse load (Fig. 11b). The dynamic load (Ndyn) is the
affecting the overall results. maximum amplitude of the dynamic in-plane impulse load the panel is
subjected to considering the two loading functions. The sinusoidal im­
3.1.3. Validation of shock spectrum pulse load is considered only in Section 3.2.3 and in the remaining
The shock spectrum of an unstiffened plate is presented in this part. A sections, only rectangular impulse load is considered. The SBL and the
square plate is subjected to dynamic impulse load. The maximum first natural period of LCSC panel is calculated using Abaqus/Standard.
deflection of the plate is observed for various loading durations. The
geometric properties and the material properties of the plate are
described in the work of Keshav et al. [19]. The boundary conditions in
the pre-buckling and the buckling stages are taken same. This is different
from the ones taken to validate the SBL of the plate in the previous

Table 2
Non-dimensional buckling load for stiffened cylindrical panel.
Analysis Non-dimensional buckling load

Stiffener along Z-axis Stiffener along X-axis

Present 348.5105 332.390

Patel et al. [33] 347.952 335.981
Fig. 8. Shock spectrum of unstiffened plate subjected to load Ndyn = 3Nst.

364
V. Keshav et al. Structures 29 (2021) 360–372

Table 3
Effective material properties of each laminae in the material direction (T300/
BSL914C Lamina).
Property Magnitude

Longitudinal modulus, E1 (GPa) 138

Transverse modulus, E2 (GPa) 11
In-plane shear modulus, G12 (GPa) 5.5
Major Poisson’s ratio, υ12 0.28
Longitudinal tensile strength, XT (MPa) 1500
Longitudinal compressive strength, XC (MPa) 900
Transverse tensile strength, YT (MPa) 27
Transverse compressive strength, YC (MPa) 200
In-plane shear strength, S12 (MPa) 80

Fig. 9. Ndyn/Nst vs w/h for orthotropic plate subjected to sinusoidal im­

pulse load.

Fig. 11a. Rectangular loading function.

Fig. 10a. Geometry of the cylindrical panel studied.

Fig. 11b. Sinusoidal loading function.

Then, for a specific duration, and different magnitudes, the transverse

displacements are evaluated with help of Abaqus/Explicit. The panel is
loaded for a certain duration and the transverse displacements are
observed after the removal of the load as well. The magnitude of load for
which the transverse displacement reaches the value of the thickness of
the panel is the DBL. In each case, Nst is the SBL of the panel and Tn is the
first natural period of the panel. With the change in the geometry and
stacking sequence of the panel, these values change. In each case, the
ratio Ndyn/Nst is unique for the panel with respective geometry and
stacking sequence. The influence of duration, loading function, aspect
ratio of stiffeners (ds/bs) and lamination scheme of the skin and the
stiffeners on the non-linear dynamic buckling of LCSC panels are pre­
sented in the following sections. The simply supported boundary con­
ditions used in the Sections 3.1.3 and 3.1.4 are used. The panel is
discretized in a similar way as done in the validation study, i.e., the four
edges of the skin of the LCSC panel are divided into 50 × 50 divisions,
Fig.10b. Boundary conditions of the stiffened cylindrical panel. the stiffener is divided into 100 divisions along the length and divisions
along the depth is 2 for ds/bs = 2, 4 for ds/bs = 4 and like this to keep the
element size same in the stiffeners. Abaqus is allowed to mesh the panel
so as to get a higher density of mesh at the junction of the skin and the

365
V. Keshav et al. Structures 29 (2021) 360–372

stiffener which contains both S4R and S3 elements. The DBL and FPFL of
the panel in all cases are expressed in terms of their corresponding SBL.

