The effect of the element length is examined in modelling crack growth in fatigue loading of an adhesive joint. This is done for a cohesive element using an expression for the damage evolution developed at the University of Skövde which is implemented using the UMAT subroutine in the FE-solver Abaqus. These analyses are done for pure mode I loading by analysing a DCB-specimen loaded by a pure moment. An expression is developed in which the critical element length is dependent on the geometry of the specimen (in the form of the wave number of the adhesive joint), the element length, the material properties of the adhesive (in form of the material parameters , , ), the load applied (in form of the stress in the crack tip), the time step used in the analysis and the crack growth rate. It is shown that the results converge by decreasing the element length and the time step used. Therefore an expression for the crack growth rate as a function of the remaining parameters can be determined. Another expression is thereafter developed for the element length needed in order to get a crack growth rate within a certain range of the critical element length. The results show a regular pattern but are not monotone. Therefor two different definitions of the critical element length are tested, either by defining the critical element length as the point where the error is greater than an arbitrary boundary of 1 % of a converged result or where a least square approximation of the error is within 1 % of the converged results. The first method shows a highly irregular result which makes it difficult to develop an expression out of these results. The second method on the other hand gives results that are predictable enough to develop a function out of them. This is done using a regression analysis with all parameters of a third order expression in order to get an expression.

With the development of the design and manufacture technology, composite materials are widely used in the aeronautical industry. But, one of the main concerns which affects the application of composites is foreign object impact. The damages induced by the Low Velocity Impact (LVI), which can significantly reduce the strength of the structures, can’t be easily inspected routinely. The so-called Barely Visible Impact Damages (BVID) due to LVI typically includes interlaminar delamination, matrix cracks and fibre fracture at the back face. Previous researches have shown that the results of LVI test are similar to that of the Quasi-Static Load (QSL) test. The initiation and propagation of delamination can be detected more easily in the QSL test and the displacement and reaction force of the impactor can be controlled and measured much more accurately. Moreover, it is easier to model QSL tests than dynamic impacts. To investigate the impact damage induced by LVI, a Finite Element (FE) model employing cohesive elements was used. At the same time, the ply clustering effect, when several plies of the same orientation were stack together, was modelled in the FE model in terms of damage resistance and damage size. A bilinear traction-separation law was introduced in the cohesive elements employed to simulate the initiation and propagation of the impact damage and delamination. Firstly, a 2D FE model of the Double Cantilever Beam (DCB) and End Notched Flexure (ENF) specimens were built using the commercial FEM software ABAQUS. The results have shown that the cohesive elements can be used to simulate mode I and mode II delamination sufficiently and correctly. Secondly, an FE model of a composite plate under QSL but without simulating damage was built using the continuum shell elements. Agreement between the FEA results with published test results is good enough to validate the capability of continuum shell elements and cohesive elements in modelling the composite laminate under the transverse load condition (QSL). Thirdly, an FE model containing discrete interface delamination and matrix cracks at the back face of the composite plate was built by pre-setting the cohesive failure elements at potential damage locations according to the experimental observation. A cross-ply laminate was modelled first where fewer interfaces could be delaminated. Good agreement was found in terms of the delamination area and impactor’s displacement-force curve. Finally, the effect of ply clustering on impact damage resistance was studied using Quasi-Isotropic (QI) layup laminates. Because of the limited time available for calculation, the simulation was only partly completed for the quasi-isotropic laminates (L2 configuration) which have more delaminated interfaces. The results showed that cohesive elements obeying the bilinear traction-separation law were capable of predicting the reaction force in quasi-isotropic laminates. However, discrepancies with the test results in terms of delamination area were observed for quasi-isotropic laminates. These discrepancies are mainly attributed to the simplification of matrix cracks simulation and compressive load at the interface in the thickness direction which is not taken into account.

