Structural Analysis.

Civil Engineering Unit 2

STRUCTURAL ANALYSIS

2. STRAIN ENERGY AND DISPLACEMENTS IN STRUCTURES

2.1 WORK AND STRAIN ENERGY

If external forces are applied to a body, it will deform until there is balance between the
applied external forces and the internal forces of the body. The external forces generate
a work (𝑊) and this work is transformed into energy, which is composed of Strain Energy
( 𝑈) and Kinetic Energy (𝐾) (neglecting the energy that is converted into heat: Thermal
Energy). Kinetic Energy is a function of the body's mass and the acceleration experienced
by the particles. If the forces are applied gradually, then Kinetic Energy is null (static
behavior).

Consider a material body, which can be view as a collection of material particles. The
particles of a deformable body can be subjected to internal and external forces. If the body
is subjected to 𝑛 external point forces 𝐅 𝑖 that displace the points of action by
displacements ∆𝐫𝑖 , then the external work (𝑊) is the sum of the work done by individual
forces is (McGuire et al., 2000; Hibbeller 2017):
𝑛

𝑊 = − ∑ 𝐅 𝑖 ∙ ∆𝐫𝑖 (2.1)
𝑖=1

The minus sign indicates that the work is expended on the body as opposed to the work
stored in the body.

The body can be subjected to a distributed force (measure per unit volume) acting in a
body with volume Ω and its particles can experience a displacement 𝐮 = 𝐮(𝐱). The work1
done on one particle is 𝐟 ∙ 𝐮. The total work done on the body is the sum of work done on
all particles occupying the body:

1
Work = External Work

1
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

𝑊 = − ∫ (𝐟 ∙ 𝐮) 𝑑Ω (2.2)
Ω

To get further understanding on the concept of Work, consider a bar with an end fixed and
the other end free (Fig. 2.1a). This bar is subjected to a load P that produces a
displacement 𝛿 at the end of the bar. If the load is gradually applied, a load-displacement
curve can be obtained (Fig. 2.1b). The External Work (𝑊) produced by the load P can be
calculated using two formulas:

• If the load is instantaneously applied:

𝑊 = −𝑃𝛿 (2.3)

• If the load is gradually applied:

𝑊 𝑔 = − ∫ 𝑃 𝑑𝛿 (2.4)

If the material of the bar is elastic, then there will be a linear relationship between the load
𝑃 and the displacement 𝛿. In this case, Eq. 2.2 yield in Ec. 2.5:

1
𝑊 𝑔 = − 𝑃𝛿 (2.5)
2

(a) Bar subjected to a load P (b) Curve load-displacement

Figure 2.1. Work produced by a load P

2
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

In Fig. 2.1b there are two areas (grey and white). W is the area under the curve (grey
area) and corresponds to the force P acting through a change of displacement d (Eq.
2.4, 𝑊 𝑔𝑟𝑎𝑑 ). The white area C corresponds to the complementary work. If the material is
linear, then C=W. It is important to notice that in the textbooks, the term Work and the
nomenclature W can be used for both definitions of Work established in this lecture notes
(Eqs. 2.3 and 2.4). In addition, the term W* is used for the complementary work.

Internal forces are generated inside the body due to the applied loads and are related to
the displacements of the particles. The strains in a body are determined from the
displacements. For example, the bar from Fig. 2.1 has an axial deformation (𝑒):

𝑒 = 𝑢|𝑥=𝐿 − 𝑢|𝑥=0 = 𝛿 − 0 = 𝛿 (2.6)

The strain (𝜀) of the bar is the axial deformation (𝑒) divided by the length (𝐿):

𝑒 𝛿
𝜀= = (2.7)
𝐿 𝐿

The stress (𝜎) can be determined using the constitutive equations that establishes a
relationship between stresses and strains. If the bar is linear elastic, the stress (𝜎) in the
bar is the strain (𝜀) times the modulus of elasticity (𝐸):

𝜎 = 𝐸𝜀 (2.8)

