Nanotechnology research and its commercial applications have experienced an exponential rise in the recent decades. Although there are a lot of studies with regards to toxicity of nanoparticles, the exposure to nanoparticles, both in terms of quality and quantity, during the life cycle of nanocomposites is very much an unknown quantity and an active area of research. Unsurprisingly, the regulations governing the use and disposal of nanomaterials during its life cycle are behind the curve. This work aims to assess the quantity of nanoparticles released along the life cycle of nanocomposites. Machining operations such as milling and drilling were chosen to simulate the manufacturing of nanocomposites parts, and impact testing to recreate the end-of-life of the materials. Several studies have tried to simulate different release scenarios, however these experiments had many variables and in general were not done in controlled environments. In this study, a reliable method was developed to assess the release of nanoparticles during machining and low velocity impact of nanocomposites. The development and validation of a new prototype used for measurement and monitoring of nanoparticles in a controlled environment is presented, as along with release experiments on different nanocomposites. Every sample tested was found to release nanoparticles irrespective of the mechanical process used or the type of material tested. Even neat polymers released nanoparticles when subjected to mechanical forces. The type of matrix was identified to play a major role on the quantity of nanoparticles release during different process. Thermoset polymers (and especially polyester) were found to release a higher number concentration of particles, mainly due to their brittle properties. A polyester sample was found to release up to 48 times more particles than a polypropylene one during drilling. The nanofiller type and percentage used to reinforce the polymer is also a key point. For example, the addition of 2 wt.% of nano-alumina into polyester increases the number concentration of particles by 106 % following an impact. The nanofiller chosen and its quantity affect the mechanical properties and machinability of the composites and therefore its nanoparticles release potential. The mechanical process and the process parameters chosen were also found to be crucial with regards to the nanoparticles released with different trends observed during drilling and impact of similar materials. Finally, thermal ageing of nanocomposites increases the number concentration of nanoparticles released (by 8 to 17 times after 6 weeks).

Low velocity impacts to the back are known to cause severe injury to crucial components such as the spine and kidneys. Researchers at Natick Soldier Center want to develop a solution that incorporates protection against low velocity impacts with the ballistic body armor (vest and plate) that is used today. The current ballistic body armor was developed to provide ballistic protection. Ballistic protection is designed to stop the penetration of bullets at velocities exceeding 300 m/s. Techniques to provide low velocity impact protection include reducing transmitted force by elongating collision time. In order to develop back protection for the soldier against low velocity impacts the performance of the ballistic body armor and impact protecting foams was evaluated. Low velocity impact tests were performed based on European standards for back protectors for horse riders (EN 13158) and motorcyclists (EN 1621-2). Performance requirements outlined by the standards and published literature established peak forces of 4 kN and 9 kN transmitted through materials under impact as minimum levels of safety before significant injury occurs. Experiments were conducted at an energy level of 4 J to compare the performance of different materials. Energy levels were then increased until maximum acceptable force transmissions were reached. At 4 J the ballistic materials showed peak transmitted forces between 11.0-16.2 kN. This indicated that the ballistic materials were not an adequate method to provide sufficient back protection. The addition of polyurethane foams to ballistic materials reduced peak force values by a factor of 15. Energy levels of 25 J and 40 J were reached with peak forces of 3.5 kN and 6.6 kN. This research provided a basis for the future development of protective equipment that provides both ballistic and low velocity impact protection.