Submitted by Viviane Lima da Cunha (viviane@biblioteca.ufpb.br) on 2015-12-07T12:44:49Z No. of bitstreams: 1 arquivototal.pdf: 7022743 bytes, checksum: 2ac4d21abb16718f361017265d0f34e2 (MD5) / Made available in DSpace on 2015-12-07T12:44:49Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 7022743 bytes, checksum: 2ac4d21abb16718f361017265d0f34e2 (MD5) Previous issue date: 2015-08-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / It is known that access to the written language mode is extremely important for the inclusion of individuals in grafocentics societies such as ours, however, we realized that mechanisms used in language teaching today still excludes the student in the teaching and learning process. To contribute with reflections in that field, this work presents the results of a survey that aims to reflect on the use of cohesive elements in the practice of textual production of a group of students belonging to a classroom of Educação de Jovens e Adultos (EJA). The study was conducted with students from ninth grade, fourth cycle of elementary school, belonging to a school municipal of Bayeux, Paraíba. Through this investigative process of qualitative, descriptive and interventionist nature, we promote the progress of these students, with regard to the acquisition of skills related to writing as a reflection on the use of cohesive elements, reviewing and rewriting texts. It was observed that the oral influence is a major reason that leads the student to confuse the use of the service system of written language. In the specific case of this study, the adversative conjunction “but” was misspelled as intensity adverb “more”. In this context, we understand that EJA students communicates and makes use of writing in their daily lives. They are, therefore, literate, however discredited because they don’t dominate the cultured language writing system, especially in relation to use of cohesive elements in the textual output. Considering such problems, we have developed writing activities, reflection and rewritten, emphasizing the importance of the written text area, working the inadequacies in order to provide learning to these students. The intervention proposal developed in the classroom was based on studies and proposals produced by Dolz, Noverraz and Schneuwly on didactic sequences and studies Gasparotto; Menegassi (2012) on the text rewritten. The work was built, taking also as reference studies by theorists such as Antunes (2003), Bagno (2010, 2013), Christiano and Silva (2005), Kleiman (2012), Kleiman and Sepulveda (2012), Koch and Elias (2014 ). Koch (2013) Marcuschi (2010), Neves (1990, 2004, 2010), Perini (2010), Possenti (1996), in addition to the PCNs and the textbook adopted by the school. Students advanced in understanding the idea of being more aware in the production of texts, not only when they are experiencing other levels of education, as well as the daily needs that emanate from the social use of writing practices. / Sabe-se que o acesso à modalidade escrita da língua é extremamente importante para a inserção dos indivíduos em sociedades grafocêntricas como a nossa, entretanto, percebemos que mecanismos utilizados no ensino da língua na atualidade ainda excluem o aluno no processo de ensino aprendizagem. Visando contribuir com reflexões nesse campo o trabalho apresenta os resultados de uma pesquisa que teve como objetivo refletir sobre o uso de elementos coesivos na prática da produção textual de um grupo de alunos, pertencentes a uma sala de aula da modalidade da Educação de Jovens e Adultos EJA. O estudo foi realizado com alunos do nono ano, quarto Ciclo do Ensino Fundamental, pertencentes de uma escola da rede municipal de Bayeux, Paraíba. Por meio desse processo investigativo de natureza qualitativa, descritiva e intervencionista, promovemos o avanço desses alunos, no que diz respeito à aquisição de habilidades relacionadas à escrita como a reflexão sobre o uso de elementos coesivos, a revisão e a reescrita de textos. Foi possível observar que a influência oral é um dos principais motivos que leva o aluno a confundir o uso do sistema culto escrito da língua. No caso específico desta pesquisa, a conjunção adversativa mas foi grafada como advérbio de intensidade mais. Nesse contexto compreendemos que o público da EJA se comunica e faz uso da escrita no seu cotidiano. São, portanto letradas, no entanto desacreditados por não dominarem o sistema escrito da língua culta, principalmente no que se refere aos usos dos elementos coesivos na produção textual. Considerando problemas dessa natureza, desenvolvemos atividades de escrita, reflexão e reescrita, enfatizando a importância do domínio do texto escrito, trabalhando as inadequações com o propósito de proporcionar aprendizagem aos referidos alunos. A proposta de intervenção desenvolvida em sala de aula, por sua vez, baseou-se nos estudos e propostas produzidas por Dolz, Noverraz e Schneuwly sobre sequências didáticas e nos estudos de Gasparotto; Menegassi (2012) sobre a reescrita textual. O trabalho foi construído, tomando também como referencial estudos produzidos por teóricos como Antunes (2003), Bagno (2010, 2013),Christiano e Silva (2005), Kleiman (2012), Kleiman e Sepulveda (2012), koch e Elias (2014). Koch (2013) Marcuschi (2010), Neves (1990, 2004, 2010), Perini (2010), Possenti (1996), além dos PCNs e o livro didático adotado pela escola. Os alunos avançaram na compreensão da ideia de estarem mais exigentes na produção de textos, não apenas quando estiverem vivenciando outros graus de escolaridade, como também nas necessidades que emanam cotidianamente das práticas sociais de uso da escrita.

