Soutenance mémoire
Wind energy in Canada
According to the Canadian Wind Energy Association (CANWEA), the total installed wind energy capacity increased from 137 MW in year 2000 to 11 205 MW as of December 2015, making Canada the seventh-largest wind energy producer in the world. Moreover, between 2011 and 2015, the sustained growth of the wind energy capacity has been 23 %. Thus, with over 1 500 MW installed in 2015, Canada was also the sixth country for added capacity for that same year (5; 6). Looking at the distribution of wind energy potential over Canada from the wind energy atlas by Environment Canada (13) — which shows the wind energy potential in Wm−2 of rotor swept surface (Figure 0.1) — or the ’Wind Energy in Canada’ map of Canadian Geographics (15), it is clear that a significant part of the best resources are set in the northernmost and remotest part of the country. However, there are little operational data for wind turbines and wind plants in cold climates such as that of Northern Canada. Combined with the fact that Canada is a large country with a low population density, the remoteness of many interesting sites would mean that in case of low turbine availability due to reliability issues, the COE could easily become excessive. The Wind Energy Strategic Network (WESNet) was founded in order to tackle the technical challenges associated with wind energy production in the distinctive climate and geography of Canada. WESNet was a five years endeavour regrouping 16 Canadian universities collaborating with 15 partners from industry, wind institutes and governments. The research programme was organized in the four following wind energy themes.
The nature of polymeric matrices and their composites
Polymer composites are made using a wide variety of matrices including both thermoset polymers and thermoplastics. Because of the possibilities for increased production rates and refusing of the polymer matrices for assembly, non-structural applications often make use of thermoplastic resins reinforced with short fibres. However, most structural applications use thermosetting resins—often simply called thermosets—for their greater stiffness and stability once cured. During manufacturing of composite parts, thermoset polymers go from a fluid state where the resin is forced through the network of reinforcing fibres using pressure (mechanical, atmospheric, autoclave,…) and capillarity. The resin is then cured through heating (selfheating due to exothermic chemical reactions or external heat source) and the initially fluid material irreversibly polymerized to an amorphous state. Cured thermoset resins are highly reticulated (cross-linked). This characteristic cross-linking of polymer branches hinders chain reorganizations and prevents the formation of a crystalline structure, but provides good strength and stiffness performances as well as good thermal and chemical stability. As opposed to thermoset, thermoplastic resins undergo a reversible hardening process and solidify mostly due to weak interaction forces and entanglement rather than through reticulation.
Because of the greater freedom of movement allowed by the absence of crosslinks in thermoplastic polymers, some thermoplastics exhibit a tendency to partially crystallize upon solidification. Manufacturing of composites using thermoplastics can either be done through in-situ polymerization (see e.g. Joncas (16)), or by diffusion of melted resin through the reinforcements. Because of the relatively high viscosity of thermoplastic melts, they are used essentially in conjunction with short fibres. Considering that most structural composites use amorphous matrices and that this class of materials behaves differently to most conventional (crystalline) engineering materials, it is useful to consider their peculiarities. 1.2 The formation, structure and properties of amorphous solids From a practical standpoint, solids are usually considered to be stable entities. This implies the assumption that their properties remain unchanged over extended periods of times, with the term ’extended’ being put in relation to the timescale that is significant for humans. For crystalline solids such as metals, the stability assumption is quite satisfying from an engineering point of view since unless a significant amount of energy is provided, the structure of crystalline solids is indeed stable. This is due to the fact that the solid phase of crystalline materials is in thermodynamic equilibrium.
Amorphous solids, of which polymers are an important subset, behave differently. But first, what is an amorphous material to start with? It is a material that has hardened from its liquid phase without undergoing a true phase change to the solid state. Upon polymerization in the case of thermosetting resins (or on cooling down through the fusion temperature Tm for semicristalline thermoplastics) the liquid phase experiences a rapid reduction of molecular mobility. On the timescale of the cooling, or polymerization in the case of thermosets, molecular rearrangement is eventually very restrained. The liquid phase becomes trapped in its configuration and starts hardening. This phenomenon is called the glass transition — even though it is not an actual phase transition in the thermodynamic sense. The result is a structure that is more or less disordered and that lacks the structural periodicity encountered in crystalline solids. Amorphous solids, although they appear stable on a first approximation, show complex time-dependent behaviour. Furthermore, their mechanical response shows a sensitivity to the history and rate of solicitation, of temperature, of pressure and of external work to which they are submitted (e.g. Struik (17) or Chow (18)). However, given a relatively short time scale, these structures may still be seen as stable, or rather ’metastable’. The fundamental reasons for this so-called metastability are yet to be fully understood (19). Still, the landmark review paper by Kauzmann (20) still provides many good insight on the physical process underlying the glass transition. It is strongly believed that the root causes for the reduced stability of glasses lies in the fact that their passage from the liquid phase to the solid state is not an actual phase change. As stated earlier, amorphous materials — also often called glasses — rather undergo a process of supercooling (cooling below the normal freezing point) and vitrification where the material passes from a liquid state at melt, to a supercooled liquid form and finally settle in a glassy, solid like state.
