Computational Fluid and Solid Mechanics

Series Editor
K.J. Bathe
Massachusetts Institute of Technology, Cambridge, MA, USA

For other titles published in this series, go to

http://www.springer.com/series/4449
Miguel Luiz Bucalem Klaus-Jürgen Bathe
•

The Mechanics of
Solids and Structures –
Hierarchical Modeling and
the Finite Element Solution

ABC
Prof. Miguel Luiz Bucalem Prof. Klaus-Jürgen Bathe
Escola Politécnica da Universidade Massachusetts Institute of Technology
de São Paulo Cambridge, MA
São Paulo, Brazil USA
mlbucale@usp.br kjb@mit.edu

ISSN 1860-482X e-ISSN 1860-4838

ISBN 978-3-540-26331-9 e-ISBN 978-3-540-26400-2
DOI 10.1007/978-3-540-26400-2
Springer Heidelberg Dordrecht London New York

c Springer-Verlag Berlin Heidelberg 2011

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting,
reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication
or parts thereof is permitted only under the provisions of the German Copyright Law of September 9,
1965, in its current version, and permission for use must always be obtained from Springer. Violations
are liable to prosecution under the German Copyright Law.
The use of general descriptive names, registered names, trademarks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective
laws and regulations and therefore free for general use.

Cover design: deblik, Berlin

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Preface

Our main objective in this book is to provide a rational, structured and

modern framework for the modeling and analysis of engineering structures.
Although engineering structures have been modeled and analyzed for cen-
turies, the mathematical models that could actually be solved were relatively
simple. This situation has dramatically changed during the last decades. To-
day, with powerful computers and reliable finite element procedures widely
available, very complex models of solids and structures can be solved, and
consequently the range and complexity of analyses has drastically increased.
In any finite element analysis, the first step for an analyst is to choose an
appropriate mathematical model, and the second step is to solve that model
using finite element procedures. In almost all analyses, the first step is most
important and also most difficult. In order to choose an appropriate model,
the analyst must be familiar with the basic mathematical models that are
available, and in particular know the hierarchy of such models. Only if the
analyst is deeply familiar with the various mathematical models available,
their hierarchy, and reliable finite element procedures, can the analyst choose
the most effective model, perform an efficient analysis, and properly interpret
the analysis results.
Many books on finite element methods have been published; however, our
aim in this book is broader. We aim to present in one treatise – both – the
basic mathematical models of solid and structural mechanics and modern
reliable finite element procedures for the solution of these models. The book
can be used for teaching, in a modern way, structural and solid mechanics, fi-
nite element methods, and for self-study – from elementary to quite advanced
material.
We draw in this treatise heavily from the material published in the books
Theory of Elasticity, by S. P. Timoshenko and J. N. Goodier (1970), and
Theory of Plates and Shells, by S. P. Timoshenko and S. Woinowsky-Krieger
(1959), regarding the mechanics of solids and structures, and from the mate-
rial in the book Finite Element Procedures, by K. J. Bathe (1996), regarding
finite element formulations and solution techniques. In essence, we try to
synthesize the presentation of the models and methods of classical mechanics
with the procedures of finite element analysis, add new insights, and show
VI Preface