3.2.1. Influence of duration of loading

The influence of loading duration on the dynamic buckling of LCSC
panel is discussed in this section. The LCSC panel with R/a = 10 and ds/
bs = 1 is considered. Two ply orientations are (0◦ /90◦ /90◦ /0◦ ) and (45◦ /
− 45◦ /− 45◦ /45◦ ). The durations of loading are: ¼ Tn, ½ Tn, Tn, 2Tn and
4Tn. The LCSC panel with ply orientation (0◦ /90◦ /90◦ /0◦ ) has Nst =
252563 N/m and Tn = 17.585 × 10− 3 s. the LCSC panel with ply
orientation 45◦ /− 45◦ /− 45◦ /45◦ ) has Nst = 387496 N/m and Tn =
13.486 × 10− 3 s. The plot of Ndyn/Nst vs v/h for the LCSC panel with R/a
= 10 and ply orientation (0◦ /90◦ /90◦ /0◦ ) for various loading durations
for rectangular impulse loading is shown in Fig. 12. The plot of Ndyn/Nst
vs v/h for the LCSC panel with R/a = 10 and ply orientation (45◦ /− 45◦ /
− 45◦ /45◦ ) for various loading durations subjected to rectangular im­ Fig. 13. Ndyn/Nst vs v/h for LCSC panel with R/a = 10 and ply orientation
(45◦ /− 45◦ /− 45◦ /45◦ ) for various durations of loading subjected to rectangular
pulse loading is shown in Fig. 13.
loading function.
It is evident from Figs. 12 and 13 that when the duration of loading is
very short, the DBL is high and when the duration of loading is
increased, the DBL decreases. No appreciable variation in response is
observed when the loads are applied for longer durations. The critical
duration of loading for the panel is its first natural period. This obser­
vation is consistent with the known fact that for a structure, when the
loading duration is near its first natural period, maximum response is
observed (during earthquakes). It is also observed that the DBL for the
stiffened panel is lesser than the SBL when the loading duration is closer
to first natural period. The SBL is higher than the DBL for all cases of
loading durations when the ply orientation (45◦ /− 45◦ /− 45◦ /45◦ ). For
the rest of the investigation, the panel is subjected to loads till the first
natural period.

3.2.2. First ply failure of LCSC panel

The FPFL of the LCSC panel is presented in this section. It is evaluated Fig. 14. Ndyn/Nst vs v/h for LCSC panel with (0◦ /90◦ /90◦ /0◦ ) laminates.
to check the precedence of the DBL and the FPFL. The LCSC panel with
R/a = 10 and ds/bs = 1 is considered. The ply orientation is (0◦ /90◦ /90◦ /
0◦ ). Fig. 14 represents the plot of Ndyn/Nst vs v/h for the LCSC panel with
(0◦ /90◦ /90◦ /0◦ ) laminates. In this figure, the horizontal line at v/h = 1
is the critical line. The load at which the maximum transverse
displacement in the LCSC panel reached the thickness of the panel, is the
DBL of the panel. Fig. 15 shows the plot of Ndyn/Nst vs Failure Index for
the panel with respect to various failure theories. Again, the horizontal
line at Failure index = 1 is the critical.
Fig. 16(a) and (c) show the deformed shape of the panel at Ndyn/Nst
= 0.7 and Ndyn/Nst = 1 respectively corresponding to maximum trans­
verse displacement. Fig. 16(b) and (d) show the deformed shape of the
panel at Ndyn/Nst = 0.7 and Ndyn/Nst = 1 respectively corresponding to
maximum failure index (Tsai-Wu criterion). The region in red represents
Fig. 15. Ndyn/Nst vs Failure Index for panel with (0◦ /90◦ /90◦ /0◦ ) laminates.

the area with maximum failure index. The maximum value of failure
index is considered for plotting Fig. 15. Ndyn = 0.7Nst is the dynamic
load near the DBL of the panel and Ndynt = Nst is the dynamic load near
the FPFL of the panel.
The failure index is calculated at all four layers of the panel (skin and
the stiffener). In this particular case (Fig. 16d), the maximum failure
index occurs at the junction of the skin and the stiffener of the panel. At
the node, the failure index at first layer is 1.1238, at second layer is
0.3406, at third layer is 0.3108 and at fourth layer is 0.6127. So, the first
ply (bottommost ply) is failing first.
From Fig. 14 it is seen that the DBL for LCSC panel is lower than the
SBL and is around 0.6–0.7Nst. In Fig. 15, although the plot for Maximum
stress criterion, Azzi-Tsai-Hill criterion and Tsai-Hill criterion merge,
the values are not the same for all cases. Some variation in the values are
Fig. 12. Ndyn/Nst vs v/h for LCSC panel with R/a = 10 and ply orientation (0◦ / observed. Also, the FPFL for the LCSC panel is lower than its SBL. Global
90◦ /90◦ /0◦ ) for various durations of loading subjected to rectangular buckling is observed in the LCSC panel at DBL (Fig. 16a). In Fig. 15
loading function.

366
V. Keshav et al. Structures 29 (2021) 360–372

Fig. 16. Deformed LCSC panel having (0◦ /90◦ /90◦ /0◦ ). (a) With respect to maximum displacement at Ndyn/Nst = 0.7. (b) With respect to Tsai-Wu failure criterion at
Ndyn/Nst = 0.7 (c) With respect to maximum displacement at Ndyn/Nst = 1. (d) With respect to Tsai-Wu failure criterion at Ndyn/Nst = 1. The scale factor is 3 for
all cases.

although higher failure indices are shown, progressive failure is not

studied. In Fig. 16(a) and (c), the blue region at the center of the panel is
the area of maximum deformation. In Fig. 16(b) and (d), the red region
is the area of the maximum failure index. From Figs. 14 and 15, it is also
observed that the DBL of the panel with stiffener aspect ratio (ds/bs) = 1
is lower than its FPFL and SBL.

