2.081J/16.

230J Plates and Shells

Professor Tomasz Wierzbicki

Contents

1 Strain-Displacement Relation for Plates 1

1.1 1-D Strain Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1.1 Engineering Strain . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1.2 Green-Lagrangian Strain . . . . . . . . . . . . . . . . . . . . . 1

1.2 3-D Strain Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2.1 Derivation of Green-Lagrangian Strain Tensor for Plates . . . 1

1.2.2 Specification of Strain-Displacement Relation for Plates . . . 5

2 Derivation of Constitutive Equations for Plates 10

2.1 Definitions of Bending Moment and Axial Force . . . . . . . . . . . . 10

2.2 Bending Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.2.1 Bending Moment . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.2.2 Bending Energy Density . . . . . . . . . . . . . . . . . . . . . 11

2.2.3 Total Bending Energy . . . . . . . . . . . . . . . . . . . . . . 13

2.3 Membrane Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.3.1 Axial Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.3.2 Membrane Energy Density . . . . . . . . . . . . . . . . . . . 14

2.3.3 Total Membrane Energy . . . . . . . . . . . . . . . . . . . . . 16

3 Development of Equation of Equilibrium and Boundary Conditions

Using Variational Approach 17

3.1 Bending Theory of Plates . . . . . . . . . . . . . . . . . . . . . . . . 17

3.1.1 Total Potential Energy . . . . . . . . . . . . . . . . . . . . . . 17

3.1.2 First Variation of the Total Potential Energy . . . . . . . . . 20

3.1.3 Equilibrium Equation and Boundary Conditions . . . . . . . 24

3.1.4 Specification of Equation for Rectangular Plate . . . . . . . . 24

3.2 Bending-Membrane Theory of Plates . . . . . . . . . . . . . . . . . . 29

3.2.1 Total Potential Energy . . . . . . . . . . . . . . . . . . . . . . 29

i
 3.2.2 First Variation of the Total Potential Energy . . . . . . . . . 29

3.2.3 Equilibrium Equation and Boundary Conditions . . . . . . . 31

4 General Theories of Plate 34

4.1 Bending Theory of Plates . . . . . . . . . . . . . . . . . . . . . . . . 34

4.1.1 Derivation of the Plate Bending Equation . . . . . . . . . . . 34

4.1.2 Reduction to a System of Two Second Order Equations . . . 35

4.1.3 Exercise 1: Plate Solution . . . . . . . . . . . . . . . . . . . . 36

4.1.4 Exercise 2: Comparison between Plate and Beam Solution . . 40

4.1.5 Exercise 3: Finite Difference Solution of the Plate Bending

Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4.2 Membrane Theory of Plates . . . . . . . . . . . . . . . . . . . . . . . 47

4.2.1 Plate Membrane Equation . . . . . . . . . . . . . . . . . . . . 47

4.2.2 Plate Equation for the Circular Membrane . . . . . . . . . . 48

4.2.3 Example: Approximation Solution for the Clamped Membrane 48

4.3 Buckling Theory of Plates . . . . . . . . . . . . . . . . . . . . . . . . 51

4.3.1 General Equation of Plate Buckling . . . . . . . . . . . . . . 51

4.3.2 Linearized Buckling Equation of Rectangular Plates . . . . . 52

4.3.3 Analysis of Rectangular Plates Buckling . . . . . . . . . . . . 54

4.3.4 Derivation of Raleigh-Ritz Quotient . . . . . . . . . . . . . . 62

4.3.5 Ultimate Strength of Plates . . . . . . . . . . . . . . . . . . . 65

4.3.6 Plastic Buckling of Plates . . . . . . . . . . . . . . . . . . . . 71

4.3.7 Exercise 1: Effect of In-Plane Boundary Conditions, δw = 0 . 74

4.3.8 Exercise 2: Raleigh-Ritz Quotient for Simply Supported Square

Plate under Uniaxial Loading . . . . . . . . . . . . . . . . . . 76

4.4 Buckling of Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

4.4.1 Transition from Global and Local Buckling . . . . . . . . . . 77

4.4.2 Local Buckling . . . . . . . . . . . . . . . . . . . . . . . . . . 80

5 Buckling of Cylindrical Shells 84

5.1 Governing Equation for Buckling of Cylindrical Shells . . . . . . . . 84

5.1.1 Special Case I: Cylinder under Axial Load P , q = 0 . . . . . 86

5.1.2 Special Case II: Cylinder under Lateral Pressure . . . . . . . 87

5.1.3 Special Case III: Hydrostatic Pressure . . . . . . . . . . . . . 88

5.1.4 Special Case IV: Torsion of a Cylinder . . . . . . . . . . . . . 89

5.2 Derivation of the Linearized Buckling Equation . . . . . . . . . . . . 89

5.3 Buckling under Axial Compression . . . . . . . . . . . . . . . . . . . 90

5.3.1 Formulation for Buckling Stress and Buckling Mode . . . . . 90

5.3.2 Buckling Coefficient and Batdorf Parameter . . . . . . . . . . 93

5.4 Buckling under Lateral Pressure . . . . . . . . . . . . . . . . . . . . 96

5.5 Buckling under Hydrostatic Pressure . . . . . . . . . . . . . . . . . . 99

5.6 Buckling under Torsion . . . . . . . . . . . . . . . . . . . . . . . . . 100

5.7 Influence of Imperfection and Comparison with Experiments . . . . . 102

ii
1 Strain-Displacement Relation for Plates
1.1 1-D Strain Measure
1.1.1 Engineering Strain
Engineering strain ε is defined as the relative displacement:
ds − ds0
ε= (1)
ds0
where ds0 is the increment of initial lenght and ds is the increment of current length.
1.1.2 Green-Lagrangian Strain
Instead of comparing the length, one can compare the square of lengths:

ds2 − ds20
E= (2)
2ds20
ds − ds0 ds + ds0
=
ds0 2ds0
Where ds → ds0 , the second term is Eq. (2) tends to unity, and the Green strain
measure and the engineering strain become identical. Equation (2) can be put into
an equivalnet form:
ds2 − ds20 = 2Eds20 (3)
which will now be generalized to the 3-D case.
1.2 3-D Strain Measure
1.2.1 Derivation of Green-Lagrangian Strain Tensor for Plates
Let define the following quanties:

• a = [ai ]: vector of the initial (material) coordinate system

• x = [xi ]: vector of the current (spatial) coordinate system

• u = [ui ]: displacement vector

where the index i = 1, 2, 3. The relation between those quantities is:

xi = ai + ui (4)
dxi = dai + dui

 du
da dx

x
a

Now, the squares of the initial and the current length increment can be written
in terms of ai and ui :
ds20 = dai daj δ ij (5)

ds2 = dxi dxj δ ij (6)

= (dai + dui ) (daj + duj ) δ ij

where the Kronecker tensor δ ij reads:

¯ ¯
¯ 1 0 0 ¯

¯ ¯
δ ij =
¯¯ 0 1 0 ¯¯
(7)

¯
0 0 1 ¯

The vector u can be considered as a function of:

• the initial (material) coordinate system, u (a), which leads to Lagrangian

descripion, or

• the current (spatial) coordinates, u (x), which leads to the Eulerian descrip
tion

In structural mechanics, the Lagrangian description is preferable:

ui = ui (ai ) (8)
∂ui
dui = dak = ui ,k dak
∂ak
∂uj
duj = dal = uj ,l dal
∂al
Let us calculated the difference in the length square:

ds2 − ds20 = (dai + dui ) (daj + duj ) δ ij − dai daj δ ij (9)

2
Using Eq. (8) and the definition of δ ij , the difference in the length square can be
transformed into:

ds2 − ds20 = (duj dai + dui daj + dui duj ) δ ij (10)

= (uj ,l dal dai + ui ,k dak daj + ui ,k dak uj ,l dal ) δ ij
= [uj ,l (daj δ jl ) dai + ui ,k (dai δ ik ) daj + ui ,k uj ,l (dai δ ik ) (daj δ jl )] δ ij
= (uj ,i +ui ,j +ui ,k uj ,k ) dai daj
= 2Eij dai daj

where, by analogy with the 1-D case, the Lagrangian or Green strain tensor Eij is
defined:
1
Eij = (ui ,j +uj ,i +uk ,i uk ,j ) (11)
2
In the case of small displacement gradient (uk ,i ¿ 1), the second nonlinear term
can be neglected leading to the defintion of the infinitesimal strain tensor:
1
εij = (ui ,j +uj ,i ) (12)
2
From the defintion, the strain tensor is symmetric εij = εji , which can be seen by
intechanign the indices i for j and j for i. In the moderately large deflection theory
of structures, the nonlinear terms are important. Therefore, Eq. (11) will be used
as a starting point in the development of the general theory of plates.
Components of Green-Lagrangian Strain Tensor Let define the following
range convention for indices:

• Greek letters: α, β, ... = 1, 2

• Roman letters: i, j, ... = 1, 2, 3

With this range convention, the Roman letters are also written as:

i = α, 3 (13)
j = β, 3

The Lagrangian or Green strain tensor can be expressed:

E11 E12 # E13 #

# Eαβ # E3β
Eij = E21 E22 # E23 = #
" " " # " " " " # "
E31 E32 # E33 Eα 3 # E33

where Eαβ is the in-plane component of strain tensor, Eα3 and E3β are out-of-plane

shear components of strain tensor, and E33 is the through-thickness component of

strain tensor. Similarily, displacement vector can be divided into two components:

u1 u
uα
ui = u2 = v =
" " "
u3 w w

where uα is the in-plane components of the displacement vector, and u3 = w is the

out-of-plane components of the displacement vector and also called as the trans
verse displacement.

Initial Undeformed Configuration

a uα x

z
u u3
x

middle
surface

ξ Deformed Configuration

Assumptions of the von Karman Theory The von Karman thoery of mod
erately large deflection of plates assumes:

1. The plate is thin. The thickness h is much smaller than the typical plate
dimension, h ¿ L.

2. The magnitude of the transverse deflection is of the same order as the thickness
of plate, |w| = O (h). In practice, the present theory is still a good engineering
approximation for deflections up to ten plate thickness.

 3. Gradients of in-plane displacements uα ,β are small so that their product or

square can be neglected.

4. Love-Kirchhoff hypothesis is satisfied. In-plane displacements are a linear

function of the z−coordiate (3-coordinate).

uα = u◦α − z u3 ,α (14)

where u◦α is the displacement of the middle surface, which is independent of

z−coordinate, i.e. u◦α ,3 = 0; and u3 ,α is the slope which is negative for the
"smiling" beam.

5. The out-of-plane displacement is independent of the z−coordiante, i.e. u3 ,3 =

0.

1.2.2 Specification of Strain-Displacement Relation for Plates

In the theory of moderately large deflections, the strain-displacement relation can
be specified for plates.
In-Plane Terms of the Strain Tensors From the general expression, Eq. (11),
the 2-D in-plane componets of the strain tensor are:
1
Eαβ = (uα ,β +uβ ,α +uk ,α uk ,β ) (15)
2
Here, consider the last, nonlinear term:

uk ,α uk ,β = u1 ,α u1 ,β +u2 ,α u2 ,β +u3 ,α u3 ,β (16)

= uγ ,α uγ ,β +u3 ,α u3 ,β

In the view of the Assumption 3, the first term in the above equation is zero,
uγ ,α uγ ,β ' 0. Therefore, the 2-D in-plane components of strain tensor reads:

1
Eαβ = (uα ,β +uβ ,α +w,α w,β ) (17)
2

5
where w = u3 . Introducing Eq. (14) into Eq. (17), i.e. applying Love-Kirchhoff
hypothesis, one gets:
1
£
¡ ¢
¤

Eαβ = (u◦α − z w,α ) ,β + u◦β − z w,β ,α +w,α w,β (18)

2
1
¡ ¢

=
u◦α ,β +u◦β ,α −2 z w,αβ +w,α w,β
2

1
¡ ¢
1

=
u◦α ,β +u◦β ,α − z w,αβ + w,α w,β
2
2

From the definiton of the curvature, one gets:

καβ = −w,αβ (19)
Now, Eq. (18) can be re-casted in the form:
◦
Eαβ = Eαβ + z καβ (20)
where the strain tensor of the middle surface Eαβ◦ is composed of a linear and a

nonlinear term:
◦ 1
¡ ¢
1
Eαβ =
u◦α ,β +u◦β ,α + w,α w,β (21)
2
2

In the limiting case of small displacements, the second term can be neglected as
compared to the first term. In the classical bending theory of plate, the in-plane
displacements are assumed to be zero uα = 0 so that strains are only due to the
curvatue:
Eαβ = z καβ (22)
where ¯ ¯
¯ ∂2w ∂2w ¯
¯ ¯ ¯ ¯
¯ κ11 κ12 ¯ ¯ ∂x2 ∂x∂y
¯
καβ ¯
=
¯
¯ =
−
¯ ¯ =
−w,αβ (23)

κ21 κ22 ¯
¯
¯ ∂2w ∂ 2 w
¯
¯
¯
∂x∂y ∂y2 ¯

In the above equation, κ11 and κ22 are curvatures of the cylindrical bending, and
κ12 is the twist which tells how the slope in the x−direction changes with the
y−direction: µ
¶

∂ ∂w

κ12 =
∂y ∂x

for a cylinder

κ12 = 0

 ³ ´
1
Interpretation of the linear terms: 2 u◦α ,β +u◦β ,α Each component can
be expressed in the followings:

1 du1
ε11 = (u1 ,1 +u1 ,1 ) = u1 ,1 = (24)
2 dx
1 du2
ε22 = (u2 ,2 +u2 ,2 ) = u2 ,2 = (25)
2 dy
µ ¶
1 1 du1 du2
ε12 = (u1 ,2 +u2 ,1 ) = + (26)
2 2 dy dx
1 du1

ε12 |if u2 =0 =
2 dy

ε11 u2 ≡ 0

ε12
x x x
y y y

ε22 u1

Therefore, ε11 and ε22 are the tensile strain in the two directions, and ε12 is the
change of angles, i.e. shear strain.
Interpretation of the nonlinear term: 12 w,α w,β Let α = 1 and β = 1.
Then, the nonlienar term reads:
¯ µ ¶
1 ¯
¯ 1 dw dw 1 dw 2
w,α w,β ¯ = = (27)
2 α=1,β=1 2 dx dx 2 dx

One can also obtain the same quantity by the defintion of 1-D Green-Lagrangian
strain:
¡ 2 ¢ µ ¶ µ ¶
ds2 − ds20 ds0 + dw2 − ds20 1 dw 2 1 dw 2
E= ' = = (28)
2ds20 2ds20 2 ds0 2 dx

 dx
x
ds0 = dx
ds0 ds 2 = ds02 + dw2
dw
z, w ds

Thus, the conclusion is that the nonlinear term 12 w,α w,β represents the change of
length of the plate element due to finite rotations.
Out-Of-Plane Terms of the Strain Tensors Refering to the definition intro
duced in Section 1.2.1, there are three other componets of the strain tensor: E3β ,
Eα3 and E33 . Using the general expression for the components of the strain tensor,
Eq. (11), it can be shown that the application of Assumption 4 and 5 lead to the
following expressions:
1
E3β = (u3 ,β +uβ ,3 +uk ,3 uk ,β ) (29)
2