how to solve the classical general models of elasticity in a modern – effective

and reliable – manner.
Towards that aim, we first develop from elementary concepts the basic
mathematical models of solids and structures, then we present modern finite
element methods for solution, and finally we give examples of applications
using the finite element program ADINA. Emphasis is given to the hierar-
chical nature of the mathematical models, from simple to complex, and on
choosing the simplest reliable and effective model for analysis, see Chapter
1. The process of hierarchical modeling and finite element solution, with the
benefits reached, is finally illustrated in the examples of Chapter 7. These
broadly indicate how we recommend modern analysis to be conducted.
To perform an effective analysis is an art. In many cases, the analyst
tries to look into the future by asking how the design of a structure will
perform; in other cases, the analyst tries to understand a phenomenon of
nature to be able to predict when and how this phenomenon will occur, how
it can be affected, and possibly be remedied. All these analyses are based on
present knowledge and the analyst endeavors to predict the future by means
of computational modeling and simulations. This clearly cannot be an easy
task.
Since any analysis depends on the knowledge of the analyst, performing
in the art of analysis is most stimulating and requires constant learning. We
hope that this book will be valuable in this learning process.
While we focus on the linear elastic static analysis of solids and structures,
with only a short introduction to nonlinear analysis in Chapter 8, the gen-
eral concepts that we present regarding modeling and analysis are equally
applicable in much more complex cases, including fluids and multi-physics
phenomena. Hence, this book not only gives valuable information regard-
ing the linear analysis of solids, but also demonstrates universal concepts of
analysis that are generally applicable.
The writing of this book required a very large effort, and we would like
to thank all of those who have supported us in this endeavor.
Miguel Luiz Bucalem is very thankful to the Department of Structural
and Geotechnical Engineering of the University of São Paulo. He was very
fortunate to first learn structural engineering from great engineers and pro-
fessors, specifically Decio Leal de Zagottis and Henrique Lindenberg Neto.
Decio Leal de Zagottis was his first mentor whose brilliant teaching and en-
couragement attracted him to the field. Since he joined the Faculty of the
Department in 1985, he has had an excellent environment for his teaching
and research and for writing this book. The interactions with his colleagues
Carlos Eduardo Nigro Mazzilli, João Cyro André and Sergio Cifu in teach-
ing subjects related to this book were very rich and helped him shape his
thoughts about some of the contents in this book. He is also very grateful
to have been the doctoral student of Klaus-Jürgen Bathe at M.I.T. Finally,
he would like to thank his students who have been a source of inspiration
 Preface VII

for his work and who have helped with this book, Daniel Carvalho Lepkison
de Oliveira, Christian Hideki Furukawa, Ivan José de Godoy Mazella, Estela
Mari Ricetti Bueno, Carlos Alberto Medeiros, Alex Neves Júnior, Krishna
Martins Leão, and Lucas Anastasi Fiorani.
Klaus-Jürgen Bathe is very thankful to the Department of Mechanical
Engineering, M.I.T. for having provided to him an excellent environment for
his teaching, research and scholarly writing. He has been fortunate to work
with many brilliant students and colleagues, including Miguel Luiz Bucalem,
who have much contributed to his research and teaching. The idea to write
this book was born when he was asked to teach the undergraduate course
in solid mechanics and he interacted with Miguel. It became apparent that
new, exciting and modern ways to teach mechanics and analysis are needed,
and we believe that this book offers one such option. K. J. Bathe is also
thankful to his students Daniel Payen and Seounghyun Ham for their help in
the proof-reading of the manuscript, and would like to acknowledge that for
his teaching and research, and work on this book, his involvement in ADINA
R & D, Inc. has been very valuable.
Finally, we both would like to express our unbounded thanks to our wives
Tamara and Zorka, and our children, for their encouragement and uncon-
ditional support over many years in all our work endeavors, including the
writing of this book.

M. L. Bucalem and K. J. Bathe

Contents

1. Mathematical models and the finite element solution. Hier-

archical modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Introductory remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Mathematical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 A demonstrative problem − a carabiner . . . . . . . . . . . . . 3
1.2.2 The case for hierarchical modeling . . . . . . . . . . . . . . . . . . 6
1.2.3 Demonstrative hierarchical modeling example − a cara-
biner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Remarks on the hierarchical modeling process . . . . . . . . . . . . . . 14
1.4 Outline of book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2. Fundamental steps in structural mechanics . . . . . . . . . . . . . . . 19

2.1 General conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.1 Motion of a deformable three-dimensional body . . . . . . 19
2.1.2 Properly supported bodies . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.1.3 Internal actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.1.4 Assumptions for static analysis . . . . . . . . . . . . . . . . . . . . . 25
2.1.5 Assumptions for a linear static analysis . . . . . . . . . . . . . 25
2.1.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2 The analysis of truss structures − to exemplify general con-
cepts of analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2.1 Model assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2.2 Kinematic conditions for a properly supported truss . . 28
2.2.3 Equilibrium conditions for a truss model . . . . . . . . . . . . 30
2.2.4 Constitutive behavior for a truss bar . . . . . . . . . . . . . . . . 34
2.2.5 Compatibility conditions for a truss . . . . . . . . . . . . . . . . . 35
2.2.6 Statically determinate and indeterminate trusses . . . . . 40
2.3 Matrix displacement method for trusses . . . . . . . . . . . . . . . . . . . 44
2.3.1 Truss bar stiffness matrix in its local system . . . . . . . . . 45
2.3.2 Solution of a two-bar truss structure using the matrix
method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.3.3 Stiffness matrix of an arbitrarily oriented truss element 52
2.3.4 Solution of the three-bar truss structure using the ma-
trix method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
X Contents

