INTRODUCTION 5

2.0

Nonlinear (0.025)
1.0 Nonlinear (0.05) (∆t = 0.025)
(∆t = 0.05)
Angular displacement

0.0

Linear (exact)
-1.0

-2.0

-3.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Time, t

Figure 1.2.1 Analytical and numerical solutions of the simple pendulum.

1.3 The Finite Element Method

As illustrated in the previous section, numerical methods are extremely
Copyright © 2004. Oxford University Press, Incorporated. All rights reserved.

powerful tools for engineering analysis. With the advent of computers, there
has been a tremendous explosion in the development and use of numerical
methods. Of these, the Þnite difference methods and the Þnite element method
and their variants are the most commonly used methods in the analysis of
practical engineering problems. In Þnite difference methods, derivatives of
various order are approximated using Taylor’s series. The traditional Þnite
difference methods suffer from two major drawbacks: (1) applying boundary
conditions of the gradient type requires additional approximation; (2) Þnite
difference formulas are traditionally developed for rectangular grids, making it
difficult to use them for irregular domains. Advances have been made in recent
years to overcome these drawbacks but the remedies are problem dependent.
The Þnite element method is based on the idea that every system is physically
composed of different parts and hence its solution may be represented in parts.
In addition, the solution over each part is represented as a linear combination
of undetermined parameters and known functions of position and possibly
time. The parts can differ from each other in shape, material properties, and
physical behavior. Even when the system is of one geometric shape and made
of one material, it is simpler to represent its solution in a piecewise manner.

Reddy, J. N. (2004). Introduction to nonlinear finite element analysis. Oxford University Press, Incorporated.
Created from UNICAF on 2024-05-17 01:46:33.
 6 NONLINEAR FINITE ELEMENT ANALYSIS

In recent years, generalizations of the Þnite element method have emerged

(e.g. the generalized Þnite element method and element-free methods or
meshless methods); however, in this study we limit our discussion to the
traditional Þnite element method [9].
The traditional Þnite element method is endowed with three basic features.
First, a domain of the system is represented as a collection of geometrically
simple subdomains, called Þnite elements. Second, over each Þnite element,
the unknown variables are approximated by a linear combination of algebraic
polynomials and undetermined parameters, and algebraic relations among
the parameters are obtained by satisfying the governing equations, often in
a weighted-integral sense, over each element. The undetermined parameters
represent the values of the unknown variables at a Þnite number of preselected
points, called nodes, in the element. Third, the algebraic relations from all
elements are assembled using continuity and “equilibrium” considerations.
There are several reasons why an engineer or scientist should study the
Þnite element method; these are listed below.
1. The Þnite element method is the most powerful numerical method ever
devised for the analysis of engineering problems. It is capable of handling
geometrically complicated domains, a variety of boundary conditions,
nonlinearities, and coupled phenomena that are common in practical
problems. The knowledge of how the method works greatly enhances the
analysis skill and provides a greater understanding of the problem being
solved.
2. Commercial software packages or “canned” computer programs based on
Copyright © 2004. Oxford University Press, Incorporated. All rights reserved.

the Þnite element method are often used in industrial, research, and
academic institutions for the solution of a variety of engineering and
scientiÞc problems. The intelligent use of these programs and a correct
interpretation of the output is often predicated on knowledge of the basic
theory underlying the method.
3. It is not uncommon to Þnd mathematical models derived in personal
research and development that cannot be evaluated using canned programs.
In such cases, an understanding of the Þnite element method and knowledge
of computer programming can help design programs to evaluate the
mathematical models.
The basic ideas underlying the Þnite element method are reviewed in
Chapter 2 using linear differential equations involving a single variable in one
and two dimensions. The main objective there is to introduce the terminology
and steps involved, e.g. weak formulation of differential equations over a
element, derivation of the interpolation functions, and numerical evaluation
of coefficients and so on. Readers who are familiar with the Þnite element
method as applied to linear differential equations may skip Chapter 2 and go
straight to Chapter 3.
Reddy, J. N. (2004). Introduction to nonlinear finite element analysis. Oxford University Press, Incorporated.
Created from UNICAF on 2024-05-17 01:46:33.