= [u3 ,β +uβ ,3 + (u1 ,3 u1 ,β +u2 ,3 u2 ,β +u3 ,3 u3 ,β )]

2

= [u3 ,β −u3 ,β + (−u3 ,1 u1 ,β −u3 ,2 u2 ,β )]

2

= (−u3 ,1 u1 ,β −u3 ,2 u2 ,β )
2

= − w,γ uγ ,β
2

1
Eα3 = (uα ,3 +u3 ,α +uk ,α uk ,3 ) (30)
2

= [uα ,3 +u3 ,α + (u1 ,α u1 ,3 +u2 ,α u2 ,3 +u3 ,α u3 ,3 )]

2

= [−u3 ,α +u3 ,α + (−u1 ,α u3 ,1 −u2 ,α u3 ,2 )]

2

= (−u1 ,α u3 ,1 −u2 ,α u3 ,2 )
2

= − w,γ uγ ,α
2

 1
E33 = (u3 ,3 +u3 ,3 +uk ,3 uk ,3 ) (31)
2
1h i
= u3 ,3 + (u1 ,3 )2 + (u2 ,3 )2 + (u3 ,3 )2
2
1h i
= (u1 ,3 )2 + (u2 ,3 )2
2
1h i
= (−u3 ,1 )2 + (−u3 ,2 )2
2
1
= w,γ w,γ
2
The above are all second order terms which vanish for small deflection theory of
plates. In the theory of moderately larege deflection of plates, the out-of-plate shear
strains as well as the through-thickness strain is not zero. Therefore, an assumption
"plane remains plane," expressed by Eq. (14), does not mean that "normal remains
normal." The existance of the out-of-plane shear strain means that lines originally
normal to the middle surface do not remain normal to the deformed plate. However,
the incremental work of these strains with the corresponding stresses is negligible:

E3β σ 3β , Eα3 σ α3 and E33 σ 33 , are small (32)

because the corresponding stress σ 3β , σ α3 and σ 33 are small as compared to the

in-plane stress σ αβ . One can conclude that the elastic strain energy (and even
plastic dissipation) is well approximated using the plane strain assumption:
Z Z
1 1
σ ij εij dz ' σ αβ εαβ dz (33)
h 2 h 2

2 Derivation of Constitutive Equations for Plates

2.1 Definitions of Bending Moment and Axial Force
Hook’s law in plane stress reads:
E
σ αβ = [(1 − ν) εαβ + ν εγγ δ αβ ] (34)
1 − ν2
In terms of components:
E
σ xx = (εxx + ν εyy ) (35)
1 − ν2
E
σ yy = (εyy + ν εxx )
1 − ν2
E
σ xy = εxy
1+ν
Here, strain tensor can be obtained from the strain-displacement relations:

εαβ = ε◦αβ + z καβ (36)

Now, define the tensor of bending moment:

Z
h
2
Mαβ ≡ σ αβ z dz (37)
− h2

and the tensor of axial force (membrane force):

Z
h
2
Nαβ ≡ σ αβ dz (38)
− h2

2.2 Bending Energy

2.2.1 Bending Moment
Let us assume that ε◦αβ = 0. The bending moment Mαβ can be calculated:
Z h
E 2
Mαβ = [(1 − ν ) εαβ + ν εγγ δ αβ ] z dz
(39)

1 − ν2 − h2
Z
h
E £ ¤ 2
= 2
(1 − ν) ε◦αβ + ν ε◦γγ δ αβ z dz

1−ν −h
2
Z h
E
2
+ [(1 − ν) καβ + ν κγγ δ αβ ] z
2 dz
1 − ν
2 −h
2

Eh3
= [(1 − ν) καβ + ν κγγ δ αβ ]
12 (1 − ν 2 )

10
Here, we define the bending rigidity of a plate D as follows:
Eh3
D= (40)
12 (1 − ν 2 )
Now, one gets the moment-curvature relations:

Mαβ = D [(1 − ν) καβ + ν κγγ δ αβ ] (41)

¯ ¯
¯ M11 M12 ¯
Mαβ ¯
= ¯
¯ (42)

M21 M22 ¯

where M12 = M21 due to symmetry.

M11 = D (κ11 + ν κ22 ) (43)
M22 = D (κ22 + ν κ11 )
M12 = D (1 − ν) κ12
2.2.2 Bending Energy Density
One -Dimensional Case Here, we use the hat notation for a function of certain
argument such as:
M11 = M̂11 (κ11 ) (44)
= D κ11
Then, the bending energy density Ūb reads :
Z
κ̄11
Ūb = M̂11 (κ11 ) dκ11 (45)
0
Z
κ̄11
=D κ11 dκ11
0
1
= D (κ̄11 )2
2

Ūb = M11 κ̄11 (46)

2

M11

D
κ11
dκ11

11

General Case General definition of the bending energy density reads:

I
Ūb = Mαβ dκαβ (47)

κ
κ11

κ22
0

Calculate the energy density stored when the curvature reaches a given value κ̄αβ .
Consider a straight loading path:

καβ = η κ̄αβ (48)

dκαβ = κ̄αβ dη

Mαβ

Mαβ η=1

η=0
καβ
καβ

Mαβ = M̂αβ (καβ ) (49)

= M̂αβ (η κ̄αβ )
= η M̂αβ (κ̄αβ )

12

where M̂αβ (καβ ) is a homogeneous function of degree one.

I
Ūb = M̂αβ (καβ ) dκαβ (50)
Z 1
= η M̂αβ (κ̄αβ ) κ̄αβ dη
0
Z 1
= M̂αβ (κ̄αβ ) κ̄αβ η dη
0
1
= M̂αβ (κ̄αβ ) κ̄αβ
2
1
= Mαβ κ̄αβ
2
Now, the bending energy density reads:
D
Ūb = [(1 − ν) κ̄αβ + ν κ̄γγ δ αβ ] κ̄αβ (51)
2
D
= [(1 − ν) κ̄αβ κ̄αβ + ν κ̄γγ κ̄αβ δ αβ ]
2
D h i
= (1 − ν) κ̄αβ κ̄αβ + ν (κ̄γγ )2
2
The bending energy density expressed in terms of components:
D n h i o
Ūb = (1 − ν) (κ̄11 )2 + 2 (κ̄12 )2 + (κ̄22 )2 + ν (κ̄11 + κ̄22 )2 (52)
2
D n h i o
= (1 − ν) (κ̄11 + κ̄22 )2 − 2 κ̄11 κ̄22 + 2 (κ̄12 )2 + ν (κ̄11 + κ̄22 )2
2
D nh i h io
= (κ̄11 + κ̄22 )2 − 2 κ̄11 κ̄22 + 2 (κ̄12 )2 − ν −2 κ̄11 κ̄22 + 2 (κ̄12 )2
2
D n h io
= (κ̄11 + κ̄22 )2 − 2 κ̄11 κ̄22 + 2 (κ̄12 )2 − ν −2 κ̄11 κ̄22 + 2 (κ̄12 )2
2
D n h io
= (κ̄11 + κ̄22 )2 + 2 (1 − ν) −κ̄11 κ̄22 + (κ̄12 )2
2

n h io
¯b = D (κ̄11 + κ̄22 )2 − 2 (1 − ν) κ̄11 κ̄22 − (κ̄12 )2
U (53)
2

2.2.3 Total Bending Energy

The total bending energy is the integral of the bending energy density over the area
of plate: Z
Ub = Ūb dA (54)
S

13
2.3 Membrane Energy
2.3.1 Axial Force
Assume that καβ = 0. The axial force can be calculated:
Z h
E
2
N
αβ =
[(1 − ν) εαβ + ν εγγ δ αβ ] dz (55)

1 − ν2 − h2
Z
h
E
2 £ ¤

=
(1 − ν) ε◦αβ + ν ε◦γγ δ αβ dz

1 − ν2 − h2
Z h
E
2
+ [(1 − ν ) καβ + ν κγγ δ αβ ] z dz
1 − ν2 − h2
Z h
E
£
◦ ◦
¤
2
=
(1 − ν) ε
αβ + ν εγγ δ αβ dz

1 − ν2 − h2
Z h
E
2
+ 2
[(1 − ν ) καβ + ν κγγ δ αβ ] z dz

1−ν − h2
Eh
£ ¤
=
2
(1 − ν ) ε◦αβ + ν ε◦γγ δ αβ
1−ν
Here, we define the axial rigidity of a plate C as follows:
Eh
C= (56)
1 − ν2
Now, one gets the membrane force-extension relation:

h i
Nαβ = C (1 − ν) ε◦αβ + ν ε◦γγ δ αβ (57)

¯ ¯
¯ N11 N12 ¯
Nαβ ¯
= ¯
¯ (58)

N21 N22 ¯

where N12 = N21 due to symmetry.

N11 = C (ε◦11 + ν ε◦22 ) (59)

N22 = C (ε◦22
+ ν ε◦11 )
N12 = C (1 − ν) ε◦11

2.3.2 Membrane Energy Density

Using the similar definition used in the calculation of the bending energy density,
the extension energy (membrane energy) reads:
I

Ūm = Nαβ dε◦αβ (60)

14
Calculate the energy stored when the extension reaches a given value ε̄◦αβ . Consider
a straight loading path:

ε◦αβ = η ε̄◦αβ (61)

dε◦αβ = ¯ε◦αβ dη

¡ ¢
Nαβ = N̂αβ ε◦αβ (62)
¡ ¢
= N̂αβ η ¯ε◦αβ
¡ ¢
= η N̂αβ ε̄◦αβ
³ ´
where N̂αβ ε◦αβ is a homogeneous function of degree one.
Z ε̄◦αβ ¡ ¢
Ūm = N̂αβ ε◦αβ dε◦αβ (63)
0
Z 1 ¡ ¢
= η N̂αβ ε̄◦αβ ε̄◦αβ dη
0
1 ˆ ¡ ◦ ¢ ◦
= N αβ ε̄αβ ε̄αβ
2
1
= Nαβ ε̄◦αβ
2
Now, the extension energy reads:
£ ¤
¯m = C (1 − ν) ε̄◦ + ν ε̄◦γγ δ αβ ε̄◦
U (64)

αβ αβ
2

C h ¡ ¢ i

2
= (1 − ν) ε̄◦αβ ε̄◦αβ + ν ε̄◦γγ
2
The extension energy expressed in terms of components:
C n h i o
Ūm = (1 − ν) (¯ε◦11 )2 + 2 (ε̄◦12 )2 + (ε̄◦22 )2 + ν (ε̄◦11 + ¯ε◦22 )2 (65)
2
C n h i o
= (1 − ν) (¯ε◦11 + ¯ε◦22 )2 − 2 ¯ε◦11¯ε◦22 + 2 (ε¯◦12 )2 + ν (¯ε◦11 + ¯ε◦22 )2
2
C n h io
= (¯ε◦11 + ¯ε◦22 )2 − 2 ¯ε◦11¯ε◦22 + 2 (ε¯◦12 )2 − ν −2 ¯ε◦11¯ε◦22 + 2 (ε¯◦12 )2
2
C n h io
= (¯ε◦11 + ¯ε◦22 )2 + 2 (1 − ν) −¯ε◦11¯ε◦22 + (ε¯◦12 )2
2

n h io
¯m = C (¯ε◦ + ¯ε◦ )2 − 2 (1 − ν) ¯ε◦ ¯ε◦ − (¯ε◦ )2
U (66)
11 22 11 22 12
2

15

2.3.3 Total Membrane Energy

The total membrane is the integral of the membrane energy density over the area
of plate:: Z
Um = Ūm dS (67)
S

16

3 Development of Equation of Equilibrium and Boundary Condi

tions Using Variational Approach
3.1 Bending Theory of Plates
3.1.1 Total Potential Energy
The total potential energy of the system Π reads:

Π = Ub − Vb (68)

where Ub is the bending energy stored in the plate, and Vb is the work of external
forces.
Bending Energy
Z
1
Ub = Mαβ καβ dS (69)
2 SZ
1
=− Mαβ w,αβ dS
2 S

where the geometrical relation καβ = −w,αβ has been used.

Work of External Forces
Plate Loading Lateral load:

q (x) = q (xα ) (70)

This is distributed load measured in [N/m2 ] or [lb/in2 ] force per unit area of the
middle surface of the plate.

q ( x)

The distributed load contains concentrated load P as a special case:

P (x0 , y0 ) = P0 δ (x − x0 ) δ (y − y0 ) (71)

where δ is the Dirac delta function, [x0, y0 ] is the coordinate of the application of
the concentrated force, and P0 is the load intensity.

17

 Dirac δ -function

x
x0

NOTE The shearing loads on the lateral surface of ice are normally not consid
ered in the theory of thin plates.

Load Classification

• Load applied at the horizontal surfaces.

transverse load

• Load applied at the lateral surfaces.

edge force

edge moment

Loads are assumed to be applied to the middle plane of the plate

NOTE Other type of loading such as shear or in-plane tension or compression
do not deflect laterally the plate and therefore are not considered in the bending
theory.

18

 in-plane tension

or compression

in-plane shear

Potential Energy due to Lateral Load q Lateral (transverse) load does

work on transverse deflection: Z
q w dS (72)
S
This is also called a work of external forces.
Potential Energy due to Edge Moment The conjugate kinematic variable
associated with the edge moment is the edge rotation dw/dxn .

dl

edge moment n
Γ

We apply only the normal bending moment Mnn :

Z
− M ¯ nn dw dl (73)
Γ dxn
where the minus sign is included because positive bending moment results in a
negative rotation and negative moment produces positive rotation.

Mnn < 0 Mnn > 0

xn
0 a
dw dw
>0 <0
dxn dxn

At the edge, Mtt = 0 and Mtn = 0.