2.3.5 Systematization of the matrix formulation for truss

structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.3.6 Principle of superposition . . . . . . . . . . . . . . . . . . . . . . . . . 72
2.3.7 Remarks about the structure stiffness matrix . . . . . . . . 73
2.3.8 Strain energy of a truss structure . . . . . . . . . . . . . . . . . . . 73
2.3.9 Properly supported truss structures in the context of
the matrix method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.4 Modeling considerations for truss structures . . . . . . . . . . . . . . . . 80

3. The linear 3-D elasticity mathematical model . . . . . . . . . . . . . 83

3.1 The analysis of a steel sheet problem . . . . . . . . . . . . . . . . . . . . . . 83
3.1.1 One-dimensional conditions . . . . . . . . . . . . . . . . . . . . . . . . 84
3.1.2 Two Dimensional Conditions . . . . . . . . . . . . . . . . . . . . . . . 91
3.2 Deformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.2.1 Displacement field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.2.2 Normal and shear strains . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.2.3 Finite and infinitesimal rigid deformations . . . . . . . . . . . 113
3.2.4 Technical or engineering notation for the strains . . . . . . 119
3.2.5 Deformation in the vicinity of a point . . . . . . . . . . . . . . . 120
3.3 Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.3.1 Classical concept of stress . . . . . . . . . . . . . . . . . . . . . . . . . 126
3.3.2 Characterization of the state of stress at a point . . . . . . 128
3.3.3 Differential equilibrium equations . . . . . . . . . . . . . . . . . . . 136
3.3.4 Principal stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
3.3.5 Principal strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
3.3.6 Infinitesimally small displacements . . . . . . . . . . . . . . . . . 149
3.3.7 Technical or engineering notation for the stresses . . . . . 150
3.4 Constitutive equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
3.4.1 Hooke’s law for three-dimensional isotropic material
conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
3.4.2 Relation between G and E, ν . . . . . . . . . . . . . . . . . . . . . . 155
3.4.3 Generalized Hooke’s law for an isotropic material in
matrix notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
3.5 Formulation of the linear elasticity problem . . . . . . . . . . . . . . . . 159
3.6 Torsion of a prismatic bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

4. Mathematical models used in engineering structural analy-

sis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
4.1 Plane elasticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
4.1.1 The plane strain model . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
4.1.2 The plane stress model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
4.1.3 The axisymmetric model . . . . . . . . . . . . . . . . . . . . . . . . . . 197
4.2 Bar models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
4.2.1 Prismatic bar subjected to axial loading . . . . . . . . . . . . . 210
 Contents XI

4.2.2 Prismatic bar subjected to transverse loading; the

Bernoulli-Euler beam model. . . . . . . . . . . . . . . . . . . . . . . . 217
4.2.3 Bar models obtained by an assemblage of bars . . . . . . . 239
4.2.4 Matrix displacement method for frames . . . . . . . . . . . . . 245
4.2.5 Bars subjected to 3-D actions . . . . . . . . . . . . . . . . . . . . . . 270
4.2.6 Thin walled bars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
4.2.7 Curved bar model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
4.2.8 The Timoshenko beam model . . . . . . . . . . . . . . . . . . . . . . 301
4.3 Plates in bending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
4.3.1 The Kirchhoff plate bending model . . . . . . . . . . . . . . . . . 307
4.3.2 The Reissner-Mindlin plate bending model . . . . . . . . . . 320
4.4 Shells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
4.4.1 Geometrical preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . 328
4.4.2 Shell mathematical models . . . . . . . . . . . . . . . . . . . . . . . . 330
4.4.3 Shells of revolution loaded axisymmetrically . . . . . . . . . 335
4.4.4 Remarks on shell modeling of engineering structures . . 353
4.5 Summary of the mathematical models for structural mechanics354

