Department of Civil Engineering

MARY MATHA COLLEGE OF ENGINEERING AND TECHNOLOGY

(AICTE, KU)

VII Semester B.Tech – Civil Engineering

Computer Methods of Structural Analysis (2003 Scheme)

MODULE III
FINITE ELEMENT METHOD
Lecture Notes
(BINOD.J: binodjohndas@gmail.com, 9645222964)
[You are free to contact me through mobile for clarification of doubts]

Syllabus: Introduction to Finite element method – Discretization of continuous systems

- Different steps in the Finite element analysis – Different types of elements. Plane
stress, Plain strain, 3D, Plate and shell elements –Shape functions for CST Element and
Beam &Truss elements - Element and structure stiffness matrix - Load Vector.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 1

1. Introduction to Finite element method
Definition: A finite element method (abbreviated as FEM) is a numerical technique to obtain
an approximate solution to a class of problems governed by elliptic partial differential
equations. Such problems are called as boundary value problems as they consist of a partial
differential equation and the boundary conditions. The finite element method converts the
elliptic partial differential equation into a set of algebraic equations which are easy to solve.
In engineering problems there are some basic unknowns. If they are found, the
behaviour of the entire structure can be predicted. The basic unknowns or the Field variable
which is encountered in structural mechanics is displacements. In a continuum, these
unknowns are infinite. The finite element procedure reduces such unknowns to a finite
number by dividing the solution region into small parts called elements and by expressing the
unknown field variables in terms of assumed approximating functions (Interpolating
functions/Shape functions) within each element. The approximating functions are defined in
terms of field variables of specified points called nodes or nodal points. Thus in the finite
element analysis the unknowns are the field variables of the nodal points. Once these are
found the field variables at any point can be found by using interpolation functions. After
selecting elements and nodal unknowns next step in finite element analysis is to assemble
element properties for each element. For example, in solid mechanics, we have to find the
force-displacement i.e. stiffness characteristics of each individual element. Mathematically
this relationship is of the form
[k]e {δ}e= {F}e
where [k]e is element stiffness matrix, {δ}e is nodal displacement vector of the
element and {F}e is nodal force vector. The element of stiffness matrix k ij represent the force
in coordinate direction ‘i’ due to a unit displacement in coordinate direction ‘j’. Four methods
are available for formulating these element properties viz.
1. direct approach,
2. variational approach,
3. weighted residual approach and
4. energy balance approach.
Any one of these methods can be used for assembling element properties. In solid mechanics
variational approach is commonly employed to assemble stiffness matrix and nodal force
vector (consistant loads). Element properties are used to assemble global properties/structure
properties to get system equations
[k] {δ} = {F}.
Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 2
 Then the boundary conditions are imposed. The solution of these simultaneous
equations gives the nodal unknowns. Using these nodal values additional calculations are
made to get the required values e.g. stresses, strains, moments, etc. in solid mechanics
problems.
2. Discretization of continuous systems
The process of modeling a structure using suitable number, shape and size of the
elements is called discretization. The modeling should be good enough to get the results as
close to actual behavior of the structure as possible.
3. Different steps in the Finite element analysis
The various steps involved in the finite element analysis are:
1. Select suitable field variables and the elements.
2. Discretize the continua.
3. Select interpolation functions.
4. Find the element properties.
5. Assemble element properties to get global properties.
6. Impose the boundary conditions.
7. Solve the system equations to get the nodal unknowns.
8. Make the additional calculations to get the required values.
Various steps in FEA can be illustrated using the following example of structural
mechanics. Consider a problem where the stress concentration is to be studies in the fillet
zone as shown in figure.

Since the problem is having symmetry about both x and y axes, only one quarter of the
tension strip may be considered. About the symmetric axes, transverse displacements of all
nodes are to be made zero.
The various steps involved in the finite element analysis of this problem are discussed
below:
Step 1: Four noded isoparametric element is selected for the analysis.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 3

 The four noded isoparametric element can take quadrilateral shape also as required for
elements 12, 15, 18, etc. As there is no bending of strip, only displacement continuity
is to be ensured but not the slope continuity. Hence displacements of nodes in x and y
directions are taken as basic unknowns in the problem.