The current thesis focuses on the deformation behaviour and strain distribution in the microstructure of ductile iron during tensile loading. Utilizing Digital Image Correlation (DIC) and in-situ tensile test under optical microscope, a method was developed to measure high resolution strain in microstructural constitutes. In this method, a pit etching procedure was applied to generate a random speckle pattern for DIC measurement. The method was validated by benchmarking the measured properties with the material’s standard properties. Using DIC, strain maps in the microstructure of the ductile iron were measured, which showed a high level of heterogeneity even during elastic deformation. The early micro-cracks were initiated around graphite particles, where the highest amount of local strain was detected. Local strain at the onset of the micro-cracks were measured. It was observed that the micro-cracks were initiated above a threshold strain level, but with a large variation in the overall strain. A continuum Finite Element (FE) model containing a physical length scale was developed to predict strain on the microstructure of ductile iron. The materials parameters for this model were calculated by optimization, utilizing Ramberg-Osgood equation. For benchmarking, the predicted strain maps were compared to the strain maps measured by DIC, both qualitatively and quantitatively. The DIC and simulation strain maps conformed to a large extent resulting in the validation of the model in micro-scale level. Furthermore, the results obtained from the in-situ tensile test were compared to a FE-model which compromised cohesive elements to enable cracking. The stress-strain curve prediction of the FE simulation showed a good agreement with the stress-strain curve that was measured from the experiment. The cohesive model was able to accurately capture the main trends of microscale deformation such as localized elastic and plastic deformation and micro-crack initiation and propagation.

Ultra high performance fibre reinforced concrete (UHPFRC) is a relatively new fibre reinforced cementitious composite and has become very popular in construction applications. Extensive experimental studies have been conducted, demonstrating its superior properties such as much higher strength, ductility and durability than conventional fibre reinforced concrete (FRC) and high performance concrete. However, the material's damage and fracture mechanisms at meso/micro scales are not well understood, limiting its wider applications considerably. This study aims at an in-depth understanding of the damage and fracture mechanisms of UHPFRC, combining microscale in-situ X-ray computed tomography (µXCT) experiments and mesoscale image-based numerical modelling. Firstly, in-situ µXCT tests of small-sized UHPFRC specimens under wedge splitting loading were carried out, probably for the first time in the world, using an in-house designed loading rig. With a voxel resolution of 16.9µm, the complicated fracture mechanisms are clearly visualised and characterised using both 2D images and 3D volumes at progressive loading stages, such as initiating of micro-cracks, arresting of cracks by fibres, bending and pulling out of fibres and spalling of mortar at the exit points of inclined fibres. Secondly, based on the statistics of pores in the µXCT images obtained for a 20mm cube specimen, an efficient two-scale analytical-numerical homogenisation method was developed to predict the effective elastic properties of the UHPFRC. The large number of small pores were first homogenised at microscale with sand and cement paste, using elastic moduli from micro-indentation tests. 3D mesoscale finite element models were built at the second scale by direct conversion of the µXCT images, with fibres and large pores were faithfully represented. The effects of the volume fraction and the orientation of steel fibres on the elastic modulus were investigated, indicating that this method can be used to optimise the material micro-structure. Thirdly, 3D mesoscale finite element models were built for the specimen used in the in-situ µXCT wedge splitting test, with embedded fibre elements directly converted from the µXCT images. The fracture behaviour in the mortar was simulated by the damage plasticity model available in ABAQUS. Finally, 2D mesoscale finite element models were developed to simulate the fracture behaviour of UHPFRC using cohesive interface elements to simulate cracks in the mortar, and randomly distributed two-noded 1D fibres and connector elements to simulate the pull-out behaviour of fibres. This approach offers a link between the fibres pull-out behaviour and the response of the whole composite at the macroscale, thus it can be used to conduct parametric studies to optimise the material properties.