3.2.3. Influence of loading function

The influence of loading functions is studied in this section. The LCSC
panel with ds/bs = 1 is considered. The ply orientation is (0◦ /90◦ /90◦ /
0◦ ) for both the skin and the stiffener of the panel. The radius of cur­
vature considered is R/a = 20, and R/a = 10. The loading duration is the
first natural period. Fig. 17 shows the plot of Ndyn/Nst vs v/h for the LCSC
panel having various curvatures and subjected to both types of impulse
loading functions. Fig. 18 shows the plot of Ndyn/Nst vs Failure Index
Fig. 18. Ndyn/Nst vs Failure Index from Tsai-Wu failure criterion for LCSC panel
with respect to Tsai-Wu failure criterion for LCSC panel having various
with various curvatures and for both loading functions.
curvatures and subjected to both types of pulse loading functions.
It is observed from Figs. 17 and 18 that the DBL and FPFL is less when
rectangular impulse load is acting on both the panel in comparison to the area under the curve is higher in rectangular loading function. In the
those with the sinusoidal pulse load. This is because, higher energy is case of dynamic loading, both the duration as well as the magnitude of
imparted into the panel due to rectangular impulse load is acting, since the load are critical. When the duration of loading is same for all cases,
the magnitude of the DBL will be the least in case of rectangular type
loading case. It is also observed that in both cases of radius of curvatures,
the DBL is lesser than the FPFL and also their corresponding SBL.

3.2.4. Influence of curvature of the LCSC panel

The influence of curvature is studied in this section. LCSC panel with
ds/bs = 1 is considered. The ply orientations are (0◦ /90◦ /90◦ /0◦ ) and
(45◦ /− 45◦ /− 45◦ /45◦ ). Three types of curvatures are considered: R/a =
20, 10 and 5. Fig. 19 shows the plot of Ndyn/Nst vs v/h of the LCSC panel
for various radius of curvatures having ply orientation (0◦ /90◦ /90◦ /0◦ )
and (45◦ /− 45◦ /− 45◦ /45◦ ). Fig. 20 shows the plot of Ndyn/Nst vs Failure
Index with respect to Tsai-Wu failure criterion of the LCSC panel for
various radius of curvatures having ply orientation (0◦ /90◦ /90◦ /0◦ ) and
(45◦ /− 45◦ /− 45◦ /45◦ ).
It is evident from Figs. 19 and 20 that the DBL and FPFL is the lowest
for R/a = 5 for panel with ply orientation (0◦ /90◦ /90◦ /0◦ ). As the ratio
Fig. 17. Ndyn/Nst vs v/h for LCSC panel with various curvatures and for both R/a increases, the DBL increases. However, when the ply orientation is
loading functions.