19
 Potential Energy due to Edge Forces
Z
V̄n w dl (74)
Γ

Q1 Q2

x2 x1
Vn

Potential Energy due to All External Forces Now, the work of external
forces reads: Z Z Z
dw
Vb = q w dS − M̄nn dl + V̄n w dl (75)
S Γ dxn Γ
3.1.2 First Variation of the Total Potential Energy
The total potential energy reads:
Z
1
Π=− Mαβ w,αβ dS (76)
2 S
Z Z Z
¯ dw
− q w dS + Mnn dl − V¯n w dl
S Γ dxn Γ

First variation of the total potential energy δΠ is expressed:

Z
δΠ = − Mαβ δw,αβ dS (77)
ZS Z µ ¶ Z
¯ dw
− q δw dS + Mnn δ dl − V¯n δw dl
S Γ dxn Γ

We shall transform now the first integral with the help of the Gauss theorem.
First note that from the rule of the product differentiation:

Mαβ δw,αβ = (Mαβ δw,α ) ,β −Mαβ ,β δw,α (78)

then Z Z Z
Mαβ δw,αβ dS = (Mαβ δw,α ) ,β dS − Mαβ ,β δw,α dS (79)
S S S
Now, the first integral on the right hand side of the above equation transforms to
the line integral:
Z Z Z
Mαβ δw,αβ dS = Mαβ δw,α nβ dl − Mαβ ,β δw,α dS (80)
S Γ S

20
The integrand of the second integral on the right hand side of the above equation
transform to:
Mαβ ,β δw,α = (Mαβ ,β δw) ,α −Mαβ ,αβ δw (81)
which results in:
Z Z
Mαβ δw,αβ dS = Mαβ δw,α nβ dl (82)
S Γ
Z Z
− (Mαβ ,β δw) ,α dS + Mαβ ,αβ δw dS
S S
upon which the application of the Gauss rule gives:
Z Z
Mαβ δw,αβ dS = Mαβ δw,α nβ dl (83)
S ΓZ Z
− Mαβ ,β δw nα dl + Mαβ ,αβ δw dS
Γ S
We can return now to the expression for δΠ and substitute there the transformed
first integral:
Z
δΠ = (−Mαβ ,αβ −q) δw dS (84)
SZ Z
+ Mαβ ,β δw nα dl − V¯n δw dl
ZΓ ZΓ
− Mαβ δw,α nβ dl + M̄nn δw,n dl
Γ Γ
³ ´
dw
where δw,n = δ dxn
. It is seen that integrals involving the prescribed forces M̄nn
and V̄n are written in a local coordinate system xγ {xn , xt } while the remaining
two integrals over the contour Γ are written in the global coordinate system xα . In
order to make comparison, we have to decide on one coordinate system. We choose
the local system.
Consider the first integral:
Z
(Mαβ ,β nα ) δw dl (85)
Γ
The term in the parenthesis is a scalar quantity and thus remain unchanged with
respect to the rotation of coordinate system. In the local system xγ , the line
integral becomes: Z
(Mγδ ,δ nγ ) δw dl (86)
Γ
where γ = 1 is the normal direction n, and γ = 2 is the tangential direction t. The
coordinates of the unit normal vector is the local system are nγ {1, 0}. Hence,
Z Z
(Mγδ ,δ nγ ) δw dl = (M1δ ,δ n1 + M2δ ,δ n2 ) δw dl (87)
Γ ZΓ

= M1δ ,δ δw dl
Γ

21

Furthermore, the integrand reads:

M1δ ,δ = M11 ,1 +M12 ,2 (88)

∂M11 ∂M12 ∂Mnn ∂Mnt
= + = +
∂x1 ∂x2 ∂xn ∂xt
and we call it the shear force in the normal direction n and denote:

Qn ≡ Mnδ ,δ δ = {1, 0} or {n, t} (89)

Now, we can combine two line integrals in the equation of first variation of the total
potential energy: Z
¡ ¢
Qn − V̄n δw dl (90)
Γ
How the remaining integral is transformed?
Z Z
(Mαβ nβ ) δw,α dl = (Mγδ nδ ) δw,γ dl (91)
Γ Γ

Because it is a scalar quantity, we simply switch indices from global system (α and
β) to local (γ and δ). As before nδ {1, 0} so after summing with respect to δ, we
have:
Z Z
(Mγ1 n1 + Mγ2 n2 ) δw,γ dl = Mγ1 δw,γ dl (92)
Γ
ZΓ
= Mγn δw,γ dl
ZΓ
= (Mnn δw,n +Mtn δw,t ) dl
Γ

The first term can be absorbed with the line integral representing potential energy
of bending moment: Z
¡ ¢
− Mnn − M ¯ nn δw,n dl (93)
Γ
There remains though one integral which does not fit to anything. Since the
boundary term must be equilibrated, it is suspected that this term might belong to
the shearing force term, at least partially:
Z
Mtn δw,t dl transverse term (94)
Γ

In order to compare this term with the shearing force term, we have to make this
term comparable as far as the kinematic quantity describing variation is concerned.
One integral involves δw and the other one δw,t . Note that ∂w,t = ∂ (δw) /∂xt is
the derivative of the function δw in the tangential direction, i.e. direction along the
curve Γ. This means that we can integrate by parts along Γ. Thus,

Mtn δw,t = (Mtn δw) ,t −Mtn ,t δw (95)

22
 Z Z Z
Mtn δw,t dl = (Mtn δw) ,t dl − Mtn ,t δw dl (96)
Γ Γ Γ
The first term in the right hand is equal to the value of the integrand calculated at
the beginning and end of the integration path:
Z
(Mtn δw) ,t dl = Mtn δw|end
beginning (97)
Γ

Consider now two cases.

• The contour Γ is a smooth closed curve, so the value at the beginning is equal
to the value at the end:

Mtn δw|end − Mtn δw|beginning = 0 (98)

The term does not give any contribution.

direction of
integration

end
Γ
begining

• The contour Γ is piece-wise linear or composed of a finite number, k, of smooth

curves with discontinuity. Therefore, the integration should be made in a
piece-wise manner. Thus, the continuation of the beginning and end of each
should be added: X
Mtn δw|end
beginning (99)
k

23

3.1.3 Equilibrium Equation and Boundary Conditions

Now, we can write the final expression for the first variation of δΠ:
Z
δΠ = (−Mαβ ,αβ −q) δw dS (100)
S
Z Z
¡ ¢ ¡ ¢
+ Qn − V̄n δw dl − Mnn − M¯ nn δw,n dl
ÃΓ Γ
Z !
X end
− Mtn δw|beginning − Mtn ,t δw dl
k Γ
Z
= (−Mαβ ,αβ −q) δw dS

Z Z
¡ ¢ ¡ ¢
+ Vn − V̄n δw dl − ¯ nn δw,n dl
Mnn − M
Γ Γ
X end

− Mtn δw|beginning

where Vn = Qn + Mtn ,t is the effective shear force.

In order to make the functional Π stationary under arbitrary variation of the
displacement field δw, there must hold:

Equation of Equilibrium
(101)
Mαβ ,αβ +q = 0 on S

Boundary Conditions

Mnn − M ¯ nn = 0 or δw,n = 0 on Γ

Vn − V̄n = 0 or δw = 0 on Γ
(102)
at corner points
Mnt = 0 or δw = 0
of the contour Γ

3.1.4 Specification of Equation for Rectangular Plate

Consider a rectangular plate.

24

 a
x→t

y→n

Boundary Conditions For edges parallel to x-axis, the normal direction is the
ydirection.
∂w
Myy − M̄yy = 0 or =0 (103)
∂y
Vy − V̄y = 0 or w=0

where
∂Myx
Vx = Qx + (104)
∂y
∂Mxy
Vy = Qy +
∂x

For edges parallel to y-axis, the normal direction is the x-direction.

∂w
Mxx − M̄xx = 0 or =0 (105)
∂x
Vx − V̄x = 0 or w=0

where
∂Mxy
Vy = Qy + (106)
∂x
∂Myx
Vx = Qx +
∂y

Interpretation of Corner Points

25

 1
segment [ k − 1]

segment [ k ] 2
3 direction of
integration
4

Boundary condition reads:

X [2] [1] [4] [3]

Mtn δw|end
beginning = Mtn δw
[2]
− Mtn δw[1] + Mtn δw[4] − Mtn δw[3] (107)

where
δw[3] = δw[2] (108)
thus
X ³ ´
[1] [2] [3] [4]
Mtn δw|end
beginning = −Mtn δw [1]
+ Mtn − Mtn δw[2] + Mtn δw[4] (109)

Consider the right angle.

x [ k − 1]

y
[k]

for the k − 1 side n = x, t = y (110)

for the k side n = y, t = x
³ ´
Mtn δw|end
beginning = (Mxy − Myx ) δw (111)
at the right angle

Interpretation of Corner Forces Plane stress:

τ xy = τ yx symmetry (112)

26

 surface element σy

σx dS τ xy

τ yx

Let us place the surface element at the corner.

x

y
z Edge
τ xy [ k −1]
τ yx

Edge M xy
[k]
M yx

The shearing stresses produce twisting moments which are in the opposite direction:
[k−1] [k]
Mxy = −Myx (113)

Therefore, the boundary condition at the corner becomes:

³ ´
Mtn δw|end
beginning = M [k−1]
xy − M [k]
yx δw = 2 Mxy δw = 0 (114)

Fcorner = 2 Mxy (115)

27

For the Entire Plate

2 M xy

2 Mxy

2 Mxy

2 M xy

Interpretation of the Effective Shear Vx

x
dy

y Qx
z

dy

M xy
∂M xy
M xy + dy
∂y

Equilibrium reads:
µ ¶
∂Mxy
Qx dy + Mxy + dy − Mxy (116)
∂y
µ ¶
∂Mxy
= Qx + dy
∂y

= Vx dy

∂Mxy
Vx = Qx + (117)
∂y

28
3.2 Bending-Membrane Theory of Plates
3.2.1 Total Potential Energy
The total potential energy of the system Π reads:

Π = Ub + Um − Vb − Vm (118)

where Ub is the bending strain energy, Um is the membrane strain energy, Vb is

the potential energy of external loading causing flexural response, and Vm is the
potential energy of external loading causing membrane response.
Membrane Strain Energy The membrane strain energy reads:
Z
1
Um = Nαβ ε◦αβ dS (119)
2 S
where
1 1
ε◦αβ =
(uα ,β +uβ ,α ) + w,α w,β (120)
2 2
Potential Energy of External Forces Evaluation of Boundary Terms
Normal in-plane loading, Nnn
Z
N̄nn un dl (121)
Γ

where un is normal in-plane displacement.

Shear in-plane loading, Ntt
Z
N̄tn ut dl (122)
Γ

where ut is shear component of the displacement vector.

Potential Energy of External Forces
Z Z
Vm = N̄nn un dl + N̄tn ut dl (123)
Γ Γ

3.2.2 First Variation of the Total Potential Energy

The first variation of the total potential energy reads:

δΠ = (δUb − δVb ) + (δUm − δVm ) (124)

The first parenthesis represent the terms considered already in the bending theory
of plates. All we have to do is to evaluate the term in the second parenthesis.
Here, the first variation of the membrane energy reads:
Z
δUm = Nαβ δλαβ dS (125)
S

29

where
1 1
δλαβ = (δuα ,β +δuβ ,α ) + (δw,α w,β +δw,β w,α ) (126)
2 2
Because of the symmetry of the tensor of membrane forces:

Nαβ = Nβα (127)

by using the characteristics of dummy indices we obtain:

Nαβ δuβ ,α = Nβα δuβ ,α = Nαβ δuα ,β (128)

Now, the first variation of the membrane strain energy reads:

Z ¿ ∙ ¸À
1 1
δUm = Nαβ (δuα ,β +δuβ ,α ) + (δw,α w,β +δw,β w,α ) dS (129)
2 2
ZS
= (Nαβ δuα ,β +Nαβ w,β δw,α ) dS

Note that the displacement vector has now three components:

{uα , w} (130)

so that there are three independent variations:

{δuα , δw} (131)

We expect those to end up with three independent equations of equilibrium. The

first term of δUm reads:
Z Z Z
Nαβ δuα ,β dS = (Nαβ δuα ) ,β dS − Nαβ ,β δuα dS (132)
S S S
Z Z
= Nαβ δuα nβ dl − Nαβ ,β δuα dS
ZΓ ZS
= Nγδ δuγ nδ dl − Nαβ ,β δuα dS
ZΓ Z S
= Nγ1 δuγ dl − Nαβ ,β δuα dS
ZΓ ZS
= Nγn δuγ dl − Nαβ ,β δuα dS
ZΓ S
Z
= (Nnn δun + Ntn δut ) dl − Nαβ ,β δuα dS
Γ S

30

 The second term of δUm reads:

Z Z Z
Nαβ w,β δw,α dS = (Nαβ w,β δw) ,α dS − (Nαβ w,β ) ,α δw dS (133)
S
Z S Z S
= Nαβ w,β δw nα dl − (Nαβ w,β ) ,α δw dS
ZΓ ZS
= Nγδ w,δ δw nγ dl − (Nαβ w,β ) ,α δw dS
ZΓ Z S
= N1δ w,δ δw dl − (Nαβ w,β ) ,α δw dS
ZΓ S Z
= (N11 w,1 +N12 w,2 ) δw dl − (Nαβ w,β ) ,α δw dS
ZΓ ZS
= (Nnn w,n +Nnt w,t ) δw dl − (Nαβ w,β ) ,α δw dS
Γ S

Now, the variation of external work reads:

Z Z
δVm = N̄nn δun dl + N̄tn δut dl (134)
Γ Γ

3.2.3 Equilibrium Equation and Boundary Conditions

The contribution of the term (δUm − δVm ) then becomes:
Z Z
δ (Um − Vm ) = (Nnn δun + Ntn δut ) dl − Nαβ ,β δuα dS (135)
Γ S
Z Z
+ (Nnn w,n +Nnt w,t ) δw dl − (Nαβ w,β ) ,α δw dS
ZΓ Z S

− N ¯ nn δun dl − N ¯tn δut dl

ZΓ Γ Z Z
¡ ¢ ¡ ¢
= − Nαβ ,β δuα dS + ¯
Nnn − Nnn δun dl + ¯tn δut dl
Ntn − N
ZS Γ Z Γ

− (Nαβ w,β ) ,α δw dS + (Nnn w,n +Nnt w,t ) δw dl

S Γ

The first three integrals involve independent variations of uα , i.e. δuα or {δun , δut }.
This gives us two independent equations of equilibrium in the plane of the plate:

Equation of Equilibrium I
(136)
Nαβ ,β = 0 on S

31

and two additional boundary conditions:

Boundary Conditions I

Nnn − N¯nn = 0 or δun = 0 on Γ (137)

¯tn = 0
Ntn − N or δut = 0 on Γ

The remaining two integrals involve variation in the out-of-plane displacement

δw and thus should be combined with the equation of equilibrium and boundary
conditions governing the flexural response. The terms involving surface integral
should be added to the equation of equilibrium:

Equation of Equilibrium II
(138)
Mαβ ,αβ + (Nαβ w,β ) ,α +q = 0 on S

where the second term in the left hand is the new term arising from the finite
rotation.
The term with the line integral should be added to the corresponding term
involving variation δw:
Z
¡ ¢
Vn + Nnn w,n +Nnt w,t −V̄n δw dl = 0 (139)
Γ

The generalized boundary conditions reads:

Boundary Conditions II- (A)

(140)
Vn + Nnn w,n +Nnt w,t −V̄n = 0 or δw = 0 on Γ

where the second and third terms in the left hand side of the first equation are the
new terms arising from the finite rotation.
If the boundaries of the plate are kept undeformed w,t = 0 (simply supported
or clamped plate), then the boundary condition is satisfied:

Vn + Nnn w,n −V̄n = 0 or δw = 0 on Γ (141)

Physically, the additional terms represent the contribution of the axial force to the
vertical equilibrium. Using the in-plane equilibrium, Nαβ ,β = 0, the out-of-plane
equilibrium can be transformed to the form:

Mαβ ,αβ +Nαβ ,α w,β +Nαβ w,αβ +q = 0 on S (142)

32
 Equation of Equilibrium II’
(143)
Mαβ ,αβ +Nαβ w,αβ +q = 0 on S

which is called as the von Karman equation. Note that Nαβ is related through the
Hook’s law with the gradient of the in-plane displacement uα , i.e. Nαβ = Nαβ (uα ).
Therefore, the new term Nαβ w,αβ represents in fact coupling between in-plane and
out-out-plane deformation.
To make derivation complete, the final boundary conditions which do not changed
from the bending theory of plate are presented:

Boundary Conditions II- (B)

¯ nn = 0
Mnn − M or δw,n = 0 on Γ
(144)
at corner points
Mnt = 0 or δw = 0
of the contour Γ

33

4 General Theories of Plate

4.1 Bending Theory of Plates
4.1.1 Derivation of the Plate Bending Equation
Then, groups of equations!

• Equilibrium
Mαβ ,αβ +q = 0 on S (145)

• Geometry
καβ = −w,αβ (146)

• Elasticity
Mαβ = D [(1 − ν) καβ + ν κγγ δ αβ ] (147)

Eliminating curvature καβ between Eq. (146) and (147), we obtain:

Mαβ = −D [(1 − ν) w,αβ +ν w,γγ δ αβ ] (148)

Substituting Eq. (148) into Eq. (145) reads:

−D [(1 − ν) w,αβ +ν w,γγ δ αβ ] ,αβ +q = 0 (149)

−D [(1 − ν) w,αβαβ +ν w,γγαβ δ αβ ] + q = 0

Note that the components of the Kronecker "δ αβ " tensor are constant and thus are
not subjected to differentiation:
¯ ¯ ½

¯ 1 0 ¯
1 if α = β
¯ ¯
δ αβ =
¯
¯
or δ αβ =
(150)

01
0 if α 6= β

Also, note that only these components:

¤αβ δ αβ = ¤αα (151)

survive in the matrix multiplication for which α = β. Therefore, Eq. (149) now
reads:
−D [(1 − ν) w,αβαβ +ν w,γγαα ] + q = 0 (152)
Because "γγ" are "dummy" indices, they can be replaced by any other indices, for
example "ββ."
−D [(1 − ν) w,αβαβ +ν w,ββαα ] + q = 0 (153)
The order of differentiation does not matter:

w,αβαβ = w,ααββ = w,ββαα

34
Thus, two terms in Eq. (153) can now be added to give the plate bending equation:
D w,ααββ = q for α, β = 1, 2 (154)
Here, the index notation can be expended:
w,ααββ = w,11ββ +w,22ββ (155)
= w,1111 +w,2211 +w,1122 +w,2222
= w,1111 +2 w,1122 +w,2222
Now, letting "1 → x", "2 → y" leads:
µ 4 ¶
∂ w ∂4w ∂4w
D +2 2 2 + = q (x, y) (156)
∂x4 ∂x ∂x ∂y4
Alternative notation can be:

D ∇4 w = q (157)

where Laplacian ∇2 w reads:

∂2w ∂2w
∇2 w = + (158)
∂x2 ∂y 2
and bi-Laplican ∇4 w reads:
¡ ¢
∇4 w = ∇2 ∇2 w (159)
∂2 ¡ ¢ ∂2 ¡ ¢
= 2 ∇2 w + 2 ∇2 w
∂x ∂y
2 µ 2 ¶ µ ¶
∂ ∂ w ∂2w ∂ 2 ∂ 2w ∂ 2w
= 2 + + +
∂x ∂x2 ∂y2 ∂y 2 ∂x2 ∂y 2
4
∂ w 4
∂ w 4
∂ w
= 4
+2 2 2 +
∂x ∂x ∂x ∂y 4
4.1.2 Reduction to a System of Two Second Order Equations
Denote
D w,αα = −M (160)
Then, from the equilibrium equation:
[D w,αα ] ,ββ = q (161)
we obtain a system of two linear partial differential equations of the second order:

M,ββ = −q
(162)
D w,αα = −M

35
or ( ∂2M ∂2M
2 + ∂y 2 = −q
³ 2
∂x ´ (163)
∂ w ∂2w
D ∂x2 + ∂y 2 = −M
What is "M " ? Let us calculate Mαα :

Mαα = M11 + M22 (164)

= D [(1 − ν) κ11 + ν (κ11 + κ22 ) δ 11 ]
+D [(1 − ν) κ22 + ν (κ11 + κ22 ) δ 22 ]
= D [(1 + ν) (κ11 + κ22 )]
= D (1 + ν) καα

or
Mαα
= D καα = −D w,αα = M (165)
1+ν
Therefore,

Mαα = M (1 + ν) (166)
= D καα (1 + ν)

Now, moment sum reads:

M = D καα (167)

and in expanded notation it reads:

M = D [κxx + κyy ] (168)

4.1.3 Exercise 1: Plate Solution

Consider a simply supported plate.

Square plate ( a× a)

a x

36

Boundary Condition General boundary condition reads:

¡ ¢
¯ nn w,n = 0 on Γ
M¡nn − M ¢ (169)
Vn − V̄n w = 0 on Γ

M̄nn = 0 ⇒ Mnn = 0 on Γ
(170)
w=0 on Γ
w=0 at x = 0 and x = a , 0 ≤ y ≤ a
(171)
w=0 at y = 0 and y = a , 0 ≤ x ≤ a
Mxx = 0 at x = 0 and x = a , 0 ≤ y ≤ a
(172)
Myy = 0 at y = 0 and y = a , 0 ≤ x ≤ a
Loading Condition Assume for simplicity the sinusoidal load distribution:
³π x´ ³π y ´
q (x, y) = q0 sin sin (173)
a a
where q0 is a pressure intensity.
Solution of Problem The solution of the form
³π x´ ³π y ´
w (x, y) = w0 sin sin (174)
a a
satisfy both the boundary conditions and the governing equations (see below).
Plate Bending Equation Substituting Eq. (173) and (174) into the plate bend
ing equation (156), one gets:
½ ∙³ ´ ³ π ´4 ³ π ´4 ¸ ¾ ³π x´ ³π y´
π 4
D w0 +2 + − q0 sin sin = 0 (175)
a a a a a
½ ³ π ´4 ¾ ³π x´ ³π y´
4 D w0 − q0 sin sin =0
a a a
In order to satisfy the above equation for all values of x and y, the coefficient in
the bracket must vanish. This gives:
q0 ³ a ´4
w0 = (176)
4D π
¡ ¢ £ ¡ ¢¤
where D = Eh3 / 12 1 − ν 2 .
Bending Moments The various bending moments are given by:
∙ 2 ¸ ³ π ´2 ³π x´ ³π y ´
∂ w ∂2w
Mxx = −D + ν = D (1 + ν) w 0 sin sin (177)
∂x2 ∂y 2 a a a
∙ 2 ¸ ³ π ´2 ³π x´ ³π y ´
∂ w ∂2w
Myy = −D + ν = D (1 + ν) w 0 sin sin
∂y2 ∂x2 a a a
2
∂ w ³ ´
π 2 ³ πx ´ ³ π y´
Mxy = −D (1 − ν) = −D (1 − ν) w0 cos cos
∂x∂y a a a

37
 Shear Components The shear components Qx and Qy are:
∂Mxx ∂Mxy
Qx = + (178)
∂x ∂y
∂Myy ∂Mxy
Qy = +
∂y ∂x
Now, using the previously obtained bending moments, we get the shear components
in the interior of the plate:
³ π ´3 ³π x´ ³π y´
Qx = 2 D w0 cos sin (179)
a a a
³ π ´3 ³π x´ ³π y´
Qx = 2 D w0 sin cos
a a a
Effective Shear Components Next, let us computer the effective shear com
ponents:
∂Mxy
Vx = Qx + (180)
∂y
∂Mxy
Vy = Qy +
∂x
Using the previous results, we get:
³ π ´3 ³π x´ ³π y ´
Vx = (3 − ν) D w0 cos sin (181)
a a a
³ π ´3 ³π x´ ³π y ´
Vy = (3 − ν) D w0 sin cos
a a a
We now need to evaluate the effective shear on the boundaries:
h ¡ ¢3 i h ¡ ¢3 i
Vx / (3 − ν) D πa Vy / (3 − ν) D πa
¡ ¢
x=0 w0 sin ¡πay ¢ 0
x=a −w0 sin a π y
0¡ ¢ (182)
y=0 0 w0 sin ¡πa
x ¢
y=a 0 −w0 sin πax

Because our sign convention is:

positive shear

38

in our case, shear along the boundary is:

From the above results, we can plot the shear distribution:

Force Balance Integrating the effective shear along the boundary, we get:
Z Z a ³ π ´3 Z a ³π y ´
R= Vn dxt = 4 Vx |x=0 dy = 4 (3 − ν) D w0 sin dy (183)
L 0 a 0 a
Then, the reduction force due to effective shear on boundaries reads:

¡ a ¢2
R = 2 (3 − ν) q0 π
(184)

Now, let us complete the total load acting on the plate:

Z Z aZ a ³π x´ ³π y ´
P = q (x, y) dS = q0 sin sin dx dy (185)
S 0 0 a a
Then, the total external load acting on the plate reads:

¡ a ¢2
P = 4 q0 π
(186)

Notice that R and P do not balance! We did not include the corner forces. These
are given by:
(Fcorner )x0 ,y0 = 2 (Mxy )|x=x0 ,y=y0 (187)

39
Because of the symmetry, all four forces are equal. So, compute the corner force
at x = y = 0, (Fcorner )0,0 :
h³π x´ ³ π y ´i¯
¯
(Fcorner )0,0 = 2 cos cos ¯ (188)
a a 0,0
³ π ´2
= − 2 D (1 − ν) w0
a
Now, the vertical force balance is satisfied:

R + 4 Fcorner = P (189)

³ a ´2 ³ π ´2 ³ a ´2
2 (3 − ν) q0 − 8 D (1 − ν) w0 = 4 q0 (190)
π a π
³ a ´2 ³ π ´2 ³ a ´2
2 (3 − ν) q0 − 2 (1 − ν) q0 = 4 q0
π a π
4.1.4 Exercise 2: Comparison between Plate and Beam Solution
Plate Solution For a square simply supported plate under loading qplate (x, y)
given by: ³π x´ ³π y ´
qplate (x, y) = (q0 )plate sin sin (191)
a a
we found that the plate deflection is:
³π x´ ³π y ´
wplate (x, y) = (w0 )plate sin sin (192)
a a
with:
(q0 )plate ³ a ´4
(w0 )plate = (193)
4¡ D ¢π
3 1 − ν (q0 )plate ³ a ´4
2
=
E h3 π
For the plate, the total load is given by:
Z aZ a ³π x´ ³π y´
Pplate = (q0 )plate sin sin dy dx (194)
0 0 a a
³ a ´2

= 4 (q0 )plate

40

Wide Beam Solution For a wide beam under line loading given by:
³π x´
qbeam (x, y) = (q0 )beam sin (195)
a
we need to compute the central deflection (w0 )beam from:
0000
E I wbeam = qbeam (x) (196)
where I = ah3 /12. Assuming the deflection wbeam (x):
³π x´
wbeam (x) = (w0 )beam sin (197)
a
we get:
³ π ´4 ³π x´ ³π x´
EI (w0 )beam sin = (q0 )beam sin (198)
a a a
Thus,
(q0 )beam ³ a ´4
(w0 )beam = (199)
EI π
12 (q0 )beam ³ a ´4
=
E a h3 π
Now, let us compute the total forces:
Z a ³π x´
Pbeam = (q0 )beam sin dx (200)
0 a
a
= 2 (q0 )beam
π
Comparison For both total forces to be equal, we need to have:
Pplate = Pbeam (201)
³ a ´2 a
4 (q0 )plate = 2 (q0 )beam
π π
a
(q0 )beam = 2 (q0 )plate (202)
π
With a concentrated load, the beam deflection now becomes:
(q0 )beam ³ a ´4
(w0 )beam = (203)
EI π
24 (q0 )plate ³ a ´4
=
π E h3 π
We now can compute the ratio of central deflections:
3 (1−ν 2 ) (q0 )plate ¡ a ¢4
(w0 )plate E h3 π
α= = ¡ ¢ (204)
(w ) 0 beam 24 (q0 )plate a 4
π E h3 π
π¡ ¢
= 1 − ν 2 ' 0.36
8
The above equation means that under the same total load, a plate is three times
stiffer than a wide beam. The ratio α will vary slightly depending on the load
distribution (sinusoidal, uniform, concentrated load, etc.).

41
4.1.5 Exercise 3: Finite Difference Solution of the Plate Bending Prob
lem
Governing Equations read:
∇2 M = −q
(205)
∇2 w = − M
D
or in the component notation they read:
∂2M ∂2M
∂x2 + ∂y 2 = −q
∂2w ∂2w M (206)
∂x2 + ∂y2 = − D

where M is the moment sum defined by:

Mαα
M= = D καα (207)
1+ν
Case of Simply Supported Plate The boundary condition of a simple sup
ported plate reads:

w=0 on Γ (208)
Mnn = 0 on Γ

a
x

a
n
y

For sides parallel to x-axis (thick lines), one gets:

Mnn = Myy = 0 (209)

dw d2 w
w=0 → =0 → =0 → κxx = 0 (210)
dx dx2
From the general constitutive equations,

Myy = D [κyy + ν κxx ] (211)

0 = D [κyy + ν · 0] → κyy = 0

Therefore,
M = D [καα + κββ ] = D [0 + 0] = 0 (212)

42
Similar derivation can be performed for two edges parallel to y-axis. Then, M =
0. It can be concluded that for a simply supported plate the following boundary
conditions hold: ( 2
∂ M 2
∂x2
+ ∂∂yM 2 = −q in S
(213)
M =0 on Γ
(
2 2
∂ w
∂x2
+ ∂∂yw2 = − M
D in S
(214)
w=0 on Γ
Therefore, the above two boundary value problems are uncoupled.
The Finite Difference Technique An approximation to the first and second
derivatives.
z

zn+1
zn

h
x
xn xn+1

m− 1

x
n−1 n, m n+1

m+1

¯
dz
¯¯backward zn − zn−1
' (215)
dx ¯n h

¯f orward
dz
¯¯ zn+1 − zn
'
dx ¯ n+1 h

43
 ¯ ∙
¸

d2 z
¯¯ d dz

=
(216)

dx2 ¯
n dx dx

¡
dz ¢
¡
dz ¢

dx n+1 − dx n
=

zn+1 −zn
h − zn −zhn−1
=
h
zn+1 − 2 zn + zn−1
=
h2
¯
d2 z
¯¯ zm+1 − 2 zm + zm−1
2 ¯
=
(217)

dy m h2

∂2z ∂2z
∇2 z = 2
+ 2 (218)
∂x ∂y
1
= 2 (zn+1 − 2 zn + zn−1 + zm+1 − 2 zm + zm−1 )
h

Top

h
h h
Left 0 Right
h

Bottom

1
∇2 z = (zT + zB + zL + zR − 4 z0 ) (219)
h2
Divide the plate into sixteen identical squares and distinguish six representative
nodes: three in the interior and three at the boundary. Because of symmetry, it is
enough to consider only an eighth of the plate.