5. The principle of virtual work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367

5.1 The principle of virtual work for the bar problem . . . . . . . . . . . 367
5.2 The principle of virtual work in 2-D and 3-D analyses . . . . . . . 380
5.2.1 The principle of virtual work for 3-D elasticity . . . . . . . 381
5.2.2 The principle of virtual work for the plane stress model 385
5.2.3 The principle of virtual work for the plane strain model388
5.2.4 The principle of virtual work for the axisymmetric
model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
5.2.5 The principle of virtual work for the Bernoulli-Euler
beam model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
5.3 Strain and potential energy in 3-D . . . . . . . . . . . . . . . . . . . . . . . 390
5.3.1 Strain energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
5.3.2 The total potential energy . . . . . . . . . . . . . . . . . . . . . . . . . 391

6. The finite element process of solution . . . . . . . . . . . . . . . . . . . . . 395

6.1 Finite element formulation of 1-D bar problem . . . . . . . . . . . . . 395
6.2 Convergence properties of 1-D finite element solutions . . . . . . . 415
6.2.1 Convergence conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
6.2.2 Distances and norms of functions . . . . . . . . . . . . . . . . . . . 421
6.2.3 Convergence properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
6.2.4 Smoothness of the solution . . . . . . . . . . . . . . . . . . . . . . . . 433
6.2.5 The h, p and h − p finite element methods . . . . . . . . . . . 433
6.3 Displacement-based finite element formulation for solids . . . . . 434
6.3.1 Discretization methodology . . . . . . . . . . . . . . . . . . . . . . . 434
6.3.2 Equilibrium properties of the finite element solutions
for 2-D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
6.3.3 Isoparametric finite elements . . . . . . . . . . . . . . . . . . . . . . . 451
XII Contents

6.3.4 Numerical integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456

6.3.5 Displacement-based finite element formulations for 3-
D solids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
6.3.6 Framework to formulate displacement-based finite el-
ements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461
6.4 Finite elements for beams, plates and shells . . . . . . . . . . . . . . . . 461
6.4.1 Bernoulli-Euler beam finite element . . . . . . . . . . . . . . . . 462
6.4.2 Matrix and finite element structural analysis for frames 464
6.4.3 Plate and shell finite elements . . . . . . . . . . . . . . . . . . . . . . 468
6.4.4 Beam finite elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473
6.5 Effective finite elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476
6.5.1 Convergence of displacement-based finite element for-
mulations and the effects of locking . . . . . . . . . . . . . . . . . 476
6.5.2 Locking for the 2-node Timoshenko beam element . . . . 479
6.5.3 Locking in a general setting . . . . . . . . . . . . . . . . . . . . . . . . 485
6.6 Finite element modeling tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489
6.6.1 Combining different type of elements in the same model489
6.6.2 Constraint equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490
6.6.3 Rigid links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492
6.6.4 Spring elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494
6.6.5 Model symmetries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495
6.6.6 Substructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497
6.6.7 Connecting different type of elements . . . . . . . . . . . . . . . 499
6.6.8 Skew systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
6.7 Meshing issues and error assessment . . . . . . . . . . . . . . . . . . . . . . 502
6.7.1 Mesh grading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503
6.7.2 Element shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503
6.7.3 Error estimation and adaptative procedures . . . . . . . . . . 505
6.8 Finite element model construction . . . . . . . . . . . . . . . . . . . . . . . . 509
6.9 A finite element modeling example . . . . . . . . . . . . . . . . . . . . . . . . 511