367
V. Keshav et al. Structures 29 (2021) 360–372

Fig. 21. Ndyn/Nst vs v/h for LCSC panel with respect to different stiffener
Fig. 19. Ndyn/Nst vs v/h for LCSC panel with both ply orientations and for aspect ratios.
different radius of curvatures subjected to rectangular impulse load.
10. Panel with ds/bs = 0 signifies unstiffened cylindrical panel. Fig. 21
depicts the plot of Ndyn/Nst vs v/h for the LCSC panel for different
stiffener aspect ratios. Fig. 22 depicts the plot of Ndyn/Nst vs Failure
Index with respect to Tsai-Wu failure criterion for the LCSC panel for
different stiffener aspect ratios.
It is observed that the DBL for LCSC panel is the lowest for ds/bs = 2
(Fig. 21). The same is true for the FPFL (Fig. 22). The DBL for the panel
with ds/bs = 1 and ds/bs = 2 is lower than the unstiffened panel. As the
stiffener aspect ratio is increased, the DBL increases. The panel with DBL
for stiffened panel with ds/bs = 4, 8 and 10 is 1.2–1.3Nst and jump in
response is observed in this region. In Fig. 22, a large difference in the
FPFL for panel with ds/bs = 1 and ds/bs = 2 is observed. Furthermore, the
DBL for panel with ds/bs = 4, 8 and 10 is more than the corresponding
SBL and corresponding FPFL.
In this case also, in Figs. 21 and 22, the Nst considered for each panel
is its own SBL. It can be seen from the figures that each panel is having
different dynamic instability performance if compared to its own static
Fig. 20. Ndyn/Nst vs Failure Index for Tsai-Wu failure criterion for LCSC panel
instability performance. The unstiffened panel is showing better dy­
with both ply orientations and for different radius of curvatures subjected to
namic performance if seen with respect to its static performance than
rectangular impulse load.
those of the stiffened panels with ds/bs = 1 and ds/bs = 2. However, for
the stiffened panels with ds/bs = 4 and ds/bs = 8, this behaviour is
(45◦ /− 45◦ /− 45◦ /45◦ ), the lowest DBL and FPFL is for the panel with R/
opposite. This kind of behaviour can be attributed to combined effect of
a = 10. In case of angle ply laminates in a panel with R/a = 5 or 10, the
stiffness and mass of the panels in case of in-plane impulse loading
DBL and FPFL of the panel is at least 50% lower than its SBL. Further­
environment. In the context of first ply failure behaviour the panel with
more, a sharp increase in displacement is observed in the panel with R/a
ds/bs = 1 is showing better results among all the panels.
= 10 and R/a = 5 for both lamination schemes. It is seen that the panel
with R/a = 20 has the highest DBL and the FPFL both ply orientations.
3.2.6. Influence of ply orientation
As the curvature of the panel decreases, the DBL and the FPFL of the
In this part, the influence of ply orientation is studied. The three
panel increases. In the Figs. 19 and 20 the Nst considered for each panel
curvatures considered are: R/a = 5, 10 and 20. The ply orientations (0◦ /
corresponds to its own static buckling load. It is clear from both the
90◦ /90◦ /0◦ ) and (45◦ /− 45◦ /− 45◦ /45◦ ) for the skin and the stiffener are
figures that each panel is showing different dynamic behavior in
considered. Aspect ratios of the stiffeners are ds/bs = 0,1,4 and 8.
connection to its static buckling behaviour. The panel with lower radius
The results of the DBL and the FPFL for the panels with various aspect
of curvature is showing lower dynamic performance (Ndyn/Nst ratio)
compared to that of the panel with higher radius of curvature. It means
the DBL of the panel in relation to its SBL with less radius of curvature is
less than that of the panel with higher radius of curvature. The stiffness
of the panel increases with the decrease in the radius of curvature. So,
the panel with lower radius of curvature should show better perfor­
mance. However, it is not the case. This is because, in dynamic loading
case both stiffness and mass of the panel come into picture. So, the
combination of mass and stiffness is making the panel with less radius of
curvature, dynamically more unstable when seen in relation to the static
instability in case of in-plane impulse loading. This behaviour is almost
similar in the context of first ply failure behaviour of the panels.

3.2.5. Influence of stiffener size

In this part, the influence of stiffener aspect ratio (size) is studied.
The panel with R/a = 10 is considered. The ply orientation is (0◦ /90◦ /
90◦ /0◦ ). The various aspect ratios considered are ds/bs = 0,1,2,4,8 and Fig. 22. Ndyn/Nst vs Failure Index for Tsai-Wu failure criterion with respect to
different stiffener aspect ratios.

368
V. Keshav et al. Structures 29 (2021) 360–372

Fig. 23. Ndyn/Nst vs v/h and Ndyn/Nst vs Failure Index with respect to Tsai-Wu failure criterion for the panel with ply orientation (0◦ /90◦ /90◦ /0◦ ) and various aspect
ratios of the stiffeners subjected to rectangular impulse loads: (a) Ndyn/Nst vs v/h for the panel with R/a = 20 (b) Ndyn/Nst vs v/h for the panel with R/a = 10 (c) Ndyn/
Nst vs v/h for the panel with R/a = 5 (d) Ndyn/Nst vs Failure Index for the panel with R/a = 20 (e) Ndyn/Nst vs Failure Index for the panel with R/a = 10 (f) Ndyn/Nst vs
Failure Index for the panel with R/a = 5.

Fig. 24. Ndyn/Nst vs v/h and Ndyn/Nst vs Failure Index with respect to Tsai-Wu failure criterion for the panel with ply orientation (45◦ /− 45◦ /− 45◦ /45◦ ) and various
aspect ratios of the stiffeners subjected to rectangular impulse loads: (a) Ndyn/Nst vs v/h for the panel with R/a = 20 (b) Ndyn/Nst vs v/h for the panel with R/a = 10
(c) Ndyn/Nst vs v/h for the panel with R/a = 5 (d) Ndyn/Nst vs Failure Index for the panel with R/a = 20 (e) Ndyn/Nst vs Failure Index for the panel with R/a = 10 (f)
Ndyn/Nst vs Failure Index for the panel with R/a = 5.