Step 2: The portion to be analysed is to be discretised. Figure below shows discretised

portion. For this 33 elements have been used. There are 48 nodes. At each node
unknowns are x and y components of displacements. Hence in this problem total
unknowns (displacements) to be determined are 48 × 2 = 96.

Step 3: The displacement of any point inside the element is approximated by suitable
functions in terms of the nodal displacements of the element. For a typical element
displacements at P are
u = ΣNiui = N1u1 + N2u2 + N3u3 + N4u4
and v = ΣNivi = N1v1 + N2v2 + N3v3 + N4v4
The approximating functions Ni are called shape functions or interpolation
functions. Usually they are derived using polynomials.

Step 4: Now the stiffness characters and consistent loads are to be found for each
element. There are four nodes and at each node degree of freedom is 2. Hence degree
of freedom in each element is 4 × 2 = 8. The relationship between the nodal

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 4

 displacements and nodal forces is called element stiffness characteristics. It is of the
form
[k]e{δ}e= {F}e
For the elements under consideration, ke is 8 × 8 matrix and δe and Fe are
vectors of 8 values. In solid mechanics element stiffness matrix is assembled using
variational approach i.e. by minimizing potential energy.

Step 5: The structure is having 48 × 2 = 96 displacement and load vector components

to be determined. Hence global stiffness equation is of the form
[k] {δ} = {F}
96 × 96 96 × 1 96 × 1
Each element stiffness matrix is to be placed in the global stiffness matrix
appropriately. This process is called assembling global stiffness matrix. In this
problem force vector F is zero at all nodes except at nodes 45, 46, 47 and 48 in x
direction. For the given loading nodal equivalent forces are found and the force vector
F is assembled.

Step 6: In this problem, due to symmetry transverse displacements along AB and BC

are zero. The system equation [k]{δ} = {F} is modified to see that the solution for {δ}
comes out with the above values. This modification of system equation is called
imposing the boundary conditions.

Step 7: The above 96 simultaneous equations are solved using the standard numerical
procedures like Gauss elimination or Choleski’s decomposition techniques to get the
96 nodal displacements.

Step 8: Now the interest of the analyst is to study the stresses at various points. In
solid mechanics the relationship between the displacements and stresses are well
established. The stresses at various points of interest may be found by using shape
functions and the nodal displacements and then stresses are calculated. The stress
concentrations may be studies by comparing the values obtained at various points in
the fillet zone with the values at uniform zone, far away from the fillet.
4. Different types of elements
Based on the shapes elements can be classified as
Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 5
 1. One dimensional elements
2. Two dimensional elements
3. Axi-symmetric elements and
4. Three dimensional elements.
 One Dimensional Elements:
These elements are suitable for the analysis of one dimensional problem and may be
called as line elements also. Figure shows some of typical one dimensional element.

 Two Dimensional Elements:

We need two dimensional elements to solve two dimensional problems. Common two
dimensional problems in stress analysis are plane stress, plane strain and plate
problems. Two dimensional elements often used is three noded triangular element. It
is the first and most used element. These elements are known as Constant Strain
Triangles (CST). Figure shows a CST element.

Six noded and ten noded triangular elements are also used by the analysts. Six noded
triangular element is known as Linear Strain Triangle (LST) or as Quadratic
Displacement Triangle.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 6

 Ten noded triangular elements are known as Quadratic Strain Triangles (QST) or
Cubic Displacement Triangles. Still higher order triangles are also in use.

Rectangular elements of higher order also can be used. Figure shows a family of
Lagrange rectangle in which nodes are in the form of grid points.

Quadrilateral Elements are also used in finite element analysis. Initially quadrilateral
elements were developed by combining triangular elements. But it has taken back
stage after isoparametric concept was developed. Isoparametric concept is based on
using same functions for defining geometries and nodal unknowns.