The Strain Energy (𝑈) of an elastic bar is:

1
𝑈= ∫ 𝜎𝜀 𝑑Ω (2.9)
2 Ω

If the bar is defined in a 1D coordinate system 𝑥 and has a cross-sectional area 𝐴 and
length 𝐿, the Strain Energy 𝑈 can be calculated as:

1 𝐿
𝑈= ∫ 𝜎𝜀 𝐴𝑑𝑥 (2.10)
2 0

The Total Potential Energy (Π) of the bar is:

Π=𝑈+𝑊 (2.11)

3
Dr. Luis Enrique Fernández Baqueiro
 Structural Analysis. Civil Engineering Unit 2

Clapeyron's theorem states that the Potential Energy2 of an elastic body, which is in
equilibrium under a given load, is equal to half of the Work done by the external forces
computed assuming these forces had remained constant from the initial state to the final
state:

1
Π= 𝑊 (2.12)
2

Remember that the Work (𝑊) has a negative sign, then Total Potential Energy (Π) is
negative.

From Eqs. 2.11 and 2.12, it can be established the Law of Conservation of Energy, which
states that the work done (gradually) by all the external forces acting on a structure
(𝑊 𝑔𝑟𝑎𝑑 ), is transformed into internal work or strain energy (𝑈), which is developed when
the structure deforms. The Law of Conservation of Energy can be stated mathematically
as:

𝑈 = −𝑊 𝑔 (2.13)

1 1
𝑈 = − 𝑊 = 𝑃𝛿 (2.14)
2 2

From Eqs. 2.10 and 2.13, it can be observed that the Strain Energy (𝑈) is always positive.
Π, 𝑊, 𝑈 are scalars.

Class activity:

Consider the slab (floor) of the second story of a building. Draw the history of loads as
𝑃 − 𝛿; 𝛿 is the deflection at the center of the slab. Draw the dead load as 𝑊 (rectangular
area) and the live loads as 𝑊 𝑔 (triangular area). The live loads are produced by the
students.

2
Total Potential Energy = Potential Energy

4
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

2.2 STRAIN ENERGY DENSITY

Consider a very small cubic particle in a coordinate system x (𝑥, 𝑦, 𝑧). The sides of the
element are parallel to the coordinate axes (𝑥, 𝑦, 𝑧). The size of this particle is 𝑑𝑉 =
∆𝑥∆𝑦∆𝑧 (Fig. 2.2). The stresses that can be acting on this particle are (Timoshenko and
Goodier, 1970):

• Normal stresses 𝜎𝑥𝑥 , 𝜎𝑦𝑦 , 𝜎𝑧𝑧

• Shear stresses 𝜎𝑥𝑦 , 𝜎𝑥𝑧 , 𝜎𝑦𝑥 , 𝜎𝑦𝑧 , 𝜎𝑧𝑥 , 𝜎𝑧𝑦 ; (𝜏𝑖,𝑗 variable can be used for shear
stresses)

Given that the particle is in equilibrium, then six shear stresses can be simplified to three:

𝜎𝑥𝑦 = 𝜎𝑦𝑥 𝜎𝑥𝑧 = 𝜎𝑧𝑥 𝜎𝑦𝑧 = 𝜎𝑧𝑦 (2.15)

Figure 2.2. Differential volume (particle)

Figure 2.3 shows the components of the stress 𝜎𝑖𝑗 . The index 𝑖 (subscript letter) indicates
the direction of the normal to the plane under consideration and the index 𝑗 indicates the
direction of the component of the stress. For example, 𝜎𝑥𝑦 (or 𝜏𝑥𝑦 ) is the stress in the 𝑥
plane (or 𝑦 − 𝑧 plane) in the 𝑦 direction. Note that there are some shear stresses that are
not drawn in Fig. 2.3 for clarity purposes.

