Finite Element Methods for Flow Problems

Finite Element Methods for Flow Problems. Jean Donea and Antonio Huerta
Copyright  2003 John Wiley & Sons, Ltd.
ISBN: 0-471-49666-9
 Finite Element
Methods for Flow Probelms

Jean Donea and Antonio Huerta

WILEY
Copyright ©2003 John Wiley & Sons Ltd, The Atrium, Southern Gate. Chichester,
West Sussex PO19 8SQ, England

Telephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected]

Visit our Home Page on www.wileyeurope.com or www.wiley.com

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning
or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the
terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London
WIT 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should
be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate,
Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (+44)
1243 770571.

This publication is designed to provide accurate and authoritative information in regard to the subject
matter covered. It is sold on the understanding that the Publisher is not engaged in rendering
professional services. If professional advice or other expert assistance is required, the services of a
competent professional should be sought.

Other Wiley Editorial Offices

John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA

Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA

Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany

John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809

John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print
may not be available in electronic books.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN 0-471-49666-9

Typeset from pdf files supplied by the author

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire
This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which
at least two trees are planted for each one used for paper production.
 Contents

Preface xi

1 Introduction and preliminaries 1

1.1 Finite elements in fluid dynamics 1
1.2 Subjects covered 2
1.3 Kinematical descriptions of the flow field 4
1.3.1 Lagrangian and Eulerian descriptions 5
1.3.2 ALE description of motion 8
1.3.3 The fundamental ALE equation 11
1.3.4 Time derivative of integrals over moving
volumes 12
1.4 The basic conservation equations 13
1.4.1 Mass equation 13
1.4.2 Momentum equation 13
1.4.3 Internal energy equation 15
1.4.4 Total energy equation 17
1.4.5 ALE form of the conservation equations 18
1.4.6 Closure of the initial boundary value problem 19
1.5 Basic ingredients of the finite element method 19
1.5.1 Mathematical preliminaries 19
Vi CONTENTS

1.5.2 Trial solutions and weighting functions 21

1.5.3 Compact integral forms 23
1.5.4 Strong and weak forms of a boundary value
problem 23
1.5.5 Finite element spatial discretization 27

2 Steady transport problems 33

2.1 Problem statement 33
2.1.1 Strong form 33
2.1.2 Weak form 35
2.2 Galerkin approximation 36
2.2.1 Piecewise linear approximation in 1D 37
2.2.2 Analysis of the discrete equation 40
2.2.3 Piecewise quadratic approximation in ID 45
2.2.4 Analysis of the discrete equations 47
2.3 Early Petrov—Galerkin methods 50
2.3.1 Upwind approximation of the convective term 50
2.3.2 First finite elements of upwind type 51
2.3.3 The concept of balancing diffusion 53
2.4 Stabilization techniques 59
2.4.1 The SUPG method 60
2.4.2 The Galerkin/Least-squares method 63
2.4.3 The stabilization parameter 64
2.5 Other stabilization techniques and new trends 65
2.5.1 Finite increment calculus 66
2.5.2 Bubble Junctions and wavelet approximations 67
2.5.3 The variational multiscale method 68
2.5.4 Complements 70
2.6 Applications and solved exercises 70
2.6.1 Construction of a bubble function method 70
2.6.2 One-dimensional transport 72
2.6.3 Convection—diffusion across a source term 74
2.6.4 Convection—diffusion skew to the mesh 75
2.6.5 Convection—diffusion-reaction in 2D 76
2.6.6 The Hemker problem 78

3 Unsteady convective transport 79

3.1 Introduction 79
3.2 Problem statement 81
 CONTENTS vii

3.3 The method of characteristics 32

3.3.1 The concept of characteristic lines 82
3.3.2 Properties of the linear convection equation 84
3.3.3 Methods based on the characteristics 87
3.4 Classical time and space discretization techniques 91
3.4.1 Time discretization 92
3.4.2 Galerkin spatial discretization 94
3.5 Stability and accuracy analysis 98
3.5.1 Analysis of stability by Fourier techniques 99
3.5.2 Analysis of classical time-stepping schemes 101
3.5.3 The modified equation method 105
3.6 Taylor—Galerkin Methods 107
3.6.1 The need for higher-order time schemes 107
3.6.2 Third-order explicit Taylor—Galerkin method 108
3.6.3 Fourth-order explicit leap-frog method 113
3.6.4 Two-step explicit Taylor—Galerkin methods 114
3.7 An introduction to monotonicity-preserving schemes 117
3.8 Least-squares-based spatial discretization 120
3.8.1 Least-squares approach for the 6 family of
methods 121
3.8.2 Taylor least-squares method 122
3.9 The discontinuous Galerkin method 124
3.10 Space—time formulations 126
3.10.1 Time-discontinuous Galerkin formulation 126
3.10.2 Time-discontinuous least-squares formulation 128
3.10.3 Space-timeGalerkin/Least-squaresformulation 128
3.11 Applications and solved exercises 129
3.11.1 Propagation of a cosine profile 129
3.11.2 Travelling wave package 133
3.11.3 The rotating cone problem 135
3.11.4 Propagation of a steep front 137

Compressible Flow Problems 147

4.1 Introduction 147
4.2 Nonlinear hyperbolic equations 149
4.2.1 Scalar equations 149
4.2.2 Weak solutions and entropy condition 152
4.2.3 Time and space discretization 156
4.3 The Euler equations 159
Viii CONTENTS