In the current thesis, the 4-probe electrical resistance of carbon fiber-reinforced polymer (CFRP) composites is utilized as a metric for sensing low-velocity impact damage. A robust method has been developed for recovering the directionally dependent electrical resistivities using an experimental line-type 4-probe resistance method. Next, the concept of effective conducting thickness was uniquely applied in the development of a brand new point-type 4-probe method for applications with electrically anisotropic materials. An extensive experimental study was completed to characterize the 4-probe electrical resistance of CFRP specimens using both the traditional line-type and new point-type methods. Leveraging the concept of effective conducting thickness, a novel method was developed for building 4-probe electrical finite element (FE) models in COMSOL. The electrical models were validated against experimental resistance measurements and the FE models demonstrated predictive capabilities when applied to CFRP specimens with varying thickness and layup. These new models demonstrated a significant improvement in accuracy compared to previous literature and could provide a framework for future advancements in FE modeling of electrically anisotropic materials. FE models were then developed in ABAQUS for evaluating the influence of prescribed localized damage on the 4-probe resistance. Experimental data was compiled on the impact response of various CFRP laminates, and was used in the development of quasi- static FE models for predicting presence of impact-induced delamination. The simulation-based delamination predictions were then integrated into the electrical FE models for the purpose of studying the influence of realistic damage patterns on electrical resistance. When the size of the delamination damage was moderate compared to the electrode spacing, the electrical resistance increased by less than 1% due to the delamination damage. However, for a specimen with large delamination extending beyond the electrode locations, the oblique resistance increased by 30%. This result suggests that for damage sensing applications, the spacing of electrodes relative to the size of the delamination is important. Finally CT image data was used to model 3-D void distributions and the electrical response of such specimens were compared to models with no voids. As the void content increased, the electrical resistance increased non-linearly. The relationship between void content and electrical resistance was attributed to a combination of three factors: (i) size and shape, (ii) orientation, and (iii) distribution of voids. As a whole, the current thesis provides a comprehensive framework for developing predictive, resistance-based damage sensing models for CFRP laminates of various layup and thickness.

This thesis concerns the analysis of dynamical systems suitable to be modelled by piecewise-smooth differential equations. In such systems the continuous-in-time dynamics is interrupted by discrete-in-time jumps in the state or governing equations of motion. Not only can this framework be used to describe existing systems with strong nonlinear behaviour such as impacts and friction, but the non-smooth properties can be exploited to design new mechanical devices. As suggested in this work it opens up the possibility of, for example, fast limit switches and energy transfer mechanisms. Particularly, the dynamics at the onset of low-velocity impacts in systems with recurrent dynamics, so called grazing bifurcations in impact-oscillators, are investigated. As previous work has shown, low-velocity impacts is a strong source of instability to the dynamics, and efforts to control the behaviour is of importance. This problem is approached in two ways in this work. One is to investigate the influence of parameter variations on the dynamic behaviour of the system. The other is to implement low-cost control strategies to regulate the dynamics at the grazing bifurcation. The control inputs are of impulsive nature, and utilizes the natural dynamics of the system to the greatest extent. The scientific contributions of this work is collected in five appended papers. The first paper consists of an experimental verification of a map that captures the correction to the smooth dynamics induced by an impact, known in the literature as the discontinuity map. It is shown that the lowest order expansion of the map accurately captures the transient growth rate of impact velocities. The second paper presents a constructive proof of a control