5
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

Figure 2.3. Stresses acting on a particle

The particle can have strains:

• Normal strains 𝜀𝑥𝑥 , 𝜀𝑦𝑦 , 𝜀𝑧𝑧

• Shear strains 𝛾𝑥𝑦 , 𝛾𝑥𝑧 , 𝛾𝑦𝑥 , 𝛾𝑦𝑧 , 𝛾𝑧𝑥 , 𝛾𝑧𝑦 ; (𝛾𝑖,𝑗 = 2𝜀𝑖,𝑗 )

Given that the particle is in equilibrium (Eq. 2.15), then six shear strains can be simplified
to three. The strains can be calculated from the displacements (u) (Reddy 2002;
Timoshenko and Goodier 1970):
𝜕𝑢 𝜕𝑣 𝜕𝑤
𝜀𝑥𝑥 = 𝜕𝑥 𝜀𝑦𝑦 = 𝜕𝑦 𝜀𝑧𝑧 = (2.16)
𝜕𝑧

𝜕𝑢 𝜕𝑣 𝜕𝑣 𝜕𝑤 𝜕𝑤 𝜕𝑢 (2.17)
𝛾𝑥𝑦 = 𝜕𝑦 + 𝜕𝑥 𝛾𝑦𝑧 = 𝜕𝑧 + 𝜕𝑦 𝛾𝑧𝑥 = + 𝜕𝑧
𝜕𝑥

Class activity:

1. Draw a 2D particle with stresses.

2. Draw a 2D deformed particle for normal strains and shear strains to explain Eqs.
2.16 and 2.17.

The strain energy density 𝑈0 (Strain energy per unit volume) for a linear elastic material
is:

6
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

1
𝑈0 = (𝜎 𝜀 + 𝜎𝑦𝑦 𝜀𝑦𝑦 + 𝜎𝑧𝑧 𝜀𝑧𝑧 + 𝜎𝑥𝑦 𝛾𝑥𝑦 + 𝜎𝑦𝑧 𝛾𝑦𝑧 + 𝜎𝑧𝑥 𝛾𝑧𝑥 ) (2.18)
2 𝑥𝑥 𝑥𝑥

The strain energy density 𝑈0 is the area under a stress-strain curve. The complementary
strain energy density 𝑈0∗ is the area under a strain-stress curve.

Class activity: Draw the stress-strain curve of concrete up to the compressive strength
(f’c) and identify 𝑈0 and 𝑈0∗ . Graph 𝜎 − 𝜀 in the 𝑦 − 𝑥 axes. Draw again this curve up to 0.6
f’c (linear elastic range). Observe that if the material is linear, then 𝑈0 = 𝑈0∗ ; if the material
is not linear, then 𝑈0 ≠ 𝑈0∗.

The internal energy stored in a body is the strain energy 𝑈 and can be determined as:

1
𝑈 = ∭ (𝜎𝑥𝑥 𝜀𝑥𝑥 + 𝜎𝑦𝑦 𝜀𝑦𝑦 + 𝜎𝑧𝑧 𝜀𝑧𝑧 + 𝜎𝑥𝑦 𝛾𝑥𝑦 + 𝜎𝑦𝑧 𝛾𝑦𝑧 + 𝜎𝑧𝑥 𝛾𝑧𝑥 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.19)
2

Class activities:

1) Simplify Eq. 2.19 to the 2D case. Consider that the body has a constant thickness
𝑡.
2) Demonstrate that the following equation (Eq. 2.20) is valid for the element
described in Fig. 2.1a. Consider that the frame element is linear elastic and has a
constant cross section area. Eq. 2.20 is an application of Clapeyron's theorem
(Eqs. 2.12 and 2.14).