ratios of the stiffeners and curvatures for ply orientations (0◦ /90◦ /90◦ / better dynamic performance with respect to their static performance
0◦ ) and (45◦ /− 45◦ /− 45◦ /45◦ ) are shown in Figs. 23 and 24 respec­ than those of unstiffened panel and stiffened panel with ds/bs = 1. It is
tively. The deformed panels with R/a = 10 at DBL and FPFL are shown in already observed in Section 3.2.5. Again, the unstiffened panel is better
Figs. 25 and 26. For these figures, the scale factor = 3 is provided in Y- than the stiffened panel with ds/bs = 1. This behaviour similar for the
direction only. Due to this reason, the depth of the stiffener may seem panels with all radius of curvature. In the matter of first play failure case,
unrealistic. However, this is done in order to visualize the deformations the unstiffened panel is performing better among all panels with all
in the LCSC panel. three curvatures. However, in case of (45◦ /− 45◦ /− 45◦ /45◦ ) stacking
In this section again, in Figs. 23 and 24, the Nst considered for each sequence, the unstiffened panel is performing better than the stiffened
panel is its own SBL. In case of the panels with (0◦ /90◦ /90◦ /0◦ ) stacking panels both in regard to dynamic buckling and first ply failure behaviour
sequence, the stiffened panels with ds/bs = 4 and ds/bs = 8, is showing in case of panels with all radius of curvatures. Also, in all the cases, the

369
V. Keshav et al. Structures 29 (2021) 360–372

Fig. 25. Deformed LCSC panel with ply orientation (0◦ /90◦ /90◦ /0◦ ) (a) With respect to maximum transverse displacement at Ndyn/Nst = 1.3 (ds/bs = 4). (b) With
respect to maximum transverse displacement at Ndyn/Nst = 1.3 (ds/bs = 8). (c) With respect to Tsai-Wu failure criterion at Ndyn/Nst = 1.1 (ds/bs = 4). (d) With respect
to Tsai-Wu failure criterion at Ndyn/Nst = 1.1 (ds/bs = 8). The scale factor is 3 for all cases.

Fig. 26. Deformed LCSC panel with ply orientation (45◦ /− 45◦ /− 45◦ /45◦ ) (a) With respect to maximum transverse displacement at Ndyn/Nst = 0.4 (ds/bs = 1). (b)
With respect to maximum transverse displacement at Ndyn/Nst = 0.325 (ds/bs = 4). (c) With respect to maximum transverse displacement at Ndyn/Nst = 0.5 (ds/bs =
8). (d) With respect to Tsai-Wu failure criterion at Ndyn/Nst = 0.45 (ds/bs = 1). (e) With respect to Tsai-Wu failure criterion at Ndyn/Nst = 0.275 (ds/bs = 4). (f) With
respect to Tsai-Wu failure criterion at Ndyn/Nst = 0.3 (ds/bs = 8). The scale factor is 3 for all cases.

DBL and the FPFL of the stiffened and unstiffened panels are lower than of LCSC panel subjected to impulse loads are investigated, and the
their corresponding SBL. following conclusions are drawn.

4. Conclusions 1. LCSC panel when subjected to loads till its first natural period or
higher duration have lower DBL than when subjected to a very
In this article, the non-linear dynamic buckling behavior and failure short duration.