44

 6 5 4

3 2

a
1 a
h=
4

Four axes of symmetry

Determination of Moment For each interior point (1, 2, 3), we write equation
∇2 M = −q. For each boundary point (4, 5, 6), we write boundary condition
M = 0 (uniform pressure).

q a2
Point 1: 4 M2 − 4 M1 = − (220)
16
q a2
Point 2: M1 + M4 + 2 M3 − 4 M2 = −
16
q a2
Point 3: 2 M5 + 2 M2 − 4 M3 = −
16
Point 4: M4 = 0
Point 5: M5 = 0
Point 6: M6 = 0

45
Substituting three last equations of Eq. (220) into the first three equations of Eq.
(220), one ends up with the following system of linear algebraic equations:
⎧
⎪ q a2
⎨ 4 M2 − 4 M1 = − 16
2
M1 + 2 M3 − 4 M2 = − q16a (221)
⎪
⎩
q a2
2 M2 − 4 M3 = − 16

whose solution is:

9
M1 = q a2 (222)
128
7
M2 = q a2
128
11
M3 = q a2
256
At the plate center, Mxx = Myy so that:
Mxx + Myy 2 Mxx
M= = (223)
1+ν 1+ν
1
Mxx = (1 + ν) M (224)
2
At the center, M = M1 , thus,
1
Mxx = (1 + ν) M1 (225)
2
1 9
= (1 + ν) q a2
2 128
= 0.0457 q a2

This is 4.6% less than the exact solution which is (Mxx )exact = 0.0479 q a2 from
the text book.
Determination of Deflection For each interior point (1, 2, 3), we write equation
∇2 w = −M/D. For each boundary point (4, 5, 6), we write boundary condition
w = 0.
µ ¶ 2
M1 a2 9 qa2 a
Point 1: 4 w2 − 4 w1 = − =− (226)
D 16 128 D 16
µ ¶ 2
M2 a2 7 q a2 a
Point 2: w1 + w4 + 2 w3 − 4 w2 = − =−
D 16 128 D 16
2
µ 2
¶ 2
M3 a 11 q a a
Point 3: 2 w5 + 2 w2 − 4 w3 = − =−
D 16 256 D 16
Point 4: w4 = 0

Point 5: w5 = 0

Point 6: w6 = 0

46
Similarly, ⎧
⎪ 9 qa4
⎨ 4 w2 − 4 w1 = − 2048 D

7 q a4 (227)

w1 + 2 w3 − 4 w2 = − 2048
⎪
⎩
11 q a 4
D
2 w2 − 4 w3 = − 4096 D
Finally, the finite difference solution is:

33 q a4 q a4
w1 = = 0.00403 (228)
8196 D D
3 q a4 q a4
w2 = = 0.00293
1024 D D
35 q a4 q a4
w3 = = 0.00214
16384 D D
On the other hand, the exact deflection of the center point is:

q a4
(w1 )exact = 0.00416 (229)
D
Thus, the error of the finite different solution is 3.1%.
4.2 Membrane Theory of Plates
4.2.1 Plate Membrane Equation
Assume that the bending rigidity is zero, D = 0. The plate becomes now a
membrane.

• Equilibrium of in-plane equation

Nαβ ,α = 0 on S (230)

Equilibrium of out-of-plane equation

Nαβ w,αβ +q = 0 on S (231)

• Strain-displacement relation
1 1
ε◦αβ = (uα ,β +uα ,β ) + w, α w,β (232)
2 2

• Constitutive equation
£ ¤
Nαβ = C (1 − ν) ε◦αβ + ν ε◦γγ δ αβ (233)
¡ ¢
where C = Eh/ 1 − ν 2 .

This is a non-linear system of equation which is difficult to solve. Note that

corresponding system of equation for the plate bending was linear.

47
4.2.2 Plate Equation for the Circular Membrane
Cylindrical coordinate system is composed of ur , uθ , uz = w.

• Equilibrium of in-plane equation

∂Nrr
r + Nrr − Nθθ = 0 on S (234)
∂r
Equilibrium of out-of-plane equation
∙ ¸
∂ ∂w
Nrr r +r q =0 on S (235)
∂r ∂r

• Strain-displacement relation
µ ¶2
∂ur 1 ∂w
λrr = + (236)
∂r 2 ∂r
ur
λθθ =
r

• Constitutive equation

Nrr = C [λrr + ν λθθ ] (237)

Nθθ = C [λθθ + ν λrr ]
¡ ¢
where C = Eh/ 1 − ν 2 .

4.2.3 Example: Approximation Solution for the Clamped Membrane

Consider a circular plate with the clamped support.
q

r
a

w0

w(r )

48

Membrane Solution From the symmetry and clamped boundary condition, the
radial displacement ur reads:

ur (r = 0) = 0 (238)
ur (r = a) = 0

Thus, as a first approximation, it is appropriate to assume:

ur ≡ 0 for 0 ≤ r ≤ a (239)

Then, the hoop strain vanishes:

εθθ = 0
Now, the radial force and the radial strain component become:

Nrr = C εrr (240)

µ ¶2
1 ∂w
εrr = (241)
2 ∂r
With the assumption ur = 0, the in-plane equilibrium equation can not be satisfied.
Consider out-of-plane equilibrium equation only. Substituting Eq. (240) and
(241) into Eq. (235), one gets:
" µ ¶ #
∂ C ∂w 2 ∂w
r = − r q on S (242)
∂r 2 ∂r ∂r

Integrating both sides once with respect to r reads:

µ ¶3
C ∂w r2 q
r =− + c1 (243)
2 ∂r 2

At the center of the membrane, the slope should be zero. Thus, one gets:
∂w
=0 at r = 0 (244)
∂r
⇒ c1 = 0

Then, Eq. (243) can be written:

r
∂w qr
=−3 (245)
∂r C
Integrating the above equation again reads:
r
3 3 q 4/3
w=− r + c2 (246)
4 C

49

The integration constant c2 can be determined from the zero deflection condition
at the clamped edge:

w=0 at r = a (247)
r
3 3 q 4/3
⇒ c2 = a
4 C
¡ ¢
Recalling the definition of the axial rigidity C = Eh/ 1 − ν 2 , Eq. (246) can be
put into a final form:
r ∙
w 3 3 (1 − ν 2 ) q a ³ r ´4/3 ¸
= 1− (248)
a 4 Eh a
r ∙ ³ r ´4/3 ¸
3
qa
' 0.73 1−
Eh a

In particular, the central deflection w (r = 0) = w0 is related to the load intensity

by: r
w0 qa
= 0.73 3 (249)
a Eh
Bending Solution It is interesting to compare the bending and membrane re
sponse of the clamped circular plate. From the page 55 of Theory of Plates and
Shells (2nd Ed.) by Timoshenko and Woinowsky-Krieger, the central deflection of
the plate is linearly related to the loading intensity:

w0 q a3
= (250)
a 64¡D ¢
3 1 − ν 2 q ³ a ´3
=
16 E h
q ³ a ´3
' 0.17
E h
Assume that a/h = 10, then Eq. (250) yields:
w0 qa
= 17 (251)
a Eh
Comparison A comparison of the bending and membrane solution is shown in
the next figure.

50

It is seen that a transition from the bending to membrane response occurs at

w0 /a = 0.15 which corresponds to w0 = 1.5 h. When the plate deflection reach
approximately plate thickness, the membrane action takes over the bending action
in a clamped plate. If the plate is not restrained from axial motion, then the
assumption ur = 0 is no longer valid, and a separate solution must be developed.
4.3 Buckling Theory of Plates
4.3.1 General Equation of Plate Buckling
• Equilibrium of in-plane equation

Nαβ ,α = 0 on S (252)

Equilibrium of out-of-plane equation

Mαβ ,αβ +Nαβ w,αβ +q = 0 on S (253)

• Strain-displacement relation
1 1
ε◦αβ = (uα ,β +uα ,β ) + w,α w,β (254)
2 2
καβ = − w,αβ

• Constitutive equation for axial force and axial strain

£ ¤
Nαβ = C (1 − ν) ε◦αβ + ν ε◦ γγ δ αβ (255)
¡ ¢
where C = Eh/ 1 − ν 2 , and another one for moment and curvature

Mαβ = D [(1 − ν) καβ + ν κγγ δ αβ ] (256)

£ ¡ ¢¤
where D = Eh3 / 12 1 − ν 2 .

51
 By combining Eq. (254) and (256), one gets:

Mαβ ,αβ = −D w,ααββ (257)

Substituting Eq. (257) into Eq. (253) leads:

−D w,ααββ +Nαβ w,αβ +q = 0 on S (258)

The buckling problem is specified by:

q≡0 (259)

Now, changing signs leads the general out-of-plane equation for the buckling of
the plates:

D w,ααββ −Nαβ w,αβ = 0 (260)

where the second term in the left hand is non-linear due to Nαβ which should be
obtained from:

Nαβ ,α = 0 (261)

4.3.2 Linearized Buckling Equation of Rectangular Plates

The nonlinear buckling equation can be separated into two linear equations: one
for in-plane equation for Nαβ and another one for w.

52

 a
x

Px Px

b
ux
y

Px

Post-buckling
(δw ≠ 0)
k
Pc 2

Pre-buckling
(δw = 0)
k
ux

Pre-Buckling Problem Recall that:

µ ¶
◦ 1 ∂uα ∂uβ 1 ∂w ∂w
εαβ = + + (262)
2 ∂xβ ∂xα 2 ∂xα ∂xβ

Eh £ ◦ ◦
¤
Nαβ = (1 − ν) εαβ + ν εγγ δ αβ (263)
1 − ν2
In the pre-buckling problem, the linear equilibrium equations are obtained by
omitting the nonlinear terms in the governing equations Eq. (260) and (261). The
resulting equations are now:
D w,ααββ = 0

Nαβ ,β = 0

For the pre-buckling trajectory, δw = 0, one gets the equilibrium equation:

Nαβ ,β = 0 (264)

53
where
Eh £
¤

Nαβ = 2
(1 − ν) ε◦αβ + ν ε◦γγ δ αβ (265)

1−ν
µ ¶

◦ 1
∂uα ∂uβ
εαβ =
+
(266)

2
∂xβ ∂xα
and boundary condition:
¡ ¢

Nnn − N̄nn δun = 0 on Γ (267)

Here, it is assumed that the unknown membrane force tensor Nαβ is equal to
◦ :
the similar quantity known from the pre-buckling solution Nαβ
◦
Nαβ = −Nαβ

where the compressive pre-buckling membrane force are defined as positive.

Post-Buckling Problem Now, the governing equation for buckling of plates
reads:
Boundary
D ∇4 w + Nαβ
◦ w,
αβ = q + (268)
Conditions

◦ is defined as:
where membrane force tensor in the pre-buckling solution Nαβ
¯ ¯
¯ Ñxx Ñxy ¯
◦
Nαβ = λ Ñαβ = λ
¯¯
¯ (269)

Ñyx Ñyy ¯

where N̄αβ is the known direction from the pre-buckling analysis, and η is unknown
load amplitude. Now, the nonlinear buckling equation becomes a linear eigenvalue
problem:

D ∇4 w + λ Ñαβ w,αβ = 0 (270)

where λ ≥ 0 is eigenvalues, and w is eigenfunctions.

4.3.3 Analysis of Rectangular Plates Buckling
Simply Supported Plate under In-Plane Compressive Loading Consider
a plate simply supported on four edges. The plate is subjected to an in-plane
compressive load Px uniformly distributed along the edges x = [0, a].

54

 a
x

Px Px

y
h

From equilibrium equations, one gets:

¯ ◦ ¯
¯ Nxx Nxy ◦ ¯
◦ ¯
Nαβ = ¯
◦ ¯ (271)

Nyx Nyy ◦ ¯

¯ ¯
¯ Ñxx Ñxy ¯
= λ
¯¯ ¯
Ñyx Ñyy ¯

¯ ¯
Px ¯¯ 1 0 ¯¯

b
¯ 0 0 ¯

Introducing Eq. (271) into Eq. (268) leads:

Px
D ∇4 w + w,xx = 0 (272)
b
Boundary condition for this simply supported plate are written as:
w=0 on Γ (273)
Mnn = 0 on Γ
where the moment components read:
Mxx = −D (w,xx +ν w,yy ) = 0 (274)
Myy = −D (w,yy +ν w,xx ) = 0
Thus, one gets:

w = w,xx = 0 on x = [0, a] (275)

w = w,yy = 0 on x = [0, b]

Equation (272) is a constant-coefficient equation, and a solution of the following

form: ³m π x´ ³n π y ´
w = c1 sin sin for m, n = 1, 2 (276)
a b

55
satisfies both the differential equation and the boundary conditions. Introduction
into Eq. (272) gives:
∙³ ³ m π ´2 ³ n π ´2 ³ n π ´4 ¸ P ³ m π ´2
m π ´4 x
D +2 + − =0 (277)
a a b b b a

Px ³ π a ´2 ∙³ m ´2 ³ n ´2 ¸2
⇒ =D + (278)
b m a b
where for the discrete values of Px Eq. (272) has nontrivial solutions. The critical
load can be determined by the smallest eigenvalue, i.e. n = 1 for all values of a:
" #
Px ³ π a ´2 ³ m ´2 µ 1 ¶2
=D + (279)
b m a b
µ ¶ " µ ¶2 #
π 2 D a b 2 ³ m ´2 1
= 2 +
b m a b
µ ¶
π2 D m b a 2
= 2 +
b a mb

Now, the critical load (Px )cr can be written as:

π2 D
(Px )cr = kc (280)
b

where µ ¶2
mb a
kc = + (281)
a mb
where coefficient kc is a function of aspect ratio a/b and wavelength parameter m.

56

 2 6 12

For a given a/b, m may be chosen to yield the smallest eigenvalue. In order to
minimize kc in Eq. (281), treating m as a continuous variable produces:
µ ¶µ ¶
∂kc mb a b a

=2 + − =0 (282)

∂m a mb a b m2

where the first bracket can not be zero, so the second bracket should be zero:
b a 1
⇒ − =0 (283)
a b m2
Now, one gets:

a
m=
b (284)
kc = 4

Here, this is valid when a/b is integer and when considering a very long plates.
Transition from m to m+1 half-waves occurs when the two corresponding curves
have equal ordinates, i.e. from Eq. (281):

kc |m = kc |m+1 (285)

mb a (m + 1) b a
⇒ + = + (286)
a mb a (m + 1) b
a p
⇒ = m (m + 1)
b

57
 a p
= m (m + 1) (287)
b

√
Example 1 For m = 1, a/b = 2
Example 2 For a very large m, i.e. a very long plate, a/b ' m. Thus, kc = 4 is
now independent of m.
A very long plate buckles in half-waves, whose lengths approach the width of the
plate: ³π x´ ³n π y ´
w = c1 sin sin
b b
Thus, the buckled plate subdivides approximately into squares.
Various Boundary Conditions of Plate under In-Plane Compressive Load
ing The critical buckling load reads:
π2 D
(Px )cr = kc
b

16

14

12

10 c c
c A B
ss

kc 8 ss c
ss C free D

ss
6 free E

4
Loaded edges clamped.
Loaded edges simply supported.
2

0
0 1 2 3 4 5
a
b

Figure by MIT OCW.