[pt] Apresentamos um método computacional na GPU que lida com eventos de fragmentação dinâmica, simulados por meio de elementos de zona coesiva. O trabalho é dividido em duas partes. Na primeira parte, tratamos o pré-processamento de informações e a verificação de corretude e eficácia da inserção dinâmica de elementos coesivos em malhas grandes. Para tal, apresentamos uma simples estrutura de dados topológica composta de triângulos. Na segunda parte, o código explícito de dinâmica é apresentado, que implementa a formulação extrínsica de zona coesiva, onde os elementos são inseridos dinamicamente quando e onde forem necessários. O principal desafio da implementação na GPU, usando a formulação de zona coesiva extrínsica, é ser capaz de adaptar dinamicamente a malha de uma forma consistente, inserindo elementos coesivos nas facetas fraturadas. Para isso, a estrutura de dados convencional usada no código de elementos finitos (baseado na incidência de elementos) é estendida, armazenando, para cada elemento, referências para elementos adjacentes. Para evitar concorrência ao acessar entidades compartilhadas, uma estratégia convencional de coloração de grafos é adotada. Na fase de pré-processamento, cada nó do grafo (elementos na malha) é associado a uma cor diferente das cores de seus nós adjacentes. Desta maneira, elementos da mesma cor podem ser processados em paralelo sem concorrência. Todos os procedimentos necessários para a inserção de elementos coesivos nas facetas fraturadas e para computar propriedades de nós são feitas por threads associados a triângulos, invocando um kernel por cor. Computações em elementos coesivos existentes também são feitas baseadas nos elementos adjacentes. / [en] A GPU-based computational framework is presented to deal with dynamic failure events simulated by means of cohesive zone elements. The work is divided into two parts. In the first part, we deal with pre-processing of the information and verify the effectiveness of dynamic insertion of cohesive elements in large meshes. To this effect, we employ a simplified topological data structured specialized for triangles. In the second part, we present an explicit dynamics code that implements an extrinsic cohesive zone formulation where the elements are inserted on-the-fly, when needed and where needed. The main challenge for implementing a GPU-based computational framework using extrinsic cohesive zone formulation resides on being able to dynamically adapt the mesh in a consistent way, inserting cohesive elements on fractured facets. In order to handle that, we extend the conventional data structure used in finite element code (based on element incidence) and store, for each element, references to the adjacent elements. To avoid concurrency on accessing shared entities, we employ the conventional strategy of graph coloring. In a pre-processing phase, each node of the dual graph (bulk element of the mesh) is assigned a color different to the colors assigned to adjacent nodes. In that way, elements of a same color can be processed in parallel without concurrency. All the procedures needed for the insertion of cohesive elements along fracture facets and for computing node properties are performed by threads assigned to triangles, invoking one kernel per color. Computations on existing cohesive elements are also performed based on adjacent bulk elements.