1 1 1
𝑈= 𝜎𝑥𝑥 𝜀𝑥𝑥 (𝐴𝐿) = − 𝑊 = 𝑃𝛿 (2.20)
2 2 2

7
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

2.3 STRAIN ENERGY IN STRUCTURAL ELEMENTS

2.3.1 AXIAL LOAD

Consider a bar directed along an axis 𝑥 with a modulus of elasticity 𝐸, cross section area
𝐴 and length 𝐿. The bar has a fixed end and the other end is free (Fig. 2.4). At the free
end an axial load 𝑁 is gradually applied. This load produces a displacement 𝛿 in the
direction of the 𝑥 axis. The axial load 𝑁 only produces stresses 𝜎𝑥𝑥 . Based on Eq. 2.19,
the strain energy of this bar can be calculated as:

1
𝑈𝑁 = ∭ (𝜎𝑥𝑥 𝜀𝑥𝑥 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.21)
2

The stress 𝜎𝑥𝑥 and the strain 𝜀𝑥𝑥 can be determine as:

𝑁
𝜎𝑥𝑥 = (2.22)
𝐴
𝜎𝑥𝑥 (2.23)
𝜀𝑥𝑥 =
𝐸

In Eq. 2.23, 𝐸 is the modulus of elasticity.

Figure 2.4. Bar subjected to an axial force.

8
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

By substituting Eqs. 2.22 and 2.23 into Eq. 2.21, the strain energy of a bar subjected to
axial load can be determined as:

1 𝑁 𝑁
𝑈𝑁 = ∭ ( 𝐸 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.24)
2 𝐸𝐴 𝐸𝐴

Given that 𝐸 is constant, that 𝐴 and 𝑁 are not function of 𝑦 and 𝑧, and the area of the bar
is 𝐴 = ∫ 𝑑𝑦𝑑𝑧, Eq. 2.24 can be simplify as:
𝐿
𝑁2
𝑈𝑁 = ∫ ( ) 𝑑𝑥 (2.25)
0 2𝐸𝐴

2.3.2 BENDING MOMENT

Consider a beam directed along an axis 𝑥 with a modulus of elasticity 𝐸, moment of inertia
𝐼 and length 𝐿. The beam is gradually subjected to a bending moment 𝑀 about the 𝑧 axis
(Fig. 2.5). The stresses and strains at any cross section caused by that bending moment
about the 𝑧 axis are directed along the 𝑥 axis of the member: 𝜎𝑥𝑥 and 𝜀𝑥𝑥 , respectively.
Based on Eq. 2.19, the strain energy of this beam can be calculated as:

1
𝑈𝑀 = ∭ (𝜎𝑥𝑥 𝜀𝑥𝑥 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.26)
2

The stress 𝜎𝑥𝑥 and the strain 𝜀𝑥𝑥 in this beam can be determine as:

𝑀𝑦
𝜎𝑥𝑥 = (2.27)
𝐼
𝜎𝑥𝑥 (2.28)
𝜀𝑥𝑥 =
𝐸

In Eq. 2.27 𝑦 is the distance from the centroid and 𝐼 is the moment of inertia of the section
about the 𝑧 axis and is calculated as:

𝐼 = ∬ 𝑦 2 𝑑𝑦𝑑𝑧 (2.29)

9
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

Figure 2.5. Beam subjected to a bending moment.

By substituting Eqs. 2.27 – 2.29 into Eq. 2.25, the strain energy of a beam subjected to
bending moment can be determined as:

1 𝑀𝑦 𝑀𝑦
𝑈𝑀 = ∭ ([ ] [ ]) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.30)
2 𝐼 𝐼𝐸

Given that 𝐸 is constant, that 𝑀 is not function of 𝑦 and 𝑧, and the moment of inertia of a
beam can be calculated with Eq. 2.28, then Eq. 2.29 can be simplify as:
𝐿
𝑀2
𝑈𝑀 = ∫ 𝑑𝑥 (2.31)
0 2𝐸𝐼

2.3.3 SHEAR FORCE

Consider a beam directed along an axis 𝑥 with a shear modulus 𝐺, a cross section area
𝐴 and length 𝐿. The beam is gradually subjected to shear forces 𝑉 in the 𝑦 axis direction
(Fig. 2.6). The stresses and strains at any cross section caused by that shear force are:
𝜏𝑥𝑦 and 𝛾𝑥𝑦 , respectively. Based on Eq. 2.19, the strain energy of this beam can be
calculated as:

1
𝑈𝑉 = ∭ (𝜏𝑥𝑦 𝛾𝑥𝑦 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.32)
2

10
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

The shear stress 𝜏𝑥𝑦 of the beam can be determined as:

𝑉𝑄
𝜏𝑥𝑦 = (2.33)
𝐼𝑏

In Eq. 2.31, 𝑄 is the first moment of area or static moment of area, 𝐼 is the moment of
inertia, 𝑏 is the width of the beam.