Influence of boundary conditions on the buckling coefficients of

plates subjected to in-plane compressive loading

58
 Various Boundary Conditions of Plate under In-Plane Shear Loading
The critical buckling load per unit length reads:

π2 D
(Nxy )cr = kc
b2
where the dimension of Nxy is [N/m].

15

2
rcr = � D
kc

13 b2h

Clamped Edges

11

kc 9

Simply Supported Edges

0 1
2 3 4 5
a

Critical values of shear stress for plates subjected to in-plane loading.

Figure by MIT OCW.

Limiting Case: Wide Plates Consider a wide plate for which a/b ¿ 1. From
the diagram, we see that if a/b < 1, then m is set to be equal to unity, i.e. just one
wavelength in the x-direction.

59
 a
x

b y

The buckling formula thus becomes:

π2 D
(Nx )cr = kc |m=1 (288)
b2
µ ¶
π2 D b a 2
= 2 +
b a b
µ ¶
π 2 D a ´2 b a 2
³
= 2 +
a b a b
∙ ³ ¸
π2 D a ´2 2
= 2 1+
a b

If a/b ¿ 1, then the second term in the bracket can be neglected so that the
buckling load per unit length becomes:

π2 D
(Nx )cr =
a2

which is called Sezawa’s formula for wide plates.

Example 3 Here, relative merits of stiffening a large panel are investigated in the

longitudinal or in the transverse direction. It is assumed that the stiffeners provide

for a simply supported boundary conditions.

Consider the case of longitudinal stiffeners.

60

σcr

From the von Karman formula, the buckling load per unit length for each divided
part reads:
4π 2 D
(Nx )cr =
s2
Now, the buckling stress can be calculated:
(Nx )cr
(σ cr )longitudinal =
h
2
4π D
= 2
s h

Consider the case of transverse stiffeners.

L

σcr s

From the Sezawa’s formula, the buckling load per unit length along the loaded edges
reads:
π2 D
(Nx )cr = 2
s
Now, the buckling stress can be calculated:
(Nx )cr
(σ cr )transverse =
h
π2D
= 2
s h

Thus, it is concluded that

(σ cr )longitudinal = 4 (σ cr )transverse

This shows advantages of longitudinal stiffeners over transverse stiffeners.

61
4.3.4 Derivation of Raleigh-Ritz Quotient
Recall the total potential energy of system and other corresponding definitions:

Π = (Ub − Vb ) + (Um − Vm ) (289)

where each term for buckling problems will be discussed in the following.
Term Relating to Plate Bending Response In the buckling problem, the
work done by external load causing bending response considered as zero:

Vb = 0 (290)

The bending energy can be expressed:

Z
1
Ub = Mαβ καβ dS (291)
2 S
Z
D
= [(1 − ν) καβ + ν κγγ δ αβ ] καβ dS
2 S
Z h i
D
= (1 − ν) καβ καβ + ν (κγγ )2 dS
2 S
Z h i
D
= (1 − ν) (κ11 κ11 + κ12 κ12 + κ21 κ21 + κ22 κ22 ) + ν (κ11 + κ22 )2 dS
2 S
Z n h io
D
= (κ11 + κ22 )2 − 2 (1 − ν) κ11 κ22 − (κ12 )2 dS
2 S

Here, the term in the square bracket is called Gaussian curvature:

κ11 κ22 − (κ12 )2 = κI κII (292)

where κI and κII are the principal curvatures. For plates with straight edges,
Gaussian curvature vanishes, so one gets:
Z
D
Ub = (κ11 + κ22 )2 dS (293)
2 S

The integrand of the above equation can be written in terms of the transverse
displacement:
¡ ¢2
(κ11 + κ22 )2 = −∇2 w (294)
2 2
=∇ w ∇ w

Now, the bending energy reads:

Z
D
Ub = ∇2 w ∇2 w dS (295)
2 S

62
Term Relating to Plate Membrane Response The work done by external
load causing membrane response reads:
Z Z
Vm = N̄nn un dl + N̄tn ut dl (296)
Γ Γ

In the buckling problem, the axial force Nαβ ◦ = λN ˜αβ is determined from the
pre-buckling solution and is considered as constant, so the membrane energy reads:
Z
Um = −λ Ñαβ ε◦αβ dS (297)
ZS ∙ ¸
1 1
= −λ Ñαβ (uα ,β +uα ,β ) + w,α w,β dS
2 2
ZS Z
λ
= −λ Ñαβ uα ,β dS − Ñαβ w,α w,β dS
S 2 S
Here, the first term can be extended in a similar way shown in Eq. (132):
Z Z ³ ´
−λ Ñαβ uα ,β dS = −λ Ñnn un + Ñtn ut dl (298)
S ZΓ
−λ Ñαβ ,β uα dS
ZS ³ ´
= −λ Ñnn un + Ñtn ut dl
Γ

where the in-plane equilibrium is applied:

−λ Ñαβ ,β = 0 (299)

Now, the membrane energy can be expressed:

Z ³ ´ Z
λ
Um = −λ Ñnn un + Ñtn ut dl − Ñαβ w,α w,β dS (300)
Γ 2 S
Thus, the term relating to membrane response can be summarized:
Z ³ ´ Z
λ
Um − Vm = −λ Ñnn un + Ñtn ut dl − Ñαβ w,α w,β dS (301)
2 S
Z Γ Z
− N ¯nn un dl − N ¯tn ut dl
Z ³ Γ Γ Z ³
´ ´
= −λ Ñnn − N̄nn un dl − −λ Ñtn − N̄tn ut dl
Γ Γ
Z
λ
− Ñαβ w,α w,β dS

2 S

Z
λ
=− Ñαβ w,α w,β
2 S
where the boundary conditions on Γ are applied.

63
Total Potential Energy and Its Variations Now, one gets the total potential
energy:
Π = (Ub − Vb ) + (Um − Vm ) (302)
Z Z
D 2 2 λ
= ∇ w ∇ w dS − Ñαβ w,α w,β dS
2 S 2 S
The first variation of the potential energy can be obtained:
Z
D £ ¡ 2 ¢ 2 ¡ ¢¤
δΠ = δ ∇ w ∇ w + ∇2 w δ ∇2 w dS (303)
2 S
Z
λ
− Ñαβ (δw,α w,β +w,α δw,β ) dS
2 S
Z Z
2
¡ 2 ¢
= D ∇ w δ ∇ w dS − λ Ñαβ w,α δw,β dS
ZS Z S
= D ∇2 w ∇2 δw dS − λ Ñαβ w,α δw,β dS
S S

where Ñαβ is considered as constant under the variation. Similarly, the second
variation of the potential energy reads:
Z Z
¡ ¢
δ 2 Π = D δ ∇2 w ∇2 δw dS − λ Ñαβ δw,α δw,β dS (304)
S
Z Z S
2 2
D ∇ δw ∇ δw dS − λ Ñαβ δw,α δw,β dS
S S
Raleigh-Ritz Quotient Application of the Trefftz condition for invertability,
δ 2 Π = 0, determines the load intensity:
R
D S ∇2 δw ∇2 δw dS
λ= R (305)
S Ñαβ δw,α δw,β dS
Here, choose a trial function for w:
w=Aφ (306)
where A is the undetermined magnitude, and φ = φ̂ (x, y) is a normalized shape
function. Then, the variation of the trial function reads:
δw = δA φ (307)
Now, the load intensity reads:
R
D S ∇2 (δA φ) ∇2 (δA φ) dS
λ= R (308)
S Ñαβ (δA φ) ,α (δA φ) ,β dS
R
D S δA ∇2 φ δA ∇2 φ dS
=R
S Ñαβ δA φ,α δA φ,β dS
R
D S ∇2 φ ∇2 φ dS
=R
S Ñαβ φ,α φ,β dS

64

The Raleigh-Ritz quotient is defined as:

R
D S ∇2 φ ∇2 φ dS
λ= R (309)

S Ñ αβ φ,α φ,β dS

Example 4 As a special case, consider 1-D case:

¯ ¯
¯1 0¯
Ñαβ = ¯¯ ¯¯
00

then, the Raleigh-Ritz quotient becomes:

R

D
S ∇2 φ ∇2 φ dS
λ =
R 2
S (φ,x ) dS
Example 5 Similarly, consider 2-D compression case:
¯ ¯
¯1 0¯
Ñαβ = δ αβ = ¯¯ ¯¯
01

then, the Raleigh-Ritz quotient becomes:

R

D
S ∇2 φ ∇2 φ dS
λ =
R

S φ,α φ,α dS
R

D
S ∇2 φ ∇2 φ dS
=
R
h i
2 2
S (φ, x ) + (φ,y ) dS

4.3.5 Ultimate Strength of Plates

The onset of buckling stress σ cr does not necessarily means the total collapse of the
plate. Usually, there is redistribution of stresses, and the plate takes additional
load until the ultimate strength σ u is reached.
Von Karman Analysis of the Effective Width For a simply supported plate,
the buckling load is:
4π 2 D
Pcr = (310)
b
4π 2 E h3
=
12 (1 − ν 2 ) b
and the corresponding buckling stress is:
Pcr
σ cr = (311)
bh
µ ¶2
4π 2 E h

=
2
12 (1 − ν ) b

µ ¶2
π2 E h

=
2
3 (1 − ν ) b

65

 The normalization of buckling stress by the yield stress reads:

µ ¶2
σ cr π2 E h
= 2
(312)
σy 3 (1 − ν ) σ y b
(1.9)2
= σ y ¡ b ¢2
E h

µ ¶2
σ cr 1.9
= (313)
σy β

where β is a non-dimensional parameter defined by:

r
σy b
β= (314)
E h
The relation between the normalized buckling stress versus β is plotted in the next
figure.

1.9

On further loading the plate beyond σ cr , a greater proportion of the load is

taken by the regions of the plate near the edge. Von Karman assumed that these
edge regions, each of the width bef f /2, carry the stress up to the yield while the
center is stress free.

66

 σ (y)
σ

x
b
y
Before Buckling After Buckling

beff beff
2 2
σy σy

σ (y)

Actual Stress von Karman Model

The edge zones are at yield, i.e. σ cr /σ y = 1, but the width of the effective
portion of the plate is unknown:

σ cr (1.9)2
= ³ ´2 = 1 (315)
σy σy bef f
E h

from which one obtains:

s
E
bef f = 1.9 h (316)
σy

Taking, for example, E/σ y = 900 for mild steel, one gets:
√
bef f = 1.9 900 = 57h (317)

This is somehow high, but there is not much difference from the empirically deter
mined values of bef f = 40h ∼ 50h.

67
 The total force at the point of ultimate load is:

P = σ y bef f h (318)

Now, the average ultimate stress can be calculated:

P
σu = (319)
bh
bef f
= σy
b
hp
= 1.9 Eσ y
b
The average ultimate stress can be normalized by the yield stress:
h
p
σu Eσ y
= 1.9 b (320)
σy σy

σu 1.9
= (321)
σy β

The average ultimate stress is plotted with respect to β in the next figure.

1.9

Comparison of the ultimate and buckling load solution is shown in the next
figure.

68

 1.9

Thick plates yield Thin plates buckle

before buckle. before reaching
ultimate load.

Under the uniaxial loading, the relation between an applied load and the cor
responding displacement is schematically shown all the way to collapse in the next
figure.
Px

x
Pu
collapse Px Px
or crash
Pcr post-buckling

ux
pre-buckling y

u
x

Empirical Formulas

• Foulkner correction
σu 2 1
= − 2 (322)
σy β β

69

 • Gerard (Handbook of elastic stability)

às !0.85
σu gh2 E
= 0.56 (323)
σy A σy

where g is the sum of the number of cuts and the number of flanges after the
cuts, A is the cross sectional area A = bh, and the coefficients 0.56 and 0.85
are empirical constants.

Example 6 Consider a plate which has one cut and two flanges.
b

Then,
g =1+2 =3
Now,
às !0.85
σu 3h2 E
= 0.56
σy bh σ y
às !0.85
h E
= 1.42
b σy
1.42
=
β 0.85

Modifications in Codes In the original von Karman formula, the effective width
ratio reads: r
bef f σ cr
= (324)
b σy
In the ANSI specification, imperfection is considered:
r µ r ¶
bef f σ cr σ cr
= 1 − 0.218 (325)
b σy σy

In UK specification, the effective width ratio is defined as:

" µr ¶4 #−0.2
bef f σ cr
= 1 + 14 − 0.35 (326)
b σy

70

 eff

A comparison of theoretical (von Karman) and experimental

predictions for the effective width in compressed steel plates

4.3.6 Plastic Buckling of Plates

Stocky plates with low b/h ratio will yield before buckling at the point B. After
additional load, the plate will deform plastically on the path BC until conditions
are met for the plate to buckle in the plastic range.

A 1.9

Stowell’s Theory for the Buckling Strain Stowell developed the theory of
plastic buckling for simply-supported square plates loaded in one direction.

71
 b x

b
thickness h
y

The critical buckling strain εcr was derived by him in the form:
µ ¶
π2 h 2 h p i
εcr = 2 + 1 + 3 (Et /Es ) (327)
9 b

where the tangent modulus Et and the secant modulus Es are defined by:
dσ σ
Et = ; Es = (328)
dε ε

Et
σy

Es
ε
For the materials obeying the power hardening low:
µ ¶n
ε
σ = σr (329)
εr

where σ r and εr are the reference stress and strain. Now, the tangent and secant
modulus are:
µ ¶
dσ σ r ε n−1
Et = =n (330)
dε εr εr
µ ¶
σ σ r ε n−1
Es = =
ε εr εr

72

Substituting these expression back into the buckling equation (327), one gets:
µ ¶2
π2 h ¡ √ ¢
εcr = 2 + 1 + 3n (331)
9 b

The exponent n varies usually between n = 0 (perfectly

¡ √plastic ¢material) and n = 1
(elastic material). This makes the coefficient 2 + 1 + 3n vary in the range
3 ∼ 4. For a realistic value of n = 0.3, the buckling strain becomes:
µ ¶2
h
εcr = 3.7 (332)
b

Having determined εcr , the corresponding buckling stress is calculated from the
power law.
Approximate Solution for the Buckling Strain Consider an elastic plane
stress relation:
E
σ xx = (εxx + νεyy ) (333)
1 − ν2
E
σ yy = (εyy + νεxx )
1 − ν2
Using the solution for the pre-buckling state, i.e. σ xx = σ cr and σ yy = 0 leads:

E
σ cr = (εxx + νεyy ) (334)
1 − ν2
E
0= (εyy + νεxx )
1 − ν2
from which one gets:

εyy = −νεxx (335)

σ cr = Eεxx = Eεcr

The critical elastic buckling stress is:

µ ¶2
π2 E h
σ cr = kc (336)
12 (1 − ν 2 ) b

So, the corresponding critical elastic buckling strain read:

µ ¶2
π2 h
εcr = kc (337)
12 (1 − ν 2 ) b

Equation (337) is more general than a similar expression Eq. (327) given by
the Stowell theory because it applies to all type of boundary conditions. At the