algorithm for a rather large class of impact oscillators. The proof is constructive in the sense that it gives control parameters which stabilizes the dynamics at the onset of low-velocity impacts. In the third paper a piecewise-smooth quarter-car model is derived, and the control strategy is implemented to reduce impact velocities in the suspension system. In the fourth and fifth papers the grazing bifurcation of an impact oscillator with dry friction type damping is investigated. It turns out that the bifurcation is triggered by the disappearance of an interval of stable stick solutions. A condition on the parameters of the system is derived which differentiates between stable and unstable types of bifurcation scenarios. Additionally, a low-cost control strategy is proposed, similar to the one previously mentioned, to regulate the bifurcation scenario. / QC 20100811

Mini unmanned Air Vehicles (UAV) have high significance among other UAV&#039 / s, in different categories, due to their ease of production, flexibility of maintenance, decrease in weight due to the elimination of landing gear system and simplicity of use. They are usually built to meet &#039 / hand launching&#039 / and &#039 / belly landing&#039 / criteria in order to have easy flight and easy landing features. Due to the hand take-off and belly landing features there is no need to have a runway and this feature is a very significant advantage in operational use. In an operation, belly landing mini UAV&#039 / s may encounter tough landing areas like gravel, concrete or hard soil. Such landing areas may create landing loads which are impulsive in character. The effect of the landing loads on the airframe of the mini unmanned air vehicle must be completely understood and the mini UAV be designed accordingly in order not to damage the mini UAV during belly landing. Typical impact speeds during belly landing is relatively low (&lt / 10 m/s) and in general belly landing phenomenon can be treated as low velocity impact. The purpose of this study is to analyze the impact loads on the composite substructures of a mini UAV due to the belly landing. &#039 / G&uuml / vent&uuml / rk&#039 / Mini UAV which is designed and built in METU Aerospace Engineering Department, is used as the analysis platform. This study is limited to the calculation of stresses and deformation that is caused by the low velocity impact forces encountered during belly landing. The main purpose of this work is to help the designer in making design decisions for a mini UAV that is tolerable to low velocity impact loads. Initial part of the thesis includes analytical treatment of low velocity impact phenomenon. In the simplified analytical approach the loading is assumed as quasistatic and comparisons of such a simplified method of analysis is made with explicit finite element solutions on isotropic and composite plate structures to investigate the applicability of simplified analytical method of analysis. Belly landing analyses of the mini UAV are done by MSC.Dytran, which is an explicit finite element solver. Model building and post processing are done via MSC.Patran. Stress and deformation response of the mini UAV is investigated by performing low velocity impact analysis using sub-structure built-up approach.

Στα πλαίσια της παρούσας διατριβής προσεγγίστηκε το πρόβλημα της κρούσης των σύνθετων υλικών μέσω μιας συνολικής μεθοδολογίας. Ο όρος συνολική μπορεί να χαρακτηρίσει την παρούσα εργασία της διατριβής γιατί η αναλυτική και αριθμητική υλοποίηση του προβλήματος που παρουσιάσθηκε επιλύει πολλά παράπλευρα προβλήματα. Η επίλυση του προβλήματος της κρούσης χαμηλής ταχύτητας σε σύνθετα υλικά κινήθηκε σε τέσσερις φάσεις. Σε πρώτη φάση αναπτύχθηκε το κύριο εργαλείο με το οποίο εξελίσσεται η διαδικασία μοντελοποίησης του προβλήματος. Η κατάστρωση του μοντέλου επίλυσης φυσικών συχνοτήτων και η αριθμητική επίλυση αυτού δίνει εκτιμήσεις για την ακρίβεια της μεθόδου Ritz με τη χρήση των αποδεκτών συναρτήσεων p-Ritz. Η μέθοδος Ritz επιτρέπει την πλήρη μοντελοποίηση του πολύστρωτου λαμβάνοντας υπόψη και τους καμπτικούς όρους σύζευξης κάτι που μέχρι στιγμής δεν έχει παρουσιαστεί στη βιβλιογραφία. Κατά τη δεύτερη φάση και μετά την τεκμηρίωση της πρώτης φάσης και τον έλεγχο της ακρίβειας των υπολογισμών των φυσικών συχνοτήτων, διατυπώθηκε το δυναμικό πρόβλημα. Η ακρίβεια των υπολογισμών ήταν εξαιρετική και σε ικανοποιητική σύγκλιση με τα αποτελέσματα του κώδικα πεπερασμένων στοιχείων LSDYNA3D. Παράλληλα αποδείχτηκε ότι το δυναμικό μοντέλο, επιλύει το δυναμικό φαινόμενο για όλα τα σχήματα