4.3.1 Strong form of the conservation equations 159

4.3.2 The quasi-linear form of the Euler equations 161
4.3.3 Basic properties of the Euler equations 163
4.3.4 Boundary conditions 165
4.4 Spatial discretization techniques 166
4.4.1 Galerkin formulation 167
4.4.2 Upwind-type discretizations 168
4.5 Numerical treatment of shocks 176
4.5.1 Introduction 176
4.5.2 Early artificial diffusion methods 178
4.5.3 High-resolution methods 180
4.6 Nearly incompressible flows 186
4.7 Fluid-structure interaction 187
4.7.1 Acoustic approximation 189
4.7.2 Nonlinear transient dynamic problems 191
4.7.3 Illustrative examples 196
4.8 Solved exercises 199
4.8.1 One-step Taylor—Galerkin solution of Burgers'
equation 199
4.8.2 The shock tube problem 202

5 Unsteady convection—diffusion problems 209

5.1 Introduction 209
5.2 Problem statement 210
5.3 Time discretization procedures 211
5.3.1 Classical methods 211
5.3.2 Fractional-step methods 213
5.3.3 High-order time-stepping schemes 215
5.4 Spatial discretization procedures 222
5.4.1 Galerkin formulation of the semi-discrete
scheme 222
5.4.2 Galerkin formulation of 0 family methods 224
5.4.3 Galerkin formulation of explicit Pade schemes 228
5.4.4 Galerkin formulation of implicit multistage
schemes 229
5.4.5 Stabilization of the semi-discrete scheme 231
5.4.6 Stabilization of multistage schemes 233
5.5 Stabilized space-time formulations 241
5.6 Solved exercises 243
 CONTENTS IX

5.6.1 Convection-diffusion of a Gaussian hill 243

5.6.2 Transient rotating pulse 246
5.6.3 Steady rotating pulse problem 250
5.6.4 Nonlinear propagation of a step 250
5.6.5 Burgers' equation in 1D 251
5.6.6 Two-dimensional Burgers' equation 252
Appendix Least-squares in transient/relaxation problems 254

Viscous incompressible flows 265

6.1 Introduction 265
6.2 Basic concepts 267
6.2.1 Strain rate and spin tensors 267
6.2.2 The stress tensor in a Newtonian fluid 268
6.2.3 The Navier—Stokes equations 269
6.3 Main issues in incompressible flow problems 272
6.4 Trial solutions and weighting functions 273
6.5 Stationary Stokes problem 275
6.5.1 Formulation in terms of Cauchy stress 275
6.5.2 Formulation in terms of velocity and pressure 278
6.5.3 Galerkin formulation 279
6.5.4 Matrix problem 281
6.5.5 Solvability condition and solution procedure 283
6.5.6 The LBB compatibility condition 284
6.5.7 Some popular velocity—pressure couples 285
6.5.8 Stabilization of the Stokes problem 287
6.5.9 Penalty method 288
6.6 Steady Navier—Stokes problem 293
6.6.1 Weak form and Galerkin formulation 293
6.6.2 Matrix problem 293
6.7 Unsteady Navier—Stokes equations 294
6.7.1 Weak formulation and spatial discretization 295
6.7.2 Stabilized finite element formulation 296
6.7.3 Time discretization by fractional-step methods 297
6.8 Applications and solved exercices 306
6.8.1 Stokes flow with analytical solution 306
6.8.2 Cavity flow problem 307
6.8.3 Plane jet simulation 313
6.8.4 Natural convection in a square cavity 317
 Preface
This book is based on lectures regularly taught in the fifth year of engineering diplomas
or basic graduate courses at the University of Liege and at the Universitat Politecnica
de Catalunya. The goal of the book is not to provide an exhaustive account of finite
elements in fluids, which is an extremely active area of research. The objective is
to present the fundamentals of stabilized finite element methods for the analysis of
steady and time-dependent convection—diffusion and fluid dynamics problems with an
engineering rather than a mathematical bias. Organized into six chapters, it combines
theoretical aspects and practical applications and covers some of the recent research
in several areas of computational fluid dynamics.
The authors wish to thank their colleagues and friends at the Joint Research Cen-
tre of the European Commission, Ispra, Alenia, Torino, ENEL Research Hydraulics
and Structures, Milano, Politecnico di Milano, Universite de Liege and Universi-
tat Politecnica de Catalunya, who contributed to the development of several of the
concepts presented in this book and provided worked examples. We are especially
grateful to Folco Casadei, Luigi Quartapelle and Vittorio Selmin who reviewed parts
of this book, provided numerous suggestions and corrections, and contributed with
several illustrative examples. We are particularly in debt to our colleagues at the
Laboratori de Calcul Numeric: Sonia Fernandez-Mendez, Esther Sala-Lardies, An-
tonio Rodriguez-Ferran, Pedro Diez, Josep Sarrate, Agusti Perez-Foguet and Xevi
Roca. Their reviews, suggestions and contributions have made this book a reality.
The second author must also thank Gilles Pijaudier-Cabot and the Ecole Centrale de
Nantes for giving him sanctuary during the last stages of this project.
Above all, however, our thanks go to our wives, Marie-Paule and Sara, for their
continuous and unfailing support and patience for many years and particularly during
the writing of this book.

J. DONEA AND A. HUERTA

XI