[pt] Apresentamos um método computacional na GPU que lida com eventos de fragmentação dinâmica, simulados por meio de zona coesiva. Implementamos uma estrutura de dados topológica simples e especializada para malhas com triângulos ou tetraedros, projetada para rodar eficientemente e minimizar ocupação de memória na GPU. Apresentamos um código dinâmico paralelo, adaptativo e distribuído que implementa a formulação de modelo zona coesiva extrínsica (CZM), onde elementos são inseridos adaptativamente, onde e quando necessários. O principal objetivo na implementação deste framework computacional reside na habilidade de adaptar a malha de forma dinâmica e consistente, inserindo elementos coesivos nas facetas fraturadas e inserindo e removendo elementos e nós no caso da malha adaptativa. Apresentamos estratégias para refinar e simplificar a malha para lidar com simulações dinâmicas de malhas adaptativas na GPU. Utilizamos uma versão de escala reduzida do nosso modelo para demonstrar o impacto da variação de operações de ponto flutuante no padrão final de fratura. Uma nova estratégia de duplicar nós conhecidos como ghosts também é apresentado quando distribuindo a simulação em diversas partições de um cluster. Deste modo, resultados das simulações paralelas apresentam um ganho de desempenho ao adotar estratégias como distribuir trabalhos entre threads para o mesmo elemento e lançar vários threads por elemento. Para evitar concorrência ao acessar entidades compartilhadas, aplicamos a coloração de grafo para malhas não-adaptativas e percorrimento nodal no caso adaptativo. Experimentos demonstram que a eficiência da GPU aumenta com o número de nós e elementos da malha. / [en] A GPU-based computational framework is presented to deal with dynamic failure events simulated by means of cohesive zone elements. We employ a novel and simplified topological data structure relative to CPU implementation and specialized for meshes with triangles or tetrahedra, designed to run efficiently and minimize memory requirements on the GPU. We present a parallel, adaptive and distributed explicit dynamics code that implements an extrinsic cohesive zone formulation where the elements are inserted on-the-fly, when needed and where needed. The main challenge for implementing a GPU-based computational framework using an extrinsic cohesive zone formulation resides on being able to dynamically adapt the mesh, in a consistent way, by inserting cohesive elements on fractured facets and inserting or removing bulk elements and nodes in the adaptive mesh modification case. We present a strategy to refine and coarsen the mesh to handle dynamic mesh modification simulations on the GPU. We use a reduced scale version of the experimental specimen in the adaptive fracture simulations to demonstrate the impact of variation in floating point operations on the final fracture pattern. A novel strategy to duplicate ghost nodes when distributing the simulation in different compute nodes containing one GPU each is also presented. Results from parallel simulations show an increase in performance when adopting strategies such as distributing different jobs amongst threads for the same element and launching many threads per element. To avoid concurrency on accessing shared entities, we employ graph coloring for non-adaptive meshes and node traversal for the adaptive case. Experiments show that GPU efficiency increases with the number of nodes and bulk elements.

Timber-concrete composite (TCC) structures are becoming more popular in several industrial applications as an efficient method for optimizing the structural performance and the cost of construction as well as lowering the emissions of carbon dioxide. TCC floors are more sustainable than pure concrete floors and more resistant to vibrations and excessive deflections than pure wooden floors. The effectiveness of a TCC floor is dependent on the connection between the materials. The stiffness and strength of the composite element increases by having a rigid connection. An example of a rigid connection is an adhesive-bonded connection, however obtaining a connection without slip is difficult considering there will always be certain amount of slippage in the connection. In this thesis adhesive connections are investigated with two different types of adhesive, one called Sikasil SG-500 and the other Sika PS. The application of the adhesives for the test specimen differ. Since Sika PS is a more fluid glue, a different approach was necessary than for the Sikasil SG-500. Five test specimen were tested of each adhesive by performing double shear push out tests and comparing the results to a numerical model, which was performed in ABAQUS. In the numerical model, the adhesive was created as a cohesive element. Furthermore the properties of the modeled adhesive was based on the experimental results for respective adhesive. The experimental results showed that both adhesives managed the estimated shear force in the serviceability limit state and the ultimate limit state and yet remained in the elastic region. Sikasil SG-500 turned out to be more flexible while Sika PS specimen resulted