και τους συνδυασμούς συνοριακών συνθηκών. Σε εξέλιξη των ανωτέρω έγινε η διατύπωση του συζευγμένου προβλήματος της κρούσης και επιλύθηκε για διάφορα υλικά και συνοριακές συνθήκες που αποτελούν μοντέλα υποδείγματα στη διεθνή βιβλιογραφία. Η ικανότητα πρόβλεψης των μοντέλων χαρακτηρίζεται ιδιαίτερα ικανοποιητική ενώ παρουσιάστηκε ο τρόπος εισαγωγής μη γραμμικών νόμων επαφής, στο συζευγμένο σύστημα διαφορικών εξισώσεων. Με σκοπό τον πλήρη χαρακτηρισμό του φαινομένου αλλά και τη μείωση των παραμέτρων που χαρακτηρίζουν το πρόβλημα σε μια προσπάθεια χαρακτηρισμού των κύριων παραμέτρων που διέπουν την κρούση παρουσιάζεται ο τρόπος υπολογισμού των αδιάστατων ομάδων που περιγράφουν την κρούση. Στα πλαίσια αυτής της ανάλυσης υπολογίστηκε το Νομογράφημα Κρουστικής Απόκρισης. Το Νομογράφημα Κρουστικής Απόκρισης μπορεί να χρησιμοποιηθεί στην οργάνωση πειραμάτων χαρακτηρισμού της κρουστικής αντοχής πολύστρωτων διατάξεων. Για την επίλυση του κύριου προβλήματος προτάθηκε μια διαδικασία ιεραρχικής αντιμετώπισης του προβλήματος. Για να καθοριστεί η κρουστική απόκριση του πολύστρωτου η διάταξη υποβάλλεται σε δυναμικό και στατικό χαρακτηρισμό με τη χρήση αριθμητικών μοντέλων που βασίζονται στη μέθοδο Ritz. Υλοποιήθηκε μια μέθοδος που δύναται να επιλύει διάφορες γεωμετρίες σε ποικιλία συνοριακών συνθηκών. Στη βιβλιογραφία μέχρι στιγμής έχουν παρουσιαστεί μοντέλα για παραλληλόγραμμες πλάκες, ειδικώς ορθότροπες και σε συνοριακές συνθήκες απλής έδρασης. Υλοποιήθηκαν προγράμματα σε Fortran που επιλύουν στατικά και δυναμικά προβλήματα σε καμπτικά φορτία και για ποικίλους συνδυασμούς συνοριακών συνθηκών. Υλοποιήθηκε το συζευγμένο πρόβλημα της κρούσης πολύστρωτης διάταξης από σφαίρα (διεισδυτή) σε χαμηλές ταχύτητες Έγινε σύγκριση ελαστοπλαστικών μη γραμμικών μοντέλων επαφής με γραμμικά μοντέλα επαφής. Με τη χρήση της θεωρίας αδιαστατοποίησης καθορίστηκαν οι αδιάστατες ομάδες που διέπουν το πρόβλημα της κρούσης και δημιουργήθηκε το Νομογράφημα Κρουστικής Απόκρισης. Αποδείχτηκε ότι το Νομογράφημα Κρουστικής Απόκρισης ισχύει για οιαδήποτε συμμετρική πολύστρωτη διάταξη σε όποια συνοριακή συνθήκη και αν είναι και οιοδήποτε σχήμα πλάκας. Αποδείχτηκε ότι το Νομογράφημα Κρουστικής Απόκρισης μπορεί να χρησιμοποιηθεί ως χάρτης αντοχής και ανοχής σε κρούση με τη χρήση της αδιάστατου Κατωφλιού Δύναμης για Διαστρωματική Αποκόλληση. / Low-velocity impact in composite laminates is a destructive loading condition since it leads to significant internal damage, no detectable by visual inspection. Numerous researchers have made significant efforts to model the impact response of composite plates and facilitate the analysis and design for impact resistance, using simple linear spring-mass models or a combination of continuum mechanics models and contact laws. Many investigators have used simple engineering structures beams, plates and shells to demonstrate the impact response of composite structures. The basic concept for modelling the impact of an object to a target was proposed by Timoshenko (1913). He introduced the procedure where the description of the interaction between the impactor and the structure (Euler beam) was implemented using the Hertzian contact law. This approach was further extended in isotropic plates and shells. Lee, 1940; Greszcuk, 1982; Lee et al., 1983 and Shivakumar et al, 1985, proposed simple models of the low-velocity impact problem. Although many important contributions exist in the bibliography for the characterization of the impact response of laminated plates, analytical solutions are considered to be very few. Most of the models proposed are for analyzing specially orthotropic plates subjected to a local dominated impact. In addition, these models do not take into account the shear coupling terms of the bending stiffness matrix, e.g. D16 and D26. Sun and Chattopadhyay (1975), Dobyns (1981), and Ramkumar and Chen (1983) employed the first order shear deformation theory developed by Whitney and Pagano (1970), and used it in conjunction with the Hertzian contact law or plasticity contact law for characterizing the impact of laminated composite plates. In their analysis they have studied the impact response of a simply supported orthotropic plate subjected to central impact using the lamination theory that includes transverse shear deformations. It is evident that the contact force history must be computed as part of the solution of the dynamic response problem solving the nonlinear integral equation. Christoforou and Swanson (1991) and Carvalho (1996), obtained an analytical solution of the impact problem using the