The shear strain 𝛾𝑥𝑦 for a linear elastic material can be calculated as:

𝜏𝑥𝑦 𝑉𝑄
𝛾𝑥𝑦 = = (2.34)
𝐺 𝐺𝐼𝑏

The shear modulus 𝐺 is determined as:

𝐸
𝐺= (2.35)
2(1 + 𝜈)

In Eq. 2.35, 𝜈 is the Poisson ratio.

Figure 2.6. Beam subjected to shear forces.

By substituting Eqs. 2.33 and 2.34 into Eq. 2.32, the strain energy of a beam subjected to
shear forces can be determined as:

1 𝑉𝑄 𝑉𝑄
𝑈𝑉 = ∭ ([ ] [ ]) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.36)
2 𝐼𝑏 𝐺𝐼𝑏

Simplifying Eq. 2.36

11
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

1 𝑉𝑄 2 1
𝑈𝑉 = ∭ ([ ] [ ]) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.37)
2 𝐼𝑏 𝐺

Substituting the radius of gyration 𝑟 (Eq. 2.39) into Eq. 2.37:

𝑉2 𝑄2
𝑈𝑉 = ∭ ( 2 2 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.38)
2𝐴𝐺 𝑟 𝐼𝑏

(2.39)
𝐼
𝑟=√
𝐴

In Eq. 2.38, the term in parenthesis is the shape factor 𝐶𝑘 (Eq. 2.40). This shape factor
has been calculated for different types of cross section, such as solid rectangle (𝐶𝑘 = 6⁄5),

solid circle (𝐶𝑘 = 10⁄9), thin-wall circle (𝐶𝑘 = 2), I shape (𝐶𝑘 = 𝐴⁄𝐴 ).
𝑤𝑒𝑏

𝑄2
𝐶𝑘 = ∬ 2 2 𝑑𝑦𝑑𝑧 (2.40)
𝑟 𝐼𝑏

By substituting Eq. 2.40 into Eq. 2.38, the strain energy of a beam subjected to bending
moment can be determined as:
𝐿
𝑉2
𝑈𝑉 = ∫ 𝐶𝑘 𝑑𝑥 (2.41)
0 2𝐺𝐴

2.3.4 TORSION

Consider a frame element directed along an axis 𝑥 with a shear modulus 𝐺, polar moment
of inertia 𝐽 and length 𝐿. The frame is gradually subjected to torsion 𝑇 about the 𝑥 axis
(Fig. 2.7). This torsion causes shear stress 𝜏 and shear strains 𝛾. The frame element has
a circular cross section. Based on Eq. 2.19, the strain energy of this frame element can
be calculated as:

1
𝑈𝑇 = ∭ (𝜏𝛾) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.42)
2

The shear stress 𝜏 of the frame element can be determined as:

12
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

𝑇𝑐
𝜏= (2.43)
𝐽

In Eq. 2.43, 𝑐 is the distance to the centroid.

The shear strain 𝛾 for a linear elastic material can be calculated as:

𝜏 𝑇𝑐
𝛾= = (2.44)
𝐺 𝐺𝐽

Figure 2.7. Frame element subjected to Torsion.