 Axi-symmetric elements
These are also known as ring type elements. These elements are useful for the
analysis of axi-symmetric problems such as analysis of cylindrical storage tanks,
shafts, rocket nozzles. Axi-symmetric elements can be constructed from one or two

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 7

 dimensional elements. One dimensional axi-symmetric element is a conical frustum
and a two dimensional axi-symmetric element is a ring with a triangular or
quadrilateral cross section as drawn in figure below.

 Three Dimensional Elements:

Similar to the triangle for two dimensional problems tetrahedron is the basic element
for three dimensional problems. Tetrahedron is having four nodes, one at each corner.
Three dimensional elements with eight nodes are either in the form of a general
hexahedron or a rectangular prism, which is a particular case of a hexahedron. The
rectangular prism element is many times called as a brick element also. In these
elements also one can think of using higher order elements.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 8

5. Miscellaneous Possible Questions
a. Isoparametric finite element
The most common types of elements used in FEM have straight edges. To take care of
curved boundaries refined meshes are to be used when straight edged elements are
employed. Refined meshes would result in unnecessary stress concentrations. Higher
order elements also do not overcome the problem of suitably approximating curved
boundaries. The isoparametric concept helps in properly mapping the curved
boundaries. It is a concept of mapping regular triangular and rectangular elements in
natural coordinate system, to arbitrary shapes in global system as shown in Figure.

The Cartesian coordinates of a point in an element is expressed as

x = N!1x1 + N!2x2 + ……
y = N!1y1 + N!2y2 + ……
z = N!1z1 + N!2z2 + …… or in matrix form {x} = [N !] {xn}. If this shape function N!
are the same as the shape function used to represent the variation of displacement in
the element, then the elements are called isoparametric elements. And if the
geometry is a higher order shape function then it is called superparametric.
b. What is equivalent joint load
In using the element approach for rigid jointed frames, it is necessary to replace the
applied loads by their statically equivalent joint loads. For this purpose every element of
the structure is considered as a fixed ended member and the two ends of the elements are
considered as joints. The loads acting on the elements are replaced by fixed end forces
imposed by loads at two joints which the element is assumed to connect. Thus the
equivalent joint load for any member is equal and opposite to the fixed end reaction.
c. Flexibility influence coefficient

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 9

 A flexibility influence coefficient Fij is defined as the displacement at point i due to unit
load acting at point j, all other points being unloaded (All other loads removed from
structure). (Give an example)
d. Stiffness influence coefficient
A stiffness influence coefficient kij is defined as the force developed at point i due to unit
displacement impressed at point j, all other points being fixed. (Give an example.)
e. What are the convergent requirements to be satisfied by a displacement model? How are
they met? Explain the significance of each.
The finite element method provides a numerical solution to t complex problem. It may,
therefore be expected that the solution must converge to the exact solution under certain
circumstances. The convergence criteria to a displacement model are
1. The displacement function must be continuous within the element. This condition can
easily be satisfied by choosing polynomials for the displacement model.
2. The displacement function must be capable of representing rigid body displacements
of the element. That is when the nodes are given such displacements corresponding to
a rigid body motion the element should not experience any strain and hence lead to
zero forces. The constant term in the polynomials used for the displacement models
would usually ensure this conditions.
3. The displacement model must be capable of representing constant strain states within
the element. If we imagine the condition when the body or structure is divided into
smaller and snaller elements the strain will also approach constant values. Hence the
assumed displacement function should include terms for representing constant strain
states.
f. Explain the difference between plain strain problem and plain stress problem. Give
examples
g. Conforming elements ?
The first convergence criteria that is to be satisfied in a displacement model of variational
formulation is that the displacement model must be continuous within the elements, and
the displacement must be compatible between adjacent elements. Elements which meet
this first convergence criteria is known as conforming elements.
h. Constitutive matrix
Constitutive matrix is an assemblage of material properties of the given problem under
study. For example, in structural mechanical problems to determine the stresses in the
members of a structure or in a deformable solid, it is necessary to know the components
Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 10
 of stress as functions of the components of strain and vice versa. We assume that the
structures of bodies under conseration are made of elastic material that obeys Hooke’s
law. Hooke’s law propose linear relationship between stress and strain. Thus for a linear
elastic material this relation can be written as below where [C] is called constitutive
matrix.

i. What is rotation matrix or transformation matrix in direct stiffness method.

