Aleksendric / Carlone Soft Computing in the Design and Manufacturing of Composite Materials

Applications to Brake Friction and Thermoset Matrix Composites
1. Auflage 2015
ISBN: 978-1-78242-180-1
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 320 Seiten

Due to problems associated with the design and manufacturing of composite materials, there is a need to introduce computational and intelligent systems engineering methodology in materials engineering. Soft Computing in the Design and Manufacturing of Composite Material offers an intelligent approach to advance material engineering, and significantly improves the process of designing and manufacturing a new material. This title includes chapters covering topics such as soft computing techniques, composite materials engineering, design and manufacturing of composite materials, numerical modeling, prediction, and optimization of the composite materials performance, development of the hybrid models, and control of the composite material performance. - Introduction of soft computing in the composite materials engineering - Includes accurate and detailed analysis of the current state of the art in the field - Development of the intelligent models for design and manufacturing of composite material - Details composite material performance prediction - Optimization of the manufacturing process of composite materials

Dragan Aleksendric is an Associate professor at Automotive Department, Faculty of Mechanical Engineering, University of Belgrade, Serbia

Autoren/Hrsg.

Aleksendric, Dragan

Carlone, Pierpaolo

Leseproben

List of figures

2.1 The process of product development 10

2.2 Top-down–bottom-up approach to the development of a braking system regarding the properties of a friction pair 11

3.1 Temperature and degree-of-cure profiles for a Shell Epon 9420/9470/537 resin 18

3.2 Temperature and viscosity profiles for a Shell Epon 9420/9470/537 resin 19

3.3 Energy intensity of composite-manufacturing processes 24

3.4 Schematic view of the pultrusion process 24

3.5 Synergetic effects of formulation and manufacturing conditions on friction and wear of a brake friction material 30

3.6 A flash mould method for brake friction material manufacturing 32

3.7 Flash mould method – a mould cavity 33

3.8 Positive moulding – compression process without a breathing cycle 34

4.1 The basic architecture of an artificial neural network 40

4.2 The learning process of an artificial neural network 41

4.3 The process of development of an artificial neural network model 43

4.4 Typical structure of a layer-recurrent neural network 44

4.5 Dynamic neural model of disc brake operation based on a layer-recurrent network 44

4.6 Geometric interpretation of the finite difference approximation of the first derivative 49

4.7 Element type and dimensionality 51

4.8 Mesh and dual mesh in vertex-centred FVM (a, b) and cell-centred FVM (c, d). Control volumes are defined by the grey-coloured areas 52

4.9 The basic cycle of a genetic algorithm 54

4.10 A hybrid ANN–GA optimization model 56

5.1 Modelling issues and reciprocal interactions in composite-manufacturing processes 65

5.2 Time–temperature–transformation diagram for a generic thermoset resin 69

5.3 Micro-, meso- and macro-scales in composite-manufacturing simulation 75

5.4 Graphical scheme of the evaluation of the Morishita index 76

5.5 Graphical scheme of the evaluation of the Ripley function, showing the definition of Ik(r) and the weight factor: (a) wk = 1, (b) wk ? 1 77

5.6 L function: comparison with the Poisson distribution 78

5.7 Lay-up of a process and corresponding finite element three-dimensional scheme 80

5.8 Lay-up of the process and corresponding finite element one-dimensional scheme 82

5.9 Temperature profiles: numerical results and reference data 84

5.10 Degree-of-cure profiles: numerical results and reference data 84

5.11 Viscosity profiles: numerical results and reference data 85

5.12 Viscosity profiles, showing the process window 86

5.13 Flow chart of the generation algorithm for the random RVEs 88

5.14 RVE perturbation: (a) 50, (b) 500, (c) 1000, (d) 5000 iterations 89

5.15 Statistical analysis of the RVEs: (a) Ripley L function; pair distribution function; (c) histogram of radius distribution; (d) Morishita number 90

5.16 Micro-scale computational domain and boundary conditions 91

5.17 Thermal flux along the transverse direction 92

5.18 Temperature and degree of cure at centre 94

5.19 Multi-physics involved in the pultrusion process and related interactions 95

5.20 Centreline pressure rise in the tapered region of the die 98

5.21 Streamlines of resin flow in the tapered region of the die 99

5.22 (a) Case study, and discretization of the cross-section: (b) FDM and (c) FEM 103

5.23 Temperature profiles in the pultrusion die 105

5.24 Cure profiles in the pultrusion die and comparison with reference data 106

5.25 Schematic view of the pultrusion domain for the composite rod. All dimensions are in millimetres 109

5.26 Temperature and degree of cure profiles: comparison of the outcomes of the present calculations with the reference data 110

5.27 Temperature and degree of cure profiles: comparison of the outcomes of the present calculations with the reference data 111

5.28 Pulling force and phase changes 112

5.29 Viscosity profiles and virtual workpiece radius 113

5.30 Flow front after (a) 10 s, (b) 30 s, (c) 60 s, (d) 120 s, (e) 180 s and (f) 300 s since the beginning of impregnation 117

5.31 Resin flow front: numerical and analytical results 117

5.32 (a) Solid model and (b) meshed computational domain 118

5.33 Resin front after (a) 15 s, (b) 3 min, (c) 6 min, (d) 9 min, (e) 18 min and (f) 30 min 119

5.34 Computational domain and boundary conditions: (a) calculation of transverse permeability; (b) calculation of longitudinal (axial) permeability 120

5.35 Transverse-permeability results 121

5.36 Axial-permeability results 122

5.37 Flow front, and temperature and degree of cure distributions at different time instants for Text = 25 °C 125

5.38 Flow front, and temperature and degree of cure distributions at different time instants for Text = 50 °C 126

5.39 Flow front, and temperature and degree of cure distributions at different time instants for Text = 100 °C 127

5.40 Viscosity distributions at t = 300 s for Text = (a) 25 °C, (b) 50 °C and (c) 100 °C 127

5.41 Numerical and analytically computed (using the mean viscosity value) resin flow fronts 128

5.42 (a) Experimental set-up and (b) results from dielectric monitoring of flow through a dual-scale porous medium 130

5.43 Mass balance in an elementary control volume including saturation effects 131

5.44 Tow saturation scheme 131

5.45 Numerical and analytically computed (using the mean viscosity value) resin flow fronts 132

5.46 (a) Pultrusion process and (b) section considered for optimization 140

5.47 Temperature profiles in the pultrusion die in the reference case 141

5.48 Cure profiles and degree of cure distribution in the final cross-section of the workpiece in the reference case 141

5.49 Convergence plots using different selection criteria: (a) uniform, (b) roulette and (c) tournament 144

5.50 Control temperatures of the die heating zones, according to (a) the genetic optimization routine and (b) the hybrid routine 147

5.51 Temperature and cure profiles and distribution of the degree of cure in the final cross-section of the workpiece, after the genetic optimization routine using the FEM 148

5.52 Temperature and cure profiles and distribution of the degree of cure in the final cross-section of the workpiece, after the hybrid optimization routine using the FEM 148

5.53 Temperature and cure profiles and distribution of the degree of cure in the final cross-section of the workpiece after the genetic optimization routine (test case Tc4, using the FEM) 151

5.54 Temperature and cure profiles and distribution of the degree of cure in the final cross-section of the workpiece after the hybrid genetic routine (test case Tc4)...