By substituting Eqs. 2.43 and 2.44 into Eq. 2.42, the strain energy of a beam subjected to
shear forces can be determined as:

1 𝑇𝑐 𝑇𝑐
𝑈𝑇 = ∭ ([ ] [ ]) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.45)
2 𝐽 𝐺𝐽

Simplifying Eq. 2.45:

1 𝑇 2𝑐2
𝑈𝑇 = ∭ ( 2 ) 𝑑𝑥𝑑𝑦𝑑𝑧 (2.46)
2 𝐺𝐽

Given that the polar moment of inertia 𝐽 can be calculated as:

13
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

𝐽 = ∬ 𝑐 2 𝑑𝐴 (2.47)

Substitute Eq. 2.47 into Eq. 2.46 to obtain 𝑈𝑇 :

𝐿
𝑇2
𝑈𝑇 = ∫ 𝑑𝑥 (2.48)
0 2𝐺𝐽

2.3.4 FRAME ELEMENT SUBJECTED TO DIFFERENT INTERNAL FORCES

The strain energy of a frame element subjected to different internal forces is calculated as
the addition of the strain energy due to axial force, bending moment, shear force and
torsion (Gere 2015; Goodno and Gere, 2017):

𝑈 = 𝑈𝑁 + 𝑈𝑀 + 𝑈𝑉 + 𝑈𝑇 (2.49)
𝐿 𝐿 𝐿 𝐿
𝑁2 𝑀2 𝑉2 𝑇2 (2.50)
𝑈=∫ 𝑑𝑥 + ∫ 𝑑𝑥 + ∫ 𝐶𝑘 𝑑𝑥 + ∫ 𝑑𝑥
0 2𝐸𝐴 0 2𝐸𝐼 0 2𝐺𝐴 0 2𝐺𝐽

Class activity: Consider a cantilever beam subjected to a uniform load w. Determine the
strain energy due to bending moments and shear forces.

2.4 MAXWELL’S AND BETTI’S THEOREM

Maxwell developed the Force Method of analysis and published a theorem that relates
the flexibility coefficients of any two points on an elastic structure. This theorem is referred
to as the theorem of reciprocal displacements and may be stated as follows:

The displacement of a point B on a structure due to a unit load acting at point A is

equal to the displacement of point A when a unit load is acting at point B; that is,
𝑓𝐴𝐵 = 𝑓𝐵𝐴 .

Maxwell’s theorem was later formalized in a more general sense by Betti. The theorem of
Betti is referred as the reciprocal work theorem and may be stated as follows for a linear
elastic structure subjected to two sets of forces (𝑃𝑖 and 𝑄𝑖 ):

14
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

The virtual work 𝑃𝑖 𝛿𝑖𝑗 done by a system of forces 𝑄𝑗 is equal to the virtual work
𝑄𝑗 𝛿𝑗𝑖 done by a system of forces 𝑃𝑖 .

To demonstrate Betti’s theorem, consider two sets of forces (𝑃𝑖 and 𝑄𝑖 ) gradually applied
to a body. If the set of forces 𝑃𝑖 is applied first and the set of forces 𝑄𝑗 is applied second,
then the work is

1 1 (2.51)
𝑊 𝑔 = 𝑃𝑖 𝛿𝑖𝑖 + 𝑄𝑗 𝛿𝑗𝑗 + 𝑃𝑖 𝛿𝑖𝑗
2 2

In Eq. 2.51, 𝛿𝑖𝑖 are the displacements in 𝑖 produced by 𝑃𝑖 , 𝛿𝑗𝑗 are the displacements in 𝑗
produced by 𝑄𝑗 , and 𝛿𝑖𝑗 are the displacements in 𝑖 produced by 𝑄𝑗 .

If the set of forces 𝑄𝑗 is applied first and the set of forces 𝑃𝑖 is applied second, then the
work is

1 1 (2.52)
𝑊𝑔 = 𝑃𝑖 𝛿𝑖𝑖 + 𝑄𝑗 𝛿𝑗𝑗 + 𝑄𝑗 𝛿𝑗𝑖
2 2

In Eq. 2.52, 𝛿𝑗𝑖 are the displacements in 𝑗 produced by 𝑃𝑖 .