7. Hierarchical modeling examples . . . . . . . . . . . . . . . . . . . . . . . . . . 519

7.1 Built-in cantilever subjected to a tip load . . . . . . . . . . . . . . . . . . 519
7.1.1 Bernoulli-Euler beam model . . . . . . . . . . . . . . . . . . . . . . . 520
7.1.2 Timoshenko beam model . . . . . . . . . . . . . . . . . . . . . . . . . 521
7.1.3 Plane stress solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523
7.2 Machine Tool Jig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540
7.2.1 Beam Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
7.2.2 The Shell Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
7.2.3 Three-Dimensional Elasticity Model . . . . . . . . . . . . . . . . 545
7.2.4 Qualitative Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546
7.2.5 Quantitative Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549
7.3 Modeling of a carabiner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
7.3.1 Straight bar model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
7.3.2 Curved bar model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
 Contents XIII

7.3.3 Three-dimensional elasticity model . . . . . . . . . . . . . . . . . 553

7.3.4 Qualitative analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554
7.3.5 Quantitative analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555

8. Modeling for nonlinear analysis. An aperçu . . . . . . . . . . . . . . . 559

8.1 Sources of nonlinearity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559
8.2 Incremental formulation for nonlinear analysis . . . . . . . . . . . . . . 568
8.3 Determination of ultimate loads leading to structural collapse 578
8.4 Modeling nonlinear problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593
1. Mathematical models and the finite element
solution. Hierarchical modeling

The objective of this chapter is to introduce a modern approach for the

modeling of structures (and in fact any physical system). The hierarchical
modeling process is the central concept. We discuss the need for such a pro-
cess, detail it in a demonstrative problem and examine the important benefits
associated with its usage in engineering. Mathematical modeling and finite
element methods are introduced as natural ingredients of the hierarchical
modeling procedure.
While the discussion in this chapter and in those to follow focuses on the
analysis of solids and structures, we also point out that the basic ideas of
the hierarchical modeling approach are directly applicable to the analysis of
fluid flows, to the solution of multi-physics and multi-scale problems, and in
fact to the analysis of any problem encountered in engineering and in the
sciences.

1.1 Introductory remarks

Historically, the design of engineering structures has progressed through a
number of stages. Engineering structures have been built much before any
rational method of analysis was available. Then a structural design was based
on trial and error and as experience was gathered for a particular type of
structure, practical guidelines were developed which helped new designs.
The development of mathematical methods and experimental approaches
have drastically altered and improved structural design methodologies. Nowa-
days, the design process is aided by several factors: a vast experience with
previous designs of similar structures, modern methods of analysis and, of
course, still experiments.
It is easy to recognize that structural design is a vast field, and involves
many architectural, engineering and societal issues that we encounter in the
design of bridges, buildings, airplanes, motor cars, ships, bio-medical devices,
any household appliances, power plants, jet engines, and so forth. However,
in order to focus the discussion on the issues that we would like to address in
this chapter, we assume that a design for a structure has been proposed. Of
course, many issues should have been addressed before a structural design is
reached. The major considerations are linked to structural safety, operational
2 1. Mathematical models and the finite element solution. Hierarchical modeling

requirements, construction/manufacturing feasibility and cost. During this

design process, the most important challenge that the structural engineer
faces is to give reliable answers to the following questions:
• Is the structure safe to be used?
• Will the structure perform in operation as required?
Of course, additional questions arise depending on the type of structure
considered. Nevertheless, the questions above highlight the main concern of
a structural engineer.
If we consider the broad spectrum of engineering structures, to build
prototypes and small-scale models for experiments can be very inefficient. Of
course experimental structural analysis is a very important field and physical
experiments in the laboratory or on prototype structures constitute usually
an important step to assure that a design is safe and functional. However, to
reach a satisfactory design it is, in many industries, common practice today,
to perform finite element analyses of proposed designs – prior to actually
building the structure or a physical (laboratory) model thereof.