Laplace transform technique. Qian and Swanson (1990) obtained analytical solutions by linearising the contact deformation law and compared this with a Rayleigh-Ritz approach with numerical integration in time. The three dimensional finite element method with an explicit formulation where also used in the dynamic analysis of laminated plates subjected to low velocity impact. Moreover, many researchers have proposed simple models for characterizing the impact phenomenon. Shivakumar et.al. (1985) developed a simple model to predict the impact force and the duration of the impact phenomenon on composite plates. The composite plates were modeled as three springs while their stiffness was calculated by the plate properties and the contact parameters. Other researchers instead of a classical Hertzian law used the statical indentation law presented by Yang and Sun (1982) and Tan and Sun (1985). The non Hertzian Contact impact was considered for combining the overall deformation of the structure to the local deformation in the contact area and was used to predict the impact response of transversely isotropic beams and plates. In general, it is impossible to obtain the exact solution for an impact problem except by ignoring the local contact deformation or pre-assuming the contact force. The first step for understanding the problem is to predict the force applied by the projectile on the structure during impact. In order to characterize the contact force history, the model should account for the motion of the target, the effective stiffness of the projectile The present work deals with the development of a numerical scheme for the calculation of the dynamic response of any type of laminated composite plates subjected to any type (and/or combination) of boundary conditions under low-velocity impact. Using the p-version Ritz polynomials any shape that can be represented in Cartesian coordinates can be formulated. The governing second order differential equations are derived and allow furnishing solutions with a variety of boundary conditions along the edges of the plate shape represented in Cartesian coordinates. Statically determined non-linear contact laws, for loading and unloading conditions, are coupled with the partial differential equations governing the dynamic response of the composite plate. These nonlinear governing equations of the contact-impact problem are decoupled according to the second order terms by the method of principal transformation and then solved numerically. The dynamic response of fully clamped cross-ply and a variation of simply support and fixed boundary conditions for angle ply and cross ply composite laminates was investigated. Rectangular, circular and elliptical plates where analysed using the developed numetrical scheme. A total analysis method is used in order to check the efficiency of the results. Starting from the homogeneous formulation of the structure the eigenfrequencies and eigenshapes were calculated for all material and boundary conditions cases. Using the “transformation of the principal” method all cases were subjected in force and pressure pulses in order to check the efficiency of the dynamic solver developed. The efficiency of the present method was investigated comparing the respective results with well-known benchmark problems, experimental results and FEM analysis using MSC-NASTRAN and LSDYNA 3D codes. Following this exhaustive analysis, the coupled contact-impact problem was formulated and numerous contact schemes were applied to the problem. For the impact problem the type of the response is intimately related to the contact stiffness, impact energy, ply thickness and orientation, geometrical configuration and boundary conditions, as well as the orthotropic material properties and the longitudinal, transverse and interlaminar strength both in compression and in tension. Three types of dynamic responses were found to exist, local, global and transitional. All types are categorised and analysed. Using Buckingham Theorem the numerous parameters of the dynamic contact-impact problem were grouped to non-dimensional groups in order to provide a flexible analysis scheme for theoretical and experimental evaluation of the low velocity impact problem. A general normalised graph for characterising the impact response of structures was constructed. The numerous advantages for analysing impact phenomena using the normalised curve developed were presented. Moreover its use for predicting composite material damage area on low velocity impact conditions was presented.