In the case of an inclined two dimensional line element the stiffness matrix will be first
calculated in the form local coordinate system. Similarly the nodal load vector due to
loads normal to the member will be the negatives of the fixed end actions with respect to
the member axes. For the assemblage of elements to form the overall equations of
equilibrium, transformation of these element properties is required to conform to the
degrees of freedom and nodal loads expressed in the global coordinate system. It is
achieved by a matrix formed by sequencing the direction cosines of the elements which is
known as the transformation matrix or rotation matrix. The force transformation relation
for a beam element can be written as {Q m} = [T] {Q}, where [T] is the transformation
matrix {Qm} is the load vector in local coordinate or member coordinate and {Q} is the
load vector in global coordinate system.
j. Geometric variance? Significance in FEM
If the variance is different in different geometric directions then the situation is referred to
as geometric variance. A finite element model has to satisfy convergence and
compatibility requirements. In addition to this one of the important considerations in
choosing proper terms in the polynomial expansion is that the element should have no
preferred direction. That is the displacement shapes will not change with a change in local
coordinate system. This property is known as geometric isotropy or geometrc invariance.
Geometric invariance is achieved if the polynomial includes all the terms, i.e., the
polynomial is a complete one.
k. What are natural coordinates, global coordinates and local coordinates. Give an example.
A local coordinate system is the one that is defined for a particular element and not

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 11

 necessarily for the entire body or structure. In local coordinate system the origin changes
for element to element. The coordinate for the entire body is called global coordinate
system. In this system all the nodes in the system are referred to a common origin. A
natural coordinate system is a local system which permits the specification of a point
within the element by a set of dimensionless numbers whose magnitude never exceeds
unity. This coordinate system generally arranged in such a way that some of the natural
coordinates have unit magnitude at primary external nodal points. For example the natural
coordinate of an one dimensional element is shown in figure below.

In the figure the natural coordinates L1 and L2 are worked out as L1 = l1/l and L2 = l2/l
l. What are the elements usually used for finite element analysis of 3D stress problem
Refer 3D elements above
m. What is the difference between hermite polynomial and hermitian polynomial necessary
for a beam element.
There are two types of interpolation functions. In the Lagrange interpolation the assumed
function takes on the same values as the given function at specified points. In the
Hermitian type of function the slopes of the function also takes the same value as the
given function at specified points. Obviously since beam is involved with slope deflection
type of deformations we need Hermitian interpolation for beam element and Lagrange
polynomial can not handle slope.
n. What are the different elements used in 2d problems
Refer 2D elements above
o. Static indeterminacy
If the unknown reactions in a structure can not be found out by the equations of static
equilibrium alone, then the structure is said to be statically indeterminate. The degree of
static indeterminacy is defined as the number of unknowns more and above the number of
equilibrium conditions available. (Give an example)
p. Compare FEM with classical methods.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 12

 In classical methods exact equations are formed and exact solutions are
obtained where as in finite element analysis exact equations are formed but
approximate solutions are obtained.
Solutions have been obtained for few standard cases by classical methods,
where as solutions can be obtained for all problems by finite element analysis.
Whenever the following complexities are faced, classical method makes the drastic
assumptions’ and looks for the solutions:
(a) Shape
(b) Boundary conditions
(c) Loading
To get the solution in the above cases, rectangular shapes, same boundary
condition along a side and regular equivalent loads are to be assumed. In FEM no
such assumptions are made. The problem is treated as it is.
When material property is not isotropic, solutions for the problems become
very difficult in classical method. Only few simple cases have been tried successfully
by researchers. FEM can handle structures with anisotropic properties also without
any difficulty.
If structure consists of more than one material, it is difficult to use classical
method, but finite element can be used without any difficulty.
Problems with material and geometric non-linearities can not be handled by
classical methods. There is no difficulty for handling non linearity in FEM.

Lecture notes on FEM, Binod J, Lr., MMCET, Paliyodu 13