370
V. Keshav et al. Structures 29 (2021) 360–372

2. The DBL of LCSC panel when subjected to rectangular impulse [10] Bisagni C, Vescovini R. Analytical formulation for local buckling and post-buckling
analysis of stiffened laminated panels. Thin-Walled Struct 2009;47(3):318–34.
loads is lower than that with the sinusoidal impulse load for panel
[Link]
with ply orientation (0◦ /90◦ /90◦ /0◦ ) due to higher energy [11] Block DL, Card MF, & Mikulas Jr MM. (1965). Buckling of eccentrically stiffened
imparted into the system. orthotropic cylinders NASA Technical note, NASA TN D-2960.
3. LCSC panel when subjected to rectangular impulse load results in [12] Bradford MA. Buckling of longitudinally stiffened plates in bending and
compression. Can J Civ Eng 1989;16(5):607–14. [Link]
lower FPFL than that when subjected to a sinusoidal impulse [13] Choi BH, Hwang MO, Yoon TY, Yoo CH. Experimental study of inelastic buckling
load. strength and stiffness requirements for longitudinally stiffened panels. Eng Struct
4. For LCSC panel with (0◦ /90◦ /90◦ /0◦ ) laminates, as the ratio R/a 2009;31(5):1141–53. [Link]
[14] Danielson DA, Cricelli AS, Frenzen CL, Vasudevan N. Buckling of stiffened plates
decreases, the DBL and the FPFL decrease when subjected to under axial compression and lateral pressure. Int J Solids Struct 1993;30(4):
rectangular impulse load. 545–51. [Link]
5. The DBL for LCSC panel with (45◦ /− 45◦ /− 45◦ /45◦ ) laminates is [15] Hinton MJKA, Soden PD, Kaddour AS (Eds.). (2004). Failure criteria in fibre
reinforced polymer composites: the world-wide failure exercise. Elsevier. DOI:
maximum for R/a = 10 in comparison to those with R/a = 5 or R/ 10.1016/B978-0-080-44475-8.X5000-8.
a = 20. However, for all the three cases of the curvatures [16] Jain HK, Upadhyay A. Buckling behavior of blade-, angle-, T-, and hat-stiffened
considered, the DBL of the panel is lower than its SBL. FRP panels subjected to in-plane shear. J Reinf Plast Compos 2010;29(24):
3614–23. [Link]
6. In the case of LCSC panel with (0◦ /90◦ /90◦ /0◦ ) laminates sub­ [17] Jansen EL. Dynamic stability problems of anisotropic cylindrical shells via a
jected to rectangular impulse load, the DBL of the panels with ds/ simplified analysis. Nonlinear Dyn 2005;39(4):349–67. [Link]
bs = 1 and ds/bs = 2 is lower their corresponding SBL. And the s11071-005-4343-1.
[18] Karachalios E, Vrettos C, Marioli-Riga Z, Bisagni C, Cordisco P, De Zarate IO,
dynamic performance of the unstiffened panel is better than these
Caruso A. Numerical simulation and testing of a composite-stiffened structure
panels. under combined buckling loads. Int J Str Stab Dyn 2010;10(04):871–84. https://
7. The dynamic performance of the panel with R/a = 20, (0◦ /90◦ / [Link]/10.1142/S0219455410003786.
90◦ /0◦ ) laminates and ds/bs = 4 is better than that of the panel [19] Keshav V, Patel SN, Kumar R. Stability and failure study of suddenly loaded
laminated composite cylindrical panel. Int J Appl Mech 2019;11(10):1950093.
with ds/bs = 8. [Link]
8. The dynamic performance of the panel with (0◦ /90◦ /90◦ /0◦ ) [20] Kowal-Michalska K, Mania RJ. Some aspects of dynamic buckling of plates under
laminates, ds/bs = 4, and R/a = 10 is similar to that of the panel in-plane pulse loading. Mech Mech Eng 2008;12(2):135–46.
[21] Kubiak T. (2013). Static and dynamic buckling of thin-walled plate structure.
with ds/bs = 8. Similar results are observed in the case of panel Springer.
with R/a = 5. [22] Kumar A, Panda SK, Kumar R. Buckling behaviour of laminated composite skew
9. In the case off panel having stacking sequence (45◦ /− 45◦ /− 45◦ / plates with various boundary conditions subjected to linearly varying in-plane edge
loading. Int J Mech Sci 2015;100:136–44. [Link]
45◦ ), the performance of the unstiffened panel in dynamic in- ijmecsci.2015.06.018.
plane impulse loading condition is better than the stiffened [23] Less H, Abramovich H. Dynamic buckling of a laminated composite
panel for all the radius of curvatures considered. stringer–stiffened cylindrical panel. Compos B Eng 2012;43(5):2348–58. https://
[Link]/10.1016/[Link].11.070.
10. LCSC panel with (0◦ /90◦ /90◦ /0◦ ) laminates and R/a = 10 shows [24] Liao CL, & Cheng CR. (1994). Dynamic stability of stiffened laminated composite
local buckling behavior in the dynamic buckling analysis when plates and shells subjected to in-plane pulsating forces. Journal of Sound and
the aspect ratio of the stiffener (ds/bs) is 4 and 8. Vibration, 174(3), 335-351. DOI:10.1006/jsvi.1994.1280.
[25] Ljubinković F, Martins JP, Gervásio H, da Silva LS, Leitão C. Experimental and