The objective of this study was to investigate the low velocity impact behavior of both monolithic and laminated aluminum alloy plates. For this purpose, a drop-weight test unit was used. The test unit included the free fall and impact of an 8 kg hammer with an 8 mm punching rod from 0.5 m to 4 m. The relationship between the change in static mechanical properties (hardness, ultimate tensile strength, yield strength, strain hardening rate) and low velocity impact behavior of monolithic aluminum plates were investigated. Tested material was AA 2024, heat treatable aluminum alloy, which was artificially aged to obtain a wide range of mechanical properties. In the second stage of the study, the relationship between the low velocity impact behavior of laminated plates was compared with that of monolithic aluminum plates at identical areal densities. For this purpose, a series of AA 2024 thin plates were combined with different types of adhesives (epoxy, polyurethane or tape). Finally, fracture surface of the samples and microstructure at the deformation zone were examined with both scanning electron microscope and optical microscope. It is found that the ballistic limit velocities of AA 2024 plates increase with increase in hardness, yield strength and ultimate tensile strength. It is also found that a linear relation exists between the ballistic limit velocity and strain hardening rate or hardness. When the low velocity impact behaviors of laminated and monolithic targets were compared, it was seen that monolithic targets have a higher ballistic limit velocity values for from the 2.5 to 10 mm thick targets. It was also observed that adhesives are not so effective to strengthen the low velocity impact performance. On the other hand, with increasing Charpy impact energy, penetration and perforation behaviors are getting worse in 10 to 30 joules energy range. Different types of failure mechanisms involving, plugging, dishing, stretching and bending were determined. For high strength and thick plates plugging type deformation was leaded. In contrast, for thinner and weaker targets bending, stretching and dishing type failures were dominating. For laminated targets also dishing type failure was determined.

L’utilisation industrielle ou militaire des explosifs est largement répandue. La sécurité est devenue un axe majeur avec notamment l’ignition involontaire des explosifs composés de HMX lors des impacts à basse vitesse. L’objectif de cette thèse est de contribuer à la compréhension des mécanismes dissipatifs à l’origine des échauffements locaux dans le matériau. Le développement d’un essai aux barres d’Hopkinson a permis de coupler de grandes vitesses de déformations à des pressions élevées pour compléter les données expérimentales. Cet essai a montré un angle de frottement quasiment identique à celui obtenu en quasistatique mais une contrainte de cohésion supérieure d’environ 25 MPa. Ensuite, pour observer les mécanismes pouvant être sources d’échauffement, un essai de compression dans la tranche a été développé avec des observations en temps réel. Celles-ci ont permis de conclure qu’il y a très peu de frottements entre les gros grains et la matrice (l’ensemble des petits grains, du liant et de la porosité). De la plasticité des grains de HMX a pu être observée mais surtout beaucoup d’endommagement dans certaines zones y compris dans la matrice. Une microfissuration très intense de certains grains a été observée. Parallèlement, une représentation numérique biphasique (gros grains de HMX et matrice) de toute la microstructure du matériau a été considérée. Une confrontation entre les observations expérimentales et les simulations a permis de déterminer le seuil de plasticité du HMX. Le comportement de la matrice a été identifié pour prendre en compte l’effet de vitesse et l’endommagement observé. Enfin, les confrontations entre les essais et les simulations de ceux-ci ont montré que les échauffements devraient plutôt se localiser dans la matrice que dans les gros grains de HMX et que le mécanisme le plus probable est le frottement de lèvres de microfissures. / Safety of industrial or military explosives is a major focus to prevent inadvertent ignition due to accidental loading as, for example, low-velocity impact. Our aim is to understand the