|
Table des matières
INTRODUCTION
0.1 Wind energy in Canada
0.2 The research problem and objective
0.3 Organization of the thesis
0.4 Contribution to solving the research problem
0.5 Significance of the work
CHAPTER 1 LITERATURE REVIEW
1.1 The nature of polymeric matrices and their composites
1.2 The formation, structure and properties of amorphous solids
1.2.1 Vitrification and the glass transition
1.2.2 Low temperature transitions
1.2.3 Time dependent mechanical response of amorphous solids
1.2.3.1 Viscoelasticity under sinusoidal loads
1.2.3.2 The time-temperature superposition principle
1.2.3.3 Ageing
1.2.3.4 Relaxation functions
1.2.3.5 Non-linear viscoelasticity
1.2.3.6 Creep failure models
1.3 Thermal stability of inorganic reinforcements
1.4 Thermomechanics of polymer matrix composites
1.4.1 Internal stresses in fibrous composites
1.4.2 Time dependent mechanical response of polymer matrix composites
1.4.3 Effects of temperature on the static strength and modulus of composites
1.5 Fatigue of composite materials
1.5.1 Modelling approaches in fatigue of composites
1.5.1.1 Empirical models
1.5.1.2 Strength degradation models
1.5.2 Statistical considerations
1.5.3 Effects of load rate and temperature on the fatigue performance of composite materials
1.6 Analysis
CHAPTER 2 TEMPERATURE AND FREQUENCY EFFECTS ON THE FATIGUE PROPERTIES OF UNIDIRECTIONAL GLASS FIBER-EPOXY COMPOSITES
2.1 Introduction
2.2 State of the art
2.2.1 Temperature effects on static properties
2.2.1.1 Constituent scale effects
2.2.1.2 Laminate scale effects
2.2.2 Temperature and frequency effects on fatigue behavior
2.3 Methodology
2.4 Results
2.4.1 Influence of temperature on static tensile properties
2.4.2 Influence of temperature on static compressive properties
2.4.3 Influence of temperature on fatigue life
2.4.4 Influence of frequency on fatigue life
2.5 Conclusion
CHAPTER 3 EFFECTS OF LOWTEMPERATURE ON THE MECHANICAL PROPERTIES OF GLASS FIBRE–EPOXY COMPOSITES: STATIC TENSION, COMPRESSION, R=0.1 AND R=−1 FATIGUE OF }45◦ LAMINATES
3.1 Background
3.2 Experimental
3.2.1 Material description
3.2.2 Test methods
3.2.3 Specimen description
3.2.4 Determination of S-N parameters
3.3 Results and discussion
3.3.1 Strength and modulus of }45◦ laminates
3.3.1.1 Tensile properties
3.3.1.2 Shear properties
3.3.1.3 Compressive properties
3.3.1.4 General considerations on static strength at low temperature
3.3.2 R = 0.1 tensile fatigue
3.3.3 R = −1 fully reversed fatigue
3.4 Conclusions
CHAPTER 4 MODELLING THE EFFECT OF TEMPERATURE ON THE PROBABILISTIC STRESS–LIFE FATIGUE DIAGRAM OF GLASS FIBRE–POLYMER COMPOSITES LOADED IN TENSION ALONG THE FIBRE DIRECTION
4.1 Introduction
4.2 Model description
4.2.1 Su(T) relationship
4.2.2 α(T) relationship
4.3 Materials and methods
4.3.1 Experimental
4.3.2 Computational approach
4.4 Results and discussion
4.4.1 Su(T) predictions for carbon–epoxy composite
4.4.2 Su(T) and fatigue life predictions for Upwind’s unidirectional glass–epoxy composite
4.4.3 Fatigue life predictions for WESNet’s unidirectional glassepoxy composite
4.4.4 Su(T) and fatigue life predictions for Sims and Gladman’s plain weave glass–epoxy composite
4.4.5 General discussion
4.4.6 Statistical considerations
4.5 Conclusions
CHAPTER 5 A NEW APPROACH FOR ASSESSING THE STORAGE MODULUS, TRANSITION TEMPERATURES AND TIME–TEMPERATURE SUPERPOSITION CHARACTERISTICS OF EPOXIES AND THEIR COMPOSITES FROM DYNAMIC MECHANICAL TESTS
5.1 Introduction
5.2 Model description
5.3 Materials and methods
5.3.1 Computational approach
5.4 Results
5.4.1 Rubber toughened epoxy storage modulus
5.4.2 Carbon–epoxy composite storage modulus
5.5 Discussion
5.6 Conclusion
CHAPTER 6 LINKING THE STORAGE MODULUS, LOSS MODULUS AND LOSS FACTOR OF POLYMERS THROUGH STATISTICAL DISTRIBUTIONS
6.1 Introduction
6.2 Theory
6.3 Results and discussion
6.4 Conclusions
CHAPTER 7 MODELLING THE EFFECTS OF TEMPERATURE ON THE INSTANTANEOUS STRENGTH OF POLYMER COMPOSITES ACROSS MULTIPLE TRANSITIONS
7.1 Introduction
7.2 Materials and methods
7.2.1 Computational approach
7.3 Results and discussion
7.4 Conclusions
CONCLUSION AND RECOMMENDATIONS
APPENDIX I MAXIMUM LIKELIHOOD ESTIMATION OF FATIGUE
CURVES INCLUDING RUNOUT DATA
APPENDIX II INTRODUCTION OF THE NORMALIZATION TEMPERATURE
TO THE GOMPERTZ DISTRIBUTION SURVIVAL
FUNCTION
APPENDIX III DEVELOPMENT OF EQUATION 5.10 BASED ON
LOCATING TG AT THE POINT OF MAXIMUM
CURVATURE UPSTREAM OF THE INFLECTION
POINT
BIBLIOGRAPHY
Télécharger le rapport complet
