Finite element method Chapter 4

CHAPTER 4

FINITE ELEMENT METHOD

4.1 INTRODUCTION
The finite element method is a powerful tool for the numerical solution of a wide
range of engineering problems. The basic idea in the finite element method is to find the
solution of a complicated problem by replacing it by a simpler one. Since the actual
problem is replaced by a simpler one in finding the solution, we will be able to find only
the approximate solution rather than the exact solution. The existing mathematical tools
will not be sufficient to find the exact solution of most of the practical problems. Thus, in
the absence of any convenient method to find approximate solution of a given problem,
we have to prefer the finite element method. Moreover, in the finite element method, it
will often be possible to improve or refine the approximate solution by spending more
computational effort.
With advances in computer technology and CAD systems, complex problems
can be modeled with relative ease. Several alternative configurations can be tried out on
a computer before the first prototype is built. In the finite element method of analysis a
complex region defining a continuum is discretized into simple geometric shapes called
finite elements. The material properties and the governing relationships are considered
over these elements and are expressed in terms of unknown values at nodes of the
element. An assembly process duly considering the loading and constraints result in a
set of equations. The solution of these equations gives us the approximate behavior of
the continuum.

11
Finite element method Chapter 4

4.2 HISTORICAL BACKGROUND

Although the name of the finite element method (FEM) was given recently, the
concept has been used several centuries back. Courant appears to be the first person to
propose the finite element method. In a mathematics lecture held during 1941, he used
the principle of stationery potential energy and piecewise polynomial interpolation over
triangular sub-regions to study the saint-Venant's torsion problem [11]. The digital
computer provided a rapid means of performing many calculations involved in FEM and
made the method practically viable. By 1953 engineers had written stiffness equations in
matrix format and solved the equations by the digital computers. Most of this work took
place is aerospace industry. In early 1960, finite element method was used for
approximate solution of problems in stress analysis, fluid flow and heat transfer.
Clough coined the name finite element in 1960. The practical value of the method
was soon obvious. New elements for stress analysis applications were developed,
largely by interactions and physical argument. Along with the development of high-speed
digital computers, the application of the finite element method also progressed at a very
impressive rate. The book by Przemieniecki presented the finite element method as
applied to the solutions of stress analysis problems. In 1967, Zienkiewz and Cheung
presented the broad interpretation of the method and its applicability to any general field
problem. The finite element equations can also be derived by using a weighed residual
method such as Galerkien method or the least square approach. In 1970 studies were
concentrated on new element development, convergence studies and mathematical
equations.

4.3 GENERAL DESCRIPTION OF FEM

The finite element method is a method of piecewise approximation in which the
actual body of matter is considered as an integrated part of small elements known as
finite elements. These elements are connected with one another at the joints called
nodes or nodal points. Since the actual variations of the field variable like displacements,
stress, temperature, pressure or velocity inside the continuum are not known, the
variations of the field variable inside the finite element can be approximated by a simple
function called interpolation. The model is defined in terms of field variables at nodes.
Field equations can be written in the form of matrix equations and are solved for nodal
values of the field variables. The approximating functions define the variables through

12
 Finite element method Chapter 4

out the assemblage of finite elements. The process of finite element analysis is shown in
Fig. 4.1.

Physical Problems Change of

Physical Problem

Mathematical model
governed by differential
operations Improve
Geometry mathematical
Kinematics model
Material constants
Loading
Boundary Conditions

Finite element solution

Choice of
Finite elements
Mesh density
Solution parameters Refine mesh,
Representation of Solution
Finite Loading parameters
element Boundary Conditions etc., etc.
solution of
mathematica
l model
Assessment of accuracy of finite element
solution and mathematical model