numerical investigation on cylindrically curved steel panels under uniform
compression. Thin-Walled Struct 2020;149:106527. [Link]
Declaration of Competing Interest tws.2019.106527.
[26] Mittelstedt C. Explicit analysis and design equations for buckling loads and
minimum stiffener requirements of orthotropic and isotropic plates under
The authors declare that they have no known competing financial compressive load braced by longitudinal stiffeners. Thin-Walled Struct 2008;46
interests or personal relationships that could have appeared to influence (12):1409–29. [Link]
the work reported in this paper. [27] Mukhopadhyay M, Mukherjee A. Finite element buckling analysis of stiffened
plates. Comput Struct 1990;34(6):795–803. [Link]
(90)90350-B.
References [28] Mustafa BAJ, Ali R. Prediction of natural frequency of vibration of stiffened
cylindrical shells and orthogonally stiffened curved panels. J Sound Vib 1987;113
(2):317–27. [Link]
[1] Abaqus (2013). Abaqus documentation. (Providence (RI): Dassault Systèmes).
[29] Nagendra S, Gürdal Z, Haftka RT, Starnes Jr JH. Buckling and failure
[2] Abramovich H, & Weller T. (2009). Buckling and postbuckling behavior of
characteristics of compression-loaded stiffened composite panels with a hole.
laminated composite stringer stiffened curved panels under axial compression:
Compos Struct 1994;28(1):1–17. [Link]
Experiments and design guidelines. Journal of Mechanics of Materials and
7.
Structures, 4(7), 1187-1207. DOI:10.2140/jomms.2009.4.1187.
[30] Ovesy HR, & Assaee H. (2002). Buckling analysis of composite stiffened plate
[3] Abramovich H, & Weller T. (2010). Repeated buckling and postbuckling behavior
structures due to interaction of in-plane compression and shear loading, using
of laminated stringer-stiffened composite panels with and without damage.
finite strip method. In Applied Mechanics: Progress and Applications (pp. 533-
International Journal of Structural Stability and Dynamics, 10(04), 807-825. DOI:
538). DOI:10.1142/9789812777973_0086.
10.1142/S0219455410003750.
[31] Patel SN, & Sheikh AH. (2016). Buckling response of laminated composite stiffened
[4] Abramovich H, Grunwald A, Pevsner P, Weller T, David A, Ghilai G, et al.
plates subjected to partial in-plane edge loading. International Journal for
Experiments on axial compression postbuckling behavior of stiffened cylindrical
Computational Methods in Engineering Science and Mechanics, 17(5-6), 322-338.
composite panels. In: 44th AIAA/ASME/ASCE/AHS/ASC Structures, Structural
DOI:10.1080/15502287.2016.1231235.
Dynamics, and Materials Conference; 2003. p. 1793. [Link]
[32] Patel SN, Bisagni C, & Datta PK. (2011). Dynamic buckling analysis of a composite
6.2003-1793.
stiffened cylindrical shell. Structural Engineering and Mechanics, 37(5), 509-527.
[5] Abramovich H, Weller T, Bisagni C. Buckling behavior of composite laminated
DOI:10.12989/sem.2011.37.5.509.
stiffened panels under combined shear-axial compression. J Aircraft 2008;45(2):
[33] Patel SN, Datta PK, Sheikh AH. Buckling and dynamic instability analysis of
402–13. [Link]
stiffened shell panels. Thin-Walled Struct 2006;44(3):321–33. [Link]
[6] Arciniega RA, Goncalves PB, & Reddy JN. (2004). Buckling and postbuckling
10.1016/[Link].2006.03.004.
analysis of laminated cylindrical shells using the third-order shear deformation
[34] Patel SN, Datta PK, Sheikh AH. Dynamic instability analysis of stiffened shell
theory. International Journal of Structural Stability and Dynamics, 4(03), 293-312.
panels subjected to partial edge loading along the edges. Int J Mech Sci 2007;49
DOI:10.1142/S0219455404001240.
(12):1309–24. [Link]
[7] Attaf B, Hollaway L. Vibrational analyses of stiffened and unstiffened composite
[35] Patel SN, Datta PK, & Sheikh AH. (2010). Effect of harmonic in-plane edge loading
plates subjected to in-plane loads. Composites 1990;21(2):117–26. [Link]
on dynamic stability of stiffened shell panels with cutouts. International Journal of
10.1016/0010-4361(90)90003-F.
Applied Mechanics, 2(04), 759-785. DOI:10.1142/S1758825110000743.
[8] Ramezannezhad Azarboni H, Darvizeh M, Darvizeh A, Ansari R. Nonlinear
[36] Anitha Priyadharshani S, Meher Prasad A, Sundaravadivelu R. Analysis of GFRP
dynamic buckling of imperfect rectangular plates with different boundary
stiffened composite plates with rectangular cutout. Compos Struct 2017;169:
conditions subjected to various pulse functions using the Galerkin method. Thin-
42–51. [Link]
Walled Struct 2015;94:577–84. [Link]
[9] Barbre R. (1939). Stability of rectangular plates with longitudinal or transverse
stiffeners under uniform compression. NACA-TM-904.

