Triangulations, and more precisely meshes, are at the heart of many problems relating to a wide variety of scientific disciplines, and in particular numerical simulations of all kinds of physical phenomena. In numerical simulations, the functional spaces of approximation used to search for solutions are defined from meshes, and in this sense these meshes play a fundamental role. This strong link between meshes and functional spaces leads us to consider advanced simulation methods in which the meshes are adapted to the behaviors of the underlying physical phenomena. This book presents the basic elements of this vision of meshing. These mesh adaptations are generally governed by a posteriori error estimators representing an increase of the error with respect to a size or metric. Independently of this metric of calculation, compliance with a geometry can also be calculated using a so-called geometric metric. The notion of mesh thus finds its meaning in the metric of its elements.
Autorentext
Paul Louis George is Director of Research at the French Institute for Research in Computer Science and Automation (Inria) and one of the most internationally recognized experts in meshing.
Houman Borouchaki is Professor at the University of Technology of Troyes (UTT) in France. He is an expert on meshing problems, geometric modeling and applications in solid mechanics.
Frédéric Alauzet is a researcher at Inria, both with particular expertise in meshing adaptation, error estimators, resolution methods (advanced solvers in fluid mechanics) and remeshing methods.
Adrien Loseille is a researcher at Inria, both with particular expertise in meshing adaptation, error estimators, resolution methods (advanced solvers in fluid mechanics) and remeshing methods.
Patrick Laug is a researcher at Inria with particular expertise in geometric modeling and the generation of curve and surface meshes.
Loïc Maréchal, a long-time collaborator at Inria as an engineer, is an essential reference on hexahedra.
Inhalt
Foreword ix
Introduction xi
Chapter 1. Metrics, Definitions and Properties 1
1.1. Definitions and properties 2
1.2. Metric interpolation and intersection 6
1.2.1. Metric interpolation 7
1.2.2. Metric intersection 13
1.3. Geometric metrics 14
1.3.1. Geometric metric for a curve 16
1.3.2. Geometric metric for a surface 17
1.3.3. Turning any metric into a geometric metric 23
1.4. Meshing metrics 23
1.5. Metrics gradation 24
1.6. Element metric 31
1.6.1. Metric of a simplicial element 31
1.6.2. Metric of a non-simplicial element 37
1.6.3. Metric of an element of arbitrary degree 38
1.7. Element shape and metric quality 38
1.8. Practical computations in the presence of a metric 46
1.8.1. Calculation of the length 46
1.8.2. The calculation of an angle, area or volume 49
Chapter 2. Interpolation Errors and Metrics 53
2.1. Some properties 54
2.2. Interpolation error of a quadratic function 55
2.3. Bézier formulation and interpolation error 62
2.3.1. For a quadratic function 63
2.3.2. For a cubic function 66
2.3.3. For a polynomial function of arbitrary degree 80
2.3.4. Error threshold or mesh density 85
2.4. Computations of discrete derivatives 86
2.4.1. The L2 double projection method 86
2.4.2. Green formula 88
2.4.3. Least square and Taylor 89
Chapter 3. Curve Meshing 93
3.1. Parametric curve meshing 95
3.1.1. Curve in R3 95
3.1.2. About metrics used and computations of lengths 99
3.1.3. Curve plotted on a patch 103
3.2. Discrete curve meshing 104
3.3. Remeshing a meshed curve 104
Chapter 4. Simplicial Meshing 107
4.1. Definitions 108
4.2. Variety (surface) meshing 109
4.2.1. Patch-based meshing 110
4.2.2. Discrete surface remeshing 119
4.2.3. Meshing using a volume mesher 120
4.3. The meshing of a plane or of a volume domain 122
4.3.1. Tree-based method 123
4.3.2. Front-based method 126
4.3.3. Delaunay-based method 129
4.3.4. Remeshing of a meshed domain 134
4.4. Other generation methods? 136
Chapter 5. Non-simplicial Meshing 141
5.1. Definitions 142
5.2. Variety meshing 143
5.3. Construction methods for meshing a planar or volume domain 145
5.3.1. Cylindrical geometry and extrusion method 147
5.3.2. Algebraic methods and block-based methods 148
5.3.3. Tree-based method 172
5.3.4. Pairing method 174
5.3.5. Polygonal or polyhedral cell meshing 176
5.3.6. Construction of boundary layers 177
5.4. Other generation methods 182
5.4.1. Q-morphism or H-morphism meshing 182
5.4.2. Meshing using a reference frame field 183
5.5. Topological invariants (quadrilaterals and hexahedra) 185
Chapter 6. High-order Mesh Construction 195
6.1. Straight meshes 196
6.1.1. Local node numbering 196
6.1.2. Overall node numeration 201
6.1.3. Node positions 204
6.1.4. On filling up matrices according to element degrees 207
6.2. Construction of curved meshes 208
6.2.1. First-degree mesh 209
6.2.2. Node creation 209
6.2.3. Deformation and validation 210
6.2.4. General scheme 211
6.3. Curved meshes on a variety, curve or surface 215
Chapter 7. Mesh Optimization 225
7.1. Toward a definition of quality 226
7.2. Optimization process 233
7.2.1. Global methods 233
7.2.1.1. Optimization of a cost function 233
7.2.1.2. Iterative relaxation of the position of vertices by duality (simplices) 234...