dissipative mechanisms at work which could heat a pressed HMX-based PBX. A test based on the Split Hopkinson Pressure Bars system is proposed to carry out a dynamic triaxial compression test. This test simultaneously associates a high strain rate and a high pressure. Data have shown almost the same friction angle as during quasi-static experiments, but a higher cohesive stress. Then, A reversed edge-on impact test has been developed. This experiment enables the real-time observation of the deformation mechanisms at the microstructural scale. No relative displacement is observed between the biggest HMX grains and the matrix made of the smallest grains, the binder and the porosity. Plasticity has been observed into some HMX grains as well as damage by microcracking. Meanwhile, a biphasic (HMX grains and matrix) numerical representation of the material microstructure has been considered. A comparison between experimental observations and simulations is used to determine the yield stress of HMX. The behavior of the matrix has been determined to account for the influence of the strain rate and of the damage. Lastly, a comparison between tests and simulations has highlighted (1) that heating should rather be located in the matrix than in the biggest HMX grains and (2) that the most likely heating mechanism is the friction of microcracks lips.

Afin de mieux comprendre et de mieux quantifier la formation des dommages d'impact et leurs conséquences sur la tenue de la structure composite, le recours à la simulation numérique semble être un complément indispensable pour enrichir les campagnes expérimentales. Cette thèse a pour objectif la mise au point d'un modèle d'impact pour la simulation numérique par éléments finis dynamique implicite, capable de prévoir les dommages induits.La première étape du travail a consisté à élaborer un modèle s'appuyant sur le modèle de comportement du pli « Onera Progressive Failure Model » (OPFM) et sur le modèle bilinéaire de zones cohésives proposé par Alfano et Crisfield, puis d'évaluer la sensibilité aux différentes composantes des lois de comportement de la réponse à un impact et des dommages prévus. Des essais d'impact et d'indentation sur des plaques stratifiées en carbone/époxy ont ensuite été réalisés, analysés et enfin confrontés aux résultats numériques, afin d'évaluer les performances à l'impact du modèle OPFM et ses limites.Ces travaux permettent d'aboutir à trois principales conclusions. Premièrement, l'usage de modèles de zones cohésives semble nécessaire pour prévoir la chute de force caractéristique de l'impact sur stratifiés. Deuxièmement, la prise en compte des contraintes hors plan, notamment les cisaillements, est indispensable pour prévoir correctement l'endommagement d'impact. Enfin, si le modèle OPFM est capable de prévoir qualitativement les dommages d'impact, l'absence de caractère adoucissant ou de viscoplasticité semble cependant limiter leur prévision quantitative. / In order to improve the understanding and the quantification of the impact damage formation and of their consequences on the composite structure behavior, numerical simulation seems to be a necessary complement to experiments. This thesis aims at designing an impact model suited for a dynamic implicit finite element numerical simulation, able to predict the induced damages.The first step of the work consisted in building an impact model using the ply behavior law “Onera Progressive Failure Model” (OPFM) and the bilinear cohesive law defined by Alfano and Crisfield, then in evaluating the impact response and the predicted damage sensitivity to the different parameters of the behavior laws. Impact and indentation tests on carbon/epoxy laminate plates have then been performed, analyzed and compared with the numerical results, in order to evaluate the impact performance of OPFM and its limits.This work points out three key results. First, the use of cohesive zone models seems necessary to predict the typical load drop. Secondly, the out-of-plane constraints, especially the shearing, must be taken into account to correctly predict impact damages. Finally, even if the OPFM model is able to qualitatively predict impact damages, the lack of softening or viscoplasticity seems to limit their quantitative prediction.

No description available.