Given that the work of an elastic body is the same independently of the order the loads
are applied, Eq. 2.51 is equal to Eq. 2.52 and the Bettis’s theorem is obtained:

𝑃𝑖 𝛿𝑖𝑗 = 𝑄𝑗 𝛿𝑗𝑖 (2.53)

Class activities:

Illustrate the implications of Maxwell’s and Betti’s theorems applied to a simply supported
beam with two loads.

Figure 2.8. Simply supported beam with two loads.

15
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

2.5 DEFORMED SHAPE OF A STRUCTURE

In the next two sections, two methods to calculate the displacement at any point of a
structure are presented. It is important to realize that after calculating the value of a linear
or angular displacement at a point of a structure, it is recommended to compare that result
with the deformed shape of the structure and verify if the results are consistent. This
means that the student must be able to draw the deformed shape ("elastic") of a structure
that is subjected to loads.

In order to draw the deformed shape of a structure it is important to consider:

1. The boundary conditions (types of supports); for example: fixed, pin, roller.
2. The type and magnitude of loads.
3. In the case of beams and frames, the nodes are rigid. Therefore, all the elements
that are connected to a node experiment the same displacements.
4. Neglect the axial deformation to draw the deformed shape.
5. If the system is too complex, try to simplify the problem or apply the superposition
principle.

Class activity: Draw deformed shapes of structures. Consider all the concepts listed
before.

2.6 CASTIGLIANO’S THEOREM

The Italian engineer Alberto Castigliano published a book in which he outlined a method
for determining the deflection or slope at a pint in an elastic structure. This method is
referred as Castigliano’s second theorem, or the method of least work (Hibbeler, 2017).
This theorem states that “the derivative of the strain energy (𝑈) in a structure with respect
to one of the force acting at the point and the direction of the displacement, is equal to the
displacement”. By applying this theorem, the linear displacement 𝑢 and the angular

16
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

displacement (rotation) 𝜑 at any point of the structure can be determined with Eqs. 2.54 y
2.55, respectively.

𝜕𝑈 (2.54)
𝑢=
𝜕𝑃
𝜕𝑈 (2.55)
𝜑=
𝜕𝑀𝑃

In Eq.2.54 𝑃 is the load applied at the point where the linear displacement is to be
determined. In Eq.2.55 𝑀𝑃 is the moment applied at the point where the angular
displacement is to be determined.

The strain energy of a frame element subjected to different internal forces is calculated
with Eq. 2.50. By substituting Eq. 2.50 into Eq. 2.54 it is obtained Eq. 2.57 to determine
the linear displacement at any point of the structure.

𝜕𝑈 𝜕 𝐿 𝑁2 𝜕 𝐿 𝑀2 𝜕 𝐿 𝑉2 𝜕 𝐿 𝑇2 (2.56)
𝑢= = ∫ 𝑑𝑥 + ∫ 𝑑𝑥 + ∫ 𝐶𝑘 𝑑𝑥 + ∫ 𝑑𝑥
𝜕𝑃 𝜕𝑃 0 2𝐸𝐴 𝜕𝑃 0 2𝐸𝐼 𝜕𝑃 0 2𝐺𝐴 𝜕𝑃 0 2𝐺𝐽
𝐿 𝐿 𝐿 𝐿
𝜕𝑁 𝑁 𝜕𝑀 𝑀 𝜕𝑉 𝑉 𝜕𝑇 𝑇 (2.57)
𝑢=∫ ( ) 𝑑𝑥 + ∫ ( ) 𝑑𝑥 + ∫ (𝐶𝑘 ) 𝑑𝑥 + ∫ ( ) 𝑑𝑥
0 𝜕𝑃 𝐸𝐴 0 𝜕𝑃 𝐸𝐼 0 𝜕𝑃 𝐺𝐴 0 𝜕𝑃 𝐺𝐽

The procedure to determine the displacement with Castigliano’s second theorem method
is:

1) Apply a force 𝑃, if a linear displacement is to be determined, or a moment 𝑀𝑃 , if an

angular displacement is to be determined, in the point where a linear or angular
displacement will be calculated.
2) Determine the internal forces equations (axial force N, shear force V, bending
moment M or torsion T) as a function of force 𝑃 or moment 𝑀𝑃 .
𝜕𝑁 𝜕𝑉 𝜕𝑀
3) Derive the internal forces equations with respect force 𝑃 or moment 𝑀𝑃 : 𝜕𝑃 , 𝜕𝑃, ,
𝜕𝑃
𝜕𝑇
.
𝜕𝑃

17
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

𝜕𝑁 𝜕𝑉 𝜕𝑀 𝜕𝑇
4) Substitute 𝑁, 𝑉, 𝑀, 𝑇, , , , in Eq. 2.57. The force 𝑃 or moment 𝑀𝑃 are
𝜕𝑃 𝜕𝑃 𝜕𝑃 𝜕𝑃

fictitious, then it can be set equal zero. If the displacement is positive, then this is
in the direction of the force 𝑃.

2.7 UNIT LOAD METHOD

The Unit Load Method is used to determine displacements in structures. This method
comes from the Principle of Virtual Work, so it is also called the Virtual Work Method. The
principle of Virtual Work states that (Gere, 2015): “If a structure in equilibrium, the total
virtual work of forces acting on the structure is zero for any virtual displacement”. The
virtual work is the work done by a real force acting through a virtual displacement or a
virtual force acting through a real displacement. A virtual displacement is any
displacement consistent with the constraints of the structure; i.e. a virtual displacement
must satisfy the boundary conditions at the supports.

The total virtual work is the sum of the internal virtual work and the external virtual work.
In consequence, the internal virtual work is equal to the external virtual work.

In order to formulate the Unit Load Method, consider two load systems (Fig. 2.9). The first
load system is the real load system, which correspond to the actual load system. The
second one is the virtual load system, which correspond to a unit load applied in the point
where the displacement will be determined.

The real load system produce internal forces in the structure: axial force (𝑁𝑅 ), bending
moment (𝑀𝑅 ), shear force (𝑉𝑅 ) and torsion (𝑇𝑅 ). The unit load system produce internal
forces in the structure: axial force (𝑁𝑈 ), bending moment (𝑀𝑈 ), shear force (𝑉𝑈 ) and torsion
(𝑇𝑈 ).

18
Dr. Luis Enrique Fernández Baqueiro
Structural Analysis. Civil Engineering Unit 2

Figure 2.9. Unit Load Method for a beam element.

The displacement in any point of the structure is determined with Eq. 2.58.
𝐿 𝐿 𝐿 𝐿
𝑁𝑈 𝑁𝑅 𝑀𝑈 𝑀𝑅 𝑉𝑈 𝑉𝑅 𝑇𝑈 𝑇𝑅 (2.58)
𝑢=∫ 𝑑𝑥 + ∫ 𝑑𝑥 + ∫ 𝐶𝑘 𝑑𝑥 + ∫ 𝑑𝑥
0 𝐸𝐴 0 𝐸𝐼 0 𝐺𝐴 0 𝐺𝐽

The procedure to determine the displacement with the Unit Load Method is:

1) Determine the internal forces equations produced by the real load system: axial
force (𝑁𝑅 ), bending moment (𝑀𝑅 ), shear force (𝑉𝑅 ) and torsion (𝑇𝑅 ).
2) Apply a unit force 𝑃 = 1, if a linear displacement is to be determined, or a unit
moment 𝑀 = 1, if an angular displacement is to be determined, in the point where
a linear or angular displacement will be calculated.
3) Determine the internal forces equations produced by the unit load system: axial
force (𝑁𝑈 ), bending moment (𝑀𝑈 ), shear force (𝑉𝑈 ) and torsion (𝑇𝑈 ).
4) Substitute 𝑁𝑅 , 𝑀𝑅 , 𝑉𝑅 , 𝑇𝑅 , 𝑁𝑈 , 𝑀𝑈 , 𝑉𝑈 , 𝑇𝑈 in Eq. 2.58. If the displacement is positive,
then this is in the direction of the unit force.

19
Dr. Luis Enrique Fernández Baqueiro