371
V. Keshav et al. Structures 29 (2021) 360–372

[37] Prusty BG, & Satsangi SK. (2001). Finite element buckling analysis of laminated [48] SudhirSastry YB, Budarapu PR, Madhavi N, Krishna Y. Buckling analysis of thin
composite stiffened shells. International Journal of Crashworthiness, 6(4), 471- wall stiffened composite panels. Comput Mater Sci 2015;96:459–71. [Link]
484. DOI:10.1533/cras.2001.0191. org/10.1016/[Link].2014.06.007.
[38] Ravi Kumar L, Datta PK, & Prabhakara DL. (2005). Vibration and stability behavior [49] Sun L, Harik IE. Buckling of stiffened antisymmetric laminated plates. J Eng Mech
of laminated composite curved panels with cutout under partial in-plane loads. 2013;139(8):1110–23. [Link]
International Journal of Structural Stability and Dynamics, 5(01), 75-94. DOI: [50] Tran KL, Douthe C, Sab K, Dallot J, Davaine L. Buckling of stiffened curved panels
10.1142/S0219455405001507. under uniform axial compression. J Constr Steel Res 2014;103:140–7. [Link]
[39] Reddy JN, Pandey AK. A first-ply failure analysis of composite laminates. Comput org/10.1016/[Link].2014.07.004.
Struct 1987;25(3):371–93. [Link] [51] Viswanathan AV, Soong TC, & Miller Jr RE. (1971). Buckling Analysis for Axially
[40] Romeo G. Experimental investigation on advanced composite-stiffened structures Compressed Flat Plates, Structural Sections, and Stiffened Plates Reinforced with
under uniaxial compression and bending. AIAA J 1986;24(11):1823–30. https:// Laminated Composites. ASA-CR-1887.
[Link]/10.2514/3.9530. [52] Volʹmir AS. (1974). The nonlinear dynamics of plates and shells (No. FTD-HC-23-
[41] Sapountzakis EJ, & Katsikadelis JT. (2002). Influence of the Inplane Boundary 851-74). Foreign Technology Division Wright-Patterson AFB Ohio.
Conditions on the Vibration Frequencies and Buckling Load of Ribbed Plates. [53] Wang D, & Abdalla MM. (2015). Global and local buckling analysis of grid-
International Journal of Structural Stability and Dynamics, 2(01), 25-43. DOI: stiffened composite panels. Composite Structures, 119, 767-776. DOI:10.1016/j.
10.1142/S0219455402000385. compstruct.2014.09.050.
[42] Singer J. (1967). The influence of stiffener geometry and spacing on the buckling of [54] Williams JG, Stein M. Buckling behavior and structural efficiency of open-section
axially compressed cylindrical and concial shells (No. SR-3). Technion-Israel Inst of stiffened composite compression panels. AIAA J 1976;14(11):1618–26. https://
Tech Haifa Dept. of Aeronautical Engineering. [Link]/10.2514/3.61497.
[43] Sinha G, Mukhopadhyay M. Finite element free vibration analysis of stiffened [55] Xu X, Chu H, & Lim CW. (2008). Hamiltonian system for dynamic buckling of
shells. J Sound Vib 1994;171(4):529–48. [Link] transversely isotropic cylindrical shells subjected to an axial impact. International
[44] Sivasubramonian B, Rao GV, Krishnan A. Free vibration of longitudinally stiffened Journal of Structural Stability and Dynamics, 8(03), 487-504. DOI:10.1142/
curved panels with cutout. J Sound Vib 1999;226(1):41–55. [Link] S0219455408002764.
10.1006/jsvi.1999.2281. [56] Yaffe R, Abramovich H. Dynamic buckling of cylindrical stringer stiffened shells.
[45] Srivastava AKL, Datta PK, & Sheikh AH. (2003a). Dynamic instability of stiffened Comput Struct 2003;81(8-11):1031–9. [Link]
plates with cutout subjected to in-plane uniform edge loadings. International 00417-0.
Journal of Structural Stability and Dynamics, 3(03), 391-403. DOI:10.1142/ [57] Zhang T, Liu TG, Zhao Y, Luo JZ. Nonlinear dynamic buckling of stiffened plates
S0219455403000963. under in-plane impact load. J Zheijang Univ-Sci 2004;5(5):609–17. [Link]
[46] Srivastava AKL, Datta PK, Sheikh AH. Dynamic instability of stiffened plates org/10.1631/jzus.2004.0609.
subjected to non-uniform harmonic in-plane edge loading. J Sound Vib 2003;262 [58] Zimmermann R, Klein H, Kling A. Buckling and postbuckling of stringer stiffened
(5):1171–89. [Link] fibre composite curved panels – Tests and computations. Compos Struct 2006;73
[47] Starnes Jr JH, Knight Jr NF, Rouse M. Postbuckling behavior of selected flat (2):150–61. [Link]
stiffened graphite-epoxy panels loaded in compression. AIAA J 1985;23(8): [59] Petry D, Fahlbusch G. Dynamic buckling of thin isotropic plates subjected to in-
1236–46. [Link] plane impact. Thin Wall Struct 2000;38(3):267–83. [Link]
S0263-8231(00)00037-9.

372