73

same time, Stowell’s equation was derived only for the simply supported boundary
conditions. In particular, for kc = 4, Eq. (337) predicts:
µ
¶2
h
εcr = 3.6 (338)
b

which should be compared with the coefficient 3.7 of Eq. (332) in the Stowell’s
theory. For a plastic material or very high hardening exponent, the prediction of
both method are much closer.
4.3.7 Exercise 1: Effect of In-Plane Boundary Conditions, δw = 0
No Constraint in In-Plane Displacement Consider no constraint in in-plane
◦ = 0.
displacement in y-direction, Nyy

expansion due to
Poison's effect
uy
a
x

ND ND

a
ux
y

The membrane force tensor reads:

¯ ◦ ¯
¯N 0¯
◦
Nαβ = ¯¯
¯ (339)

0 0 ¯

From the constitutive equation, one gets:

Eh
£
◦ ¤

Nyy = 2
εyy + ν ε◦xx = 0 = −Nyy
◦
(340)

1−ν
=
⇒
ε◦yy = −ν ε◦xx

By applying the geometric equation between strain and the displacement and con
sidering δw = 0, here, one gets the relation between ux and uy :

∂uy ∂ux
= −ν (341)
∂y ∂x

74
Integrating both sides over the plate length leads:
Z a Z a
∂uy ∂ux
dy = −ν dx (342)
0 ∂y 0 ∂x
uy = −ν ux

Constraint in In-Plane Displacement Consider a plate fully constrained in

the y-direction, uy = 0.
D
N yy = νN D

a
x

D
ND ND N xx = ND

a
ux
y

Consequently, one also gets:

∂uy
ε◦yy = =0 (343)
∂y
Under the uniform compression, ε◦xx reads:

∂ux ux
ε◦xx = = (344)
∂x a
From the constitutive relation, the membrane forces reads:
Eh
£
◦ ¤ Eh ux
Nxx = 2
εxx + ν ε◦yy =
◦
= −Nxx (345)
1−ν 1 − ν2 a
Eh £
◦ ¤ Eh ux
Nyy = 2
εyy + ν ε◦xx =
2
ν
◦
= −Nyy
1−ν 1−ν a

Finally, the membrane force tensor reads:

¯ ◦ ¯ ¯ ¯
¯ Nxx 0 ¯ ¯ ¯
Nαβ =
¯¯

◦ ¯ = λ
¯ 1 0 ¯
(346)

0 Nyy◦ ¯
¯
0 ν
¯

where
Eh ux
λ= (347)
1 − ν2 a

75
4.3.8 Exercise 2: Raleigh-Ritz Quotient for Simply Supported Square
Plate under Uniaxial Loading
Consider a simply supported square plate subjected to uniform compressive load in
the x-direction.
a
x

ND ND

Then, the membrane force tensor reads:

¯ ¯ ¯ ¯
¯ ¯
◦ ¯1 0¯
¯1 0¯
◦
Nαβ = N ¯ ¯ = λ ¯ ¯ (348)
00 ¯0 0¯

The plate will deform into a dish, so for the trial function take the following:
³π x´ ³ π y ´
φ = φ̂ (x, y) = sin sin (349)
a a
Now, in order to obtain the load intensity, we first calculate φ,x and ∇2 φ :
π ³π x´ ³π y´
φ,x = cos sin (350)
a ³ π ax ´ ³ πay ´
π
φ,y = sin cos
a a a

³ π ´2³π x´ ³π y ´
φ,xx = − sinsin (351)
a a a
³ π ´2 ³π x´ ³π y ´
φ,yy = − sin sin
a a a

∇2 φ = φ,xx +φ,yy (352)

³ π ´2 ³π x´ ³π y ´
= −2 sin sin
a a a
³ π ´4 ³π x´ ³π y ´
∇2 φ ∇2 φ = 4 sin2 sin2 (353)
a a a

76
Now, the Raleigh-Ritz quotient is calculated:
R
◦ D S ∇2 φ ∇2 φ dS
N =λ = R 2 (354)
S (φ,x ) dS
¡ ¢4 R ¡ ¢ ¡ ¢
4D πa S sin2 πax sin2 πay dS
= ¡ ¢2 R ¡ ¢ ¡ ¢
π 2 π x sin2 π y dS
a S cos a a
³ π ´2
= 4D
a

µ ¶2
◦ π
Ncr = λ = 4D (355)
a

This is the classical buckling solution, and it is exact because of the right guess of
the displacement field.
4.4 Buckling of Sections
4.4.1 Transition from Global and Local Buckling
Euler buckling load of a simply-supported column reads:
π 2 EI
(Pcr )column = (356)
l2
where I is the bending rigidity of the column. Consider a section column which is
composed of several thin plates, then the Euler buckling load can be considered as
a global buckling load of the column.
( Pcr )column

l EI

77

 On the other hands, local buckling force of a simply-supported plate reads:

kc π 2 D
Ncr = (357)
b2
£ ¡ ¢¤
where D = Eh3 / 12 1 − ν 2 and kc = [(mb) /a + a/ (mb)]2 . Thus, the total local
buckling load can be obtained:

kc π 2 D
(Pcr )plate = (358)
b

N cr

a
b

Transition from global to local buckling can be calculated by (Pcr )column =

(Pcr )plate :

EI π2 D
π2 = kc (359)
l2 b
kc π 2 E h3
=
12 (1 − ν 2 ) b
µ ¶2
I kc l
3
= (360)
bh 12 (1 − ν 2 ) b

Example 7 Consider a square box column.

78

b
l b

Then, one obtains:

kc = 4
2
I = hb3
3
Now, the global buckling load reads:

I 2π 2 E hb3
(Pcr )column = π 2 E =
l2 3 l2
and the local buckling load from four plates can be calculated:

4kc π 2 E h3
(Pcr )f our plates =
12 (1 − ν 2 ) b

Then, applying (Pcr )column = (Pcr )f our plates , one gets:

µ ¶2
b2
' 2.20
hl

Thus, the local and global buckling loads become same when

b2 ' 1.5hl

For example, if b = 40h, then l = 60b.

Transition from the local to global buckling for an open channel section with
lips is shown in the figure below.

79

 Buckling coefficients and modes for a hat section

4.4.2 Local Buckling

The remainder of this section deals only with the local buckling. Dividing both
sides of Eq. (357) by the plate thickness b gives the expression of the buckling stress
σ cr : µ ¶2
Ncr kc π 2 E h
σ cr = = 2
(361)
h 12 (1 − ν ) b
Consider two adjacent plates of a section of the prismatic column.

P late 1

h1

b1

h2

b2 P late 2

Compatibility and equilibrium conditions at junction

of adjoining walls of a section

In general, there will be a restraining moment acting at the corner line between

80

P late 1 and P late 2. The buckling stresses for those two plates are:
µ ¶2
k1 π 2 E h1
(σ cr )1 = 2
(362)
12 (1 − ν ) b1
µ ¶2
k2 π 2 E h2
(σ cr )2 = 2
12 (1 − ν ) b2
Before buckling, stresses in the entire cross-section are the same. So, at the point
of buckling, one gets:
(σ cr )1 = (σ cr )2 (363)
from which the buckling coefficient k2 is relating to k1 :
µ ¶
h1 b2 2
k2 = k1 (364)
h2 b1
The total buckling load on the angle element is:

Pcr = (σ c )1 h1 b1 + (σ c )2 h2 b2 (365)
" µ ¶ µ ¶2 #
π2 E h1 2 h2
= k1 h1 b1 + k2 h2 b2
12 (1 − ν 2 ) b1 b2
" µ ¶ #
π2 E (h1 )3 h1 b2 2 (h2 )3
= k1 + k1
12 (1 − ν 2 ) b1 h2 b1 b2
µ ¶
π 2 Ek1 h1 2
= A
12 (1 − ν 2 ) b1
where A is the sectional area of two plate A = b1 h1 +b2 h2 . From this derivation, the
conclusion is that only one buckling coefficient is needed to calculate the buckling
load of the section consisting of several plates.
Determination of the buckling coefficient is a bit more complicated because of
the existence of the edge bending moment. This can be illustrated in an example
of a box column with a rectangular cross section with the same thickness h.
b1

b2

uniform thickness h

The wider flange will be ready to buckling first while the narrow plate is not ready

81

to buckle. When the second plate buckles, the first plate would have buckled long
before. Thus, there is an interaction between left plates, and a compromise must
be established because left plates must buckle at the same time. The buckling
coefficient as a function of the ratio b2 /b1 is plotted in the figure next.

Buckling coefficients for box sections

In the limiting case of a square box (b1 = b2 = b), k1 = 4 and the edge interactive
moment between adjacent plates is zero.

Some useful graphs and formulas for typical sections are given next.

82

83

5 Buckling of Cylindrical Shells

5.1 Governing Equation for Buckling of Cylindrical Shells
The starting point of the analysis is the strain-displacement relation for plates:
1 1
ε◦αβ = (uα ,β +uβ ,α ) + w,α w,β (366)
2 2
καβ = − w,αβ

Consider a flat plate (x, y) and a segment of a cylinder (x, aθ), where a is the radius
of a cylinder and θ is the hoop coordinate.
a

x x

y aθ

Can the strain-displacement relation for a cylinder be derived from similar re

lation for a flat plate?
x→x ; dx → dx (367)
∂ 1 ∂
y → aθ ; → (368)
∂y a ∂θ
Consider component by component. Use the notation:

u1 = ux → u
u2 = uy → v (369)
u3 → w

Then, one gets:

1
ε◦xx = u,x + (w,x )2 (370)
2
u,θ 1 ³ w,θ ´2 w
ε◦θθ = + + (371)
a 2 a a
1 ³ u,θ ´ 1 w,θ
ε◦xθ = + u,x + w,x (372)
2 a 2 a
There is a new term w/a in the expression for the hoop strain. The physical
meaning of this new term becomes clear if we consider axisymmetric deformation
with v = 0 and w independent of θ. Define the hoof strain as a relative change

84

in the length of circumference when the original circle has a radius of "a" before
deformation and a new circle has a radius of "a + w" after deformation:
2π (a + w) − 2πa w
ε◦θθ = = (373)
2πa a

a
θ w

Mathematically, Eq. (370)-(372) can be derived from its counterparts by trans

forming the rectangular coordinate system into the curvilinear coordinate system.

x = r sin θ (374)
y = r cos θ

The step-by-step derivation can be found, for example, in the book by Y.C. Fung,
"First Course in the Continuum Mechanics." The expression for curvature are
transformed in a similar way:

κxx = w,xx (375)

w,θθ
κθθ = 2
a
w,xθ
κxθ =
a
Using the variational approach explained in details for the plate problem, one can
see that the only new term in the expression for δΠ = 0 is
Z
δw
Nθθ dS (376)
S a
Therefore, the new term should be added in the equation for out-of-plane equilib
rium: µ ¶
4 1 2 1
D∇ w + Nθθ − Nxx w,xx + Nxθ w,xθ + 2 Nθθ w,θθ = q (377)
a a a
where
2 1
∇4 w = w,xxxx + 2
w,xxθθ + 4 w,θθθθ (378)
a a

85

The above equations are the nonlinear equilibrium equations for quasi-shallow cylin
drical shells. The linear equilibrium equations are obtained by omission the non
linear terms, i.e. terms in the parenthesis. The resulting equations are:

aNxx ,x +Nxθ ,θ = 0 (379)

aNθθ ,x +Nθθ ,θ = 0
1
D∇4 w + Nθθ = q
a
with
³ w´
Nxx = C u,x +ν (380)
³
w a ´

0
Nθθ = C + ν u,x
a
¡ ¢
where C is the axial rigidity, C = Eh/
1 − ν 2 .
The pre-buckling solution should
satisfy the system Eq. (379) and (380). These solutions will be denoted by Nαβ =

−Nαβ◦ .

5.1.1 Special Case I: Cylinder under Axial Load P , q = 0

The membrane forces in the pre-buckling state are:
¯ ¯
◦ P
¯¯ 1 0 ¯¯

Nαβ = (381)

2πa
¯
0 0 ¯

and the corresponding displacement field reads:

P
u (x) = − x (382)
2πahE
ν Pl
w=
E 2πah

u ( l)

l
w

aθ

It is easy to prove that the above solution satisfies all field equation.

86
5.1.2 Special Case II: Cylinder under Lateral Pressure
¯ ¯
¯0 0¯
Nαβ = Nθθ ¯¯ ¯¯
◦
(383)

01

where from the constitutive equations Eq. (380):

w
Nθθ = Eh (384)
a
Substituting Eq. (384) into Eq. (379) leads the following linear forth order inho
mogeneous ordinary differential equation for w (x):
Ehw
Dw0000 + =q (385)
a2
or in a dimensionless form:
d4 w q
4
+ 4β 4 w = (386)
dx D
where ¡
¢

4 Eh 3 1 − ν2
β = 2 = (387)

a D a
2 h4
The dimension of β is [L−1 M 0 T 0 ], so βx is dimensionless. There are four boundary
conditions for a simply supported cylinder:

Mxx = w = 0 at x=0 (388)

Mxx = w = 0 at x=l

where one gets from the moment-curvature relation:

d2 w
Mxx = −D (389)
dx2
Mθθ = 0

The general solutions of the above boundary value problem is:

q
w (x) = e−βx [c1 sin βx + c2 cos βx] + eβx [c3 sin βx + c4 cos βx] − (390)
D4 β 4
The four integration constants can be found from the boundary conditions. A
typical term of the solution is a rapidly decaying function of x.

87

 It can be conducted that the curvature and bending is confined to a narrow

boundary zone of the width xb = π/ (2β). The remainder of the shell undergoes a
uniform radial contraction:
q
w0 = − (391)
D 4β 4

For the sake of simplicity, this localized bending can be neglected (D → 0).
Then, from Eq. (379), the hoop membrane force is related to the lateral pressure
by Nθθ = qa and the pre-buckling solution is:
¯ ¯
¯ 0 0 ¯

Nαβ = qa ¯¯
¯¯

◦
(392)

01

5.1.3 Special Case III: Hydrostatic Pressure

For a cylinder subjected to the hydrostatic pressure, the total axial compressive
force is:
P = qπa2 (393)

The pre-buckling solution is:

¯ ¯
¯ 1/2 0 ¯

◦
Nαβ = qa ¯¯
¯ (394)

0 1 ¯

which is a classical membrane stress state in a thin cylinder.