Interpretation of results Refine analysis

Design improvements of
the structure

Fig. 4.1 The finite element analysis procedure

13
Finite element method Chapter 4

The solution of a general continuum problem by the finite element method always
follows an orderly step by step process [12]. With references to static structural problems
the step by step procedure can be stated as follows:
Step 1 Discretization of the structure
The first step in the finite element method is to divide the structure or solution
region into elements. Hence, the structure is to be modeled with suitable finite elements.
The number, size and arrangement of the elements are the input parameters.
Step 2 Selection of proper interpolation or displacement model
Since the displacement solution of a complex structure under any specified load
conditions cannot be predicted exactly, some suitable solutions within an element is
assumed to approximate the unknown solution. The assumed solution must be simple
from computational point of view, but should satisfy certain convergence requirements.
Step 3 Derivations of element stiffness matrix and load vectors
The stiffness matrix [k(e)] and the load vector {F} of element ‘e’ are derived from
the assumed displacement model by using either equilibrium conditions or a suitable
variations principle.
Step 4 Assemblage of element equations to obtain the overall equilibrium
equations
The individual element stiffness matrices and load vectors are assembled in a
suitable manner as the structure in an assemblage of these elements. The overall
equilibrium equations are formulated as
[K] {q} = {F} (4.1)
where [K] is called the assembled stiffness matrix, {F} is the vector of nodal
displacement and {F} is the vector of nodal forces for the complete structure.
Step 5 Solutions for the unknown nodal displacements
The overall equilibrium equations are modified to account for the boundary
conditions. After the incorporation of the boundary conditions, equilibrium equations can
be expressed as
[K] {q} = {F}
For linear problems, the vector q can be solved very easily. But for non linear
problems the solution is obtained in a sequence of steps, each step involving the
modifications of the stiffness matrix [K] and/or the load vector {F}

14
Finite element method Chapter 4

Step 6 Computation of element stress and stain

From the known displacement {q} the element stress and strains can be
computed by using the necessary equations of solutions or structural mechanics. The
finite element method is used to solve physical problems in engineering analysis and
design. The physical problem involved an actual structure or structural component
subjected to certain loads.
The idealization of physical problem to a mathematical model requires certain
assumptions that together lead to differential equations governing the mathematical
mode. The finite element analysis solves this mathematical mode. Since the finite
element solution technique is a numerical procedure, it is necessary to assess the
solution accuracy, if the accuracy criteria are not met, the numerical solution has to be
repeated by refining the solution parameters such as finer mesh element until a sufficient
accuracy is reached.

4.4 APPLICATIONS
Finite element method (FEM) was developed originally for the analysis of aircraft
structures. However, the general nature of its concept makes it applicable to a wide
variety of boundary value problems in engineering. The specific applications of the finite
element method are in three major categories of boundary value problem, namely
 Equilibrium or steady state of time independent problem
 Eigen value problems
 Propagation or transient problems
In equilibrium problem steady state displacement or stress distribution is found
for solid mechanics problem, temperature or heat flux distribution in the case of a heat
transfer problem.
Referring to eigen value problems natural frequencies, buckling loads and mode
shapes in solid mechanics or structural problem, stability of laminar flows in field
mechanics problem, resonance characteristics if it is an electrical circuit problem will be
found.
FEM finds its application in the field of civil engineering in carrying out the static
analysis of trusses, frames and bridges, the dynamic behavior of the structures for
obtaining natural frequencies, modes and response of structures to periodic loads. FEM
is the best tool in investigation of aircraft structures involving static analysis of wings,

15
Finite element method Chapter 4

structures of rockets and missiles, dynamic analysis, response to random loads and
periodic loads.
In solving heat conduction problems, steady state temperature distribution in
solids, fluids, transfer heat flow systems, FEM is the better method. It can be applied to
analysis of excavations, underground, dynamic analysis of dam reservoir systems, which
come under geo-mechanics.
Nuclear engineering also uses the concept, FEM in static and dynamic
characterization of its system viz., nuclear pressure vessels, containment structures and
dynamic response of reactor containment structures. In mechanical design, stress
concentration problems, stress analysis of pressure vessels, dynamic analysis of
mechanical linkages and gears can be effectively dealt using FEM. Even the biomedical
engineering applies FEM for impact analysis of skulls. The transient behavior of electro
mechanical devices comes under other fields of FEM applications.

4.5 GOVERNING EQUATION IN FEM

Large size problems handled by modern digital computers connected with the
static and dynamic linear analysis of complicated machines or structures are generally of
the form,
{M} {U} + [C] {U} + [K] {U} = {f (t)} (4.2)
where
{M} represents the global mass matrix,
[C] represents the global damping matrix,
[K]represents the global stiffness matrix,
f(t) represents force function vector in time,
{U} represents the nodal displacement vector respectively.
Case (i): when {M} and [C] are absent, [K] and f (t) are constants, results it is a static
linear problem.
Case (ii): when {M} and [C] are absent, [K] is function of U and f (t) is constant, the
result is Non-linear problem.

Case (iii): when f (t) and [C] are absent, {M} and [K] are constant, one gets an eigen
value problem.
Case (iv): when {M}, [C], [K] are constant and f (t) is a transient function of time, the
result is a transient vibration problem.

16