1.2 Mathematical models

This book is concerned with methods that permit to predict the behavior of a
structure without requiring that the structure has already been built. Even if
the structure already exists, such methods do not demand physical experi-
mentation with the structure. It is important to appreciate that to have a
method with this characteristic constitutes a major step since it implicitly
assumes that we are able to construct an abstract representation of a struc-
ture which can capture its structural behavior. In other words, for a given
structure we need to represent its geometry, materials, the loads acting on it
and the governing relations linking these quantities, all of which allows us to
simulate its structural behavior. In general, we are interested in predicting
the deformations and internal forces of the structure subjected to various
loadings and whether it would “break” or not.
The abstract representation of the structure which allows us to make pre-
dictions of its behavior is called a mathematical model of the structure at
hand. We summarize the process described above in Figure 1.1. Note that we
intentionally refer to a physical problem in this figure to place mathematical
modeling into a wide perspective. Our physical problem will be a structural
problem and its modeling constitutes the main objective of this book. How-
ever, the discussion that we will undertake is also valid in the broad context of
predicting the behavior of other, very general, physical systems, for example
involving also fluids and the interactions of structures and fluids.
To further consider this point, the structures that we primarily think of
analyzing are of course the traditional structures like bridges, motor cars,
 1.2 Mathematical models 3

Fig. 1.1. Concepts involved in the construction of a mathematical model

airplanes, that we see around us in everyday life. We are aiming to predict

the stiffness and strength characteristics of these structures. However, our
discussion is also directly applicable to the analysis of structures of much
smaller and much larger scales, see e.g. Bathe, 2005. For example, on the
much smaller scale, we may design a new composite material based on micro-
mechanical considerations, or analyze, on the nano-scale, cells moving in the
human body. On the much larger scale, geological problems like the eroding of
beaches and the construction of tunnels through mountains may be analyzed,
or in weather forecasting, the development and movement of storms and
their effect on structures may be simulated. It is apparent that all these
“structures” can be exposed to important multi-physics events, that may
involve fluid flows, and thermal, chemical and electro-magnetic effects. In
such cases, the full coupling between very different media, that need to be
modeled at different scales, might need to be considered.
While we do not discuss in this book the analysis of fluid flows, multi-
physics and multi-scale problems, it is important to note, though, that the
basic concepts of hierarchical modeling − as we expose them in detail in
this book − are directly applicable to the analysis of such problems as well.
Hence the basic modern approach of analysis presented here is very general
and indeed can be followed in any field of analysis.
In order to further develop our discussion of modeling, we now will fo-
cus on an example analysis that we use to exemplify the basic concepts of
hierarchical modeling.

1.2.1 A demonstrative problem − a carabiner

A photograph of the chosen “structure” is shown in Figure 1.2. It is called

a carabiner and it is used in rock climbing. A schematic representation of
its usage is shown in Figure 1.3a. From a structural point of view, the cara-
biner should transfer the load between ropes without “breaking” and without
deforming “too much”, i.e., the deformations must not jeopardize its perfor-
4 1. Mathematical models and the finite element solution. Hierarchical modeling

mance. Therefore, an important objective for the mathematical modeling of

this structure is to determine the magnitude of the load that can safely be
transferred by the carabiner.

Fig. 1.2. Photograph of a carabiner

Fig. 1.3. Schematic representation of the carabiner usage and of a situation that
makes it to work open

Let us focus on the situation when the carabiner works open which could
arise as schematically shown in Figure 1.3b. In order to discuss the modeling
of the carabiner we need to use some structural mechanics terminology and
concepts but do so in an introductory manner. We will revisit this problem
in much more detail in Chapter 7.
The task at hand is to construct a mathematical model for the carabiner
which allows us to predict the carabiner’s structural response. Hence, there
are some choices to be made that, although interconnected, can be organized
as follows:
 1.2 Mathematical models 5

• How to represent the loads.

• How to represent the restrictions to motion (the displacement boundary
conditions).
• How to represent the actual structure.
These are addressed below.

How to represent the loads

The load is transferred from the lower rope to the carabiner by the contact
between the rope and the carabiner. A simplified way of modeling this action
is by a concentrated load applied to the carabiner whose magnitude represents
the load applied by the rope.

How to represent the restrictions to motion

We note that the motion is not clearly restricted at any specific point of
the structure. Actually, the opposing ropes and the carabiner find an equi-
librated position. Hence to restrict the motion we may consider the final
equilibrated position. At one side the load is imposed as explained above and
at the opposite side we prevent the motion. Of course, the reactions associ-
ated with restricting the motion equilibrate the applied load, see Figure 1.4.

Fig. 1.4. Model representation of loading and boundary conditions

We note that since we are considering the carabiner to be open, Figure

1.4 shows only the part which is structurally relevant.