88
5.1.4 Special Case IV: Torsion of a Cylinder
The pre-buckling stress in a cylinder subjected to the total torque of the magnitude
T is: ¯ ¯
◦ M ¯¯ 0 1 ¯¯

Nαβ = (395)

2πa
¯
1 0 ¯

aθ

N θx

5.2 Derivation of the Linearized Buckling Equation

We are now in the position to linearize the nonlinear buckling equation Eq. (377).
It is assumed that the state of membrane forces does not change at the point of
buckling from the pre-buckling value. Thus,
◦
Nαβ = −Nαβ (396)

and Eq. (377) becomes:

∙ ¸

4 1 ◦ 2 ◦ 1 ◦
D∇ w + Nθθ + Nxx w,xx + Nxθ w,xθ + 2 Nθθ w,θθ = q
(397)

a a a

where the hoop membrane force Nθθ , the second term in Eq. (397), depends linearly
on three component of the displacement vector (u, v, w):
µ
¶

1 w

Nθθ =
C
v,θ + + νu,x (398)

a a

89
Therefore, the out-of-plane equilibrium equation is coupled with the in-plane dis
placement (u, v) through the presence of the term Nθθ . Note that in the plate
buckling problem the in-plane and out-of-plane response was uncoupled. It is
possible to eliminate the terms involving in-plane components using the full set of
equilibrium and constitutive equations in the in-plane direction. By doing this, the
order of the governing equation has to be raised by four to give eight:
∙ ¸
8 1 − ν2 4 ◦ 2 ◦ 1 ◦
D∇ w + Cw,xxxx +∇ Nxx w,xx + Nxθ w,xθ + 2 Nθθ w,θθ = 0 (399)
a2 a a

The above equation is called the Donnell stability equation in the uncoupled form.
Not that w (x, θ) in the above equation represents additional lateral deflection over
and above those produced by the pre-buckling solution. The total deflection is a
sum of the two.
5.3 Buckling under Axial Compression
5.3.1 Formulation for Buckling Stress and Buckling Mode
We are now in a position to develop solutions to the buckling equations, Eq. (399)
for four different loading cases discussed in the previous section. Consider first
Case I of a simply-supported cylindrical shell in which the pre-buckling solution is
given by Eq. (381). In this case, Eq. (399) reduces to:

1 − ν2 P 4
D∇8 w + Cw,xxxx + ∇ w,xx = 0 (400)
a2 2πa
The buckling deflection of the shell is assumed in the following form:
³ mπx ´
w (x, θ) = c1 sin sin (nθ) (401)
l
where c1 is the magnitude, and the integer numbers (m, n) denote the number
of half-waves respectively in the axial and circumferential direction. The above
deformation satisfies both simply-supported boundary conditions at the ends, x = 0
and x = L, and periodicity conditions along the circumference.
Here, the half-length of the buckling wave is defined:
l
λ= (402)
m
It is convenient to introduce a dimensionless buckling number, m̄:
mπx ³ mπa ´ x x
= = m̄ (403)
l l a a
mπa
⇒ m̄ =
l

90

 Using the dimensionless buckling number, substituting the solution Eq. (401)
into the governing equation Eq. (400) leads:
∙ ¸ ³ x´
D¡ 2 ¢
2 4 4
¡ 2
¢ P ¡ 2 ¢
2 2 2 c1
m̄ + n + m̄ 1 − ν C − m̄ + n m̄ sin m̄ sin (nθ) = 0
a2 2πa a6 a
(404)
By setting the coefficient in the square bracket to zero, the critical buckling mem
brane force per unit length becomes:
Pcr
Ncr = (405)
2πa
¡ ¢2
D m̄2 + n2 ¡ 2
¢ m̄2
= 2 + 1 − ν C
a m̄2 (m̄2 + n2 )2
Here, by introducing the dimensionless parameter χ:
¡ 2 ¢2
m̄ + n2
χ= (406)
m̄2
Eq. (405) reads:
D ¡ ¢ 1
Ncr = 2 χ + 1 − ν 2 C (407)
a χ
The dependence of the buckling force on the parameter χ is shown in the figure
below.

91

Treating χ as a continuous variable, one can find an analytical minimum:

dNcr D ¡ ¢ 1
= 2 − 1 − ν2 C 2 = 0 (408)
dχ a χ
from which the optimum value of the parameter χ is:
r
(1 − ν 2 ) Ca2
χopt = (409)
D
ap
= 12 (1 − ν 2 )
h
a
' 3.3
h
Introducing the expression for the optimum parameter χ into Eq. (407) and using
definitions of bending and axial rigidity, one gets:

E h2
(Ncr )min = p (410)
3 (1 − ν 2 ) a

The buckling stress is obtained by dividing the critical membrane fore by the shell
thickness h:
(Ncr )min
σ cr = (411)
h
E h
=p
3 (1 − ν ) a
2

h
σ cr ' 0.605 E (412)
a

This is the classical solution for the buckling stress of a cylindrical shell subjected
to axial compression. While the buckling load is unique and does not depend on
(m̄, n), the buckling mode is not unique as:
¡ 2 ¢2
m̄ + n2 a
χopt = = 3.3 (413)
m̄2 h
There are infinity of combinations of m̄ and n that give the same expression.

Example 8 Let a/h = 134, then

¡ 2 ¢2
m̄ + n2
χopt = = 3.3 × 134 ' 442
m̄2
m̄2 + n2
⇒ ' 21
m̄

92
 p
n= 21m̄ − m̄2

5.3.2 Buckling Coefficient and Batdorf Parameter

Let us introduce the dimensionless buckling stress or the buckling coefficient kc :

hl2
kc = σ cr (414)
π2 D
Additionally, the Batdorf parameter Z is defined:
p l2
Z= 1 − ν2 (415)
ah
Using Eq. (411), the buckling coefficient becomes:

E h hl2
kc = p × 2 (416)
3 (1 − ν 2 ) a π D
12 p l2
=√ 1 − ν2
3π 2 ah
12
=√ Z
3π 2

kc ' 0.702 Z (417)

This relation between kc and Z are shown in the figure below together with two
limiting cases of very short and very long cylindrical shells. These two limiting
cases are discussed below.

93
Limiting Cases: Short Cylinders, a À l Consider a case of the following
conditions:
aÀl
a
⇒ →∞ (418)
l

Then, it is natural to assume that the number of half-waves in the axial direction
is unity:
m=1
Consequently, one gets:
a
m̄ = π →∞ (419)
l
Since m̄ is much larger than n, one also gets:

m̄2 + n2 ' m̄2 (420)

94

 Now, from Eq. (405), the membrane force reads:

P
Ncr = (421)
2πa
D ¡ ¢ 1
= 2 m̄2 + 1 − ν 2 C 2
a m̄
m̄ D2
= 2
a
π2 D
= 2
l
The buckling stress reads:
Ncr π2D
σ cr = =
h hl2
Now, from the definition of the buckling coefficient, Eq. (414), the buckling coeffi
cient for the very short cylindrical shells reads:

kc = 1

This solution indicated in previous figure as the lower bound cut-off value. The
upper bound cut-off values is given by the Euler buckling load.
Limiting Cases: Very Long Cylinders, l À a The cylindrical shell is becom
ing the Euler column. The buckling load of the column is:
π 2 EI
Pcr = (422)
l2
where I = πa3 h for cylinder sections.

95

The buckling stress reads:

Pcr
σ cr = (423)
2πah
π 2 ³ a ´2
= E
2 l
For this very long cylindrical shells, the buckling coefficient can be written:
¡ ¢ ³ a ´2
kc = 6 1 − ν 2 (424)
h

µ ¶4
a
kc = 6 Z2 (425)
l

From Eq. (411) and (423), the transition between the local shell buckling and
global thin-walled column buckling occurs when

(σ cr )shell = (σ cr )column (426)

which gives:
E h π 2 ³ a ´2
p = E (427)
3 (1 − ν 2 ) a 2 l
p
hl2 π 2 3 (1 − ν 2 )
⇒ 3 = ' 8.15 (428)
a 2
5.4 Buckling under Lateral Pressure
From the pre-buckling solution, Eq. (392), the governing equation Eq. (399) be
comes:
1 − ν2 q
D∇8 w + 2
Cw,xxxx + ∇4 w,θθ = 0 (429)
a a
Assuming the double sine buckling deflection function, similar to the case of axial
compression, the governing equation becomes:
∙ ¸ ³ x´
D¡ 2 ¢
2 4 4
¡ 2
¢ 2
¡ 2 ¢
2 2 c1
m̄ + n + m̄ 1 − ν C − qa n m̄ + n sin m̄ sin (nθ) = 0
a2 a6 a
(430)
By setting the coefficient in the square bracket to zero, the equation for the buckling
pressure becomes:
¡ ¢2
D m̄2 + n2 ¡ 2
¢ m̄4
qa = 2 + 1 − ν C (431)
a n2 n2 (m̄2 + n2 )2

96

It can be shown that the smallest eigenvalue is obtained when m = 1 or m̄ = πa/l.

With this observation, the solution becomes a function of the parameter n, or
dimensionless parameter n̄:
nl
n̄ = (432)
πa
Additionally, we define the dimensionless buckling pressure, q̄:

l2 a
q̄ = q (433)
π2 D
Now, substituting m̄, n̄ and p̄ into Eq. (431) leads:
¡ ¢2 ¡ ¢ µ ¶
1 + n̄2 1 1 − ν2 C l 2
q̄ = + (434)
n̄2 n̄2 (1 + n̄2 )2 a2 a

By introducing the Batdorf parameter Z, one gets:

¡ ¢2
1 + n̄2 1 12 2
q̄ = + 2 π Z (435)
n̄2 2 2
n̄ (1 + n̄ )

For any value of the geometrical parameter Z, there exists a preferred n̄ which
minimize the buckling pressure. Treating n̄ as a continuous variable, the optimum
n̄ can be found analytically from d¯
p/d¯
n = 0. Substituting this back into Eq. (432),
there will be a unique relation between the buckling pressure and the Batdorf pa
rameter. The solution is shown graphically in the figure below.

Limiting Cases: Very Long Cylinders, l À a In the limiting case of a long

tube (l À a), one gets:
a
m̄ = mπ → 0 (436)
l

97

In this case, the last term in Eq. (431) vanishes and the buckling pressure becomes:

n2 D
q= (437)
a3
Imagine a long cylinder consisting of a change of ring, each of the height b. The
moment of inertia of the ring along the axial axis reads:

bh3
I= (438)
12
From Eq. (437) and (438), the intensity of the line load Q = qb can be written:

Q = qb (439)
n2 Eh3
= 3 b
a 12 (1 − ν 2 )
n2 EI
= 3
a (1 − ν 2 )
The above approximation is due to Donnell. The smallest integer value is n = 1
which gives the following buckling mode.

98

 The Donnell solution should be compared with more exact solution of the ring
buckling problem which take into account a more complex incremental displacement
field with both w (θ) and v (θ). Here, a distinction should be made between the
centrally directed pressure (as in all preceding analysis) and the field -pressure
loading where pressure is always directed normal to the deformed surface.
In the later case, the ring buckling occurs at:
¡ ¢ EI
Qc = n2 − 1 3 (440)
a
where the smallest integer n = 2 so that Q = 3EI/a3 . The buckling mode has now
eight nodal points rather than four.

The solution of centrally directed pressure loading case is:

¡ 2 ¢2
n − 1 EI
Qc = (441)
n2 − 2 a3
where again the smallest n = 2. Thus the smallest buckling load intensity is:
EI
Qc = 4.5 (442)
a3
It is seen that for a realistic value n=2, the present solution, Eq. (439) gives
the buckling pressure between the two cases of field-pressure loading and centrally
directed loading, 3 < 4 < 4.5.
5.5 Buckling under Hydrostatic Pressure
Special case of the combined loading in which the total axial load P is:

P = πa2 q (443)

99

The pre-buckling solution is given by Eq. (394). The solution of the buckling equa
tion still can be sought through the sinusoidal function, Eq. (401). The optimum
solution can be found by a trial and error method varying parameters m̄ and n̄. A
graphical representation of the solution is shown in figure below.

5.6 Buckling under Torsion

A twist applied to one end of a cylindrical shell process a twisting force Nxθ :

◦ M
Nxθ = (444)
2πa
◦ = N◦ =
The other two components of the pre-buckling membrane forces vanish Nxx θθ
0. Moreover, the force is constant.

100

Under these conditions, the governing equation reduces to:

1 − ν2 2 ◦
D∇8 w + 2
C w,xxxx + Nxθ ∇4 w,xθ = 0 (445)
a a
In view of the presence of odd-ordered derivatives in the above equation, the sep
arable form of the solution for w (x, θ), assumed previously, does not satisfy the
equation.
Under torsional loading, the buckling deformation consists of circumferential
waves that spiral around the cylindrical shell from one end to the other. Such
waves can be represented by a deflection function of the form:
³ x ´
w (x, θ) = C sin m̄ − nθ (446)
a
with
mπa
m̄ = (447)
L
where m and n are integers. The alone displacement field satisfy the differential
equation and the periodical condition is the circumferential direction, but does not
satisfy any commonly used boundary at the cylinder ends. Consequently, this
simple examples can be used only for long cylinders.
For such cylinders, introduction of the additional displacement field into the
governing equation, yields:
¡ 2 ¢2
◦ m̄ + n2 D m̄3 ¡ ¢
Nxθ = 2
+ 2 1 − v2 C (448)
2 m̄ n a 2 (m̄2 + n2 ) n
For sufficiently long cylinders, the shell buckles in two circumferential waves, n = 2.
Also, the term m̄2 is small compared with 4. Then, the approximate expression is:

◦ 4 D m̄3 ¡ ¢
Nxθ = 2
+ 1 − v2 C (449)
m̄ a 64
An analytical minimization of the alone experiment with repeat to m̄ gives:
µ ¶2
4 64 h
m̄ = 2
(450)
9 (1 − ν ) a

101
Upon substitution, the final expression for the buckling force, or better, critical
shear strain causing buckling is:
µ ¶3/2
N◦ 0.272 E h
τ cr = xθ = 3/4
(451)
h 2
(1 − ν ) a

The above solution was given by Donnell. As noted, the above solution is invalid
for short shells due to the difficulties in satisfying boundary condition. A more
rigorous analytical-numerical solution is shown in the figure below.

As the radius of the shell approaches infinity, the critical stress coefficient for
simply supported and clamped edge approaches respectively the value 5.35 and 8.98
corresponding to plates under the shear loading.
5.7 Influence of Imperfection and Comparison with Experiments
Because of the presence of unavoidable imperfection in real shells, the experimen
tally measured buckling load are much smaller than the ones found theoretically.

Reduction in the
buckling strength
due to imperfection

Comparison of theoretical and experimental results for four different type of loads:

102

 • Axial compression

• Torsion

• Lateral pressure

• Hydrostatic pressure
are shown in the subsequent two pages. Note that the graphs were presented
in log − log scale. Replotting the results for axially loaded plate yields the graph
shown below.

1.1

1.0 Theory

0.8
�cr = 0.605 Eh/a
�exp

0.6

A design recommendation

0.4

0.2

0
0 500 1,000 1,500 a 2,000 2,500 3,000 3,500
h
Distribution of test data for cylinders subjected to axial compression.

Figure by MIT OCW.

The differences are shown to be very large. Design curves for cylindrical and
other shells are based on the theoretical solution modified by empirical prediction
factors called "know-down factors". For example, the solid curve shown in the
receding page is a "90 percent probability curve." For a/h = 50, the reduction
factor predicted by this probability curve is 0.24. Thus, the theoretical solution
σ cr = 0.605 Eh/a should be multiplied by 0.24 for the design stress σ = 0.15 Eh/a.
Cylindrical shells loaded in different way are been sensitive to imperfections and
the resulting knock-down factors are smaller.
Most industrial organization establish their own design criteria. They are fre
quently loose-leaf and are continuously updated on the volume of the experimented
evidence increase. The situation can be converted to the firmly established manual
103

of steel constructions for the design of columns and beam. The reason is that col
umn are not sensitive to imperfection as far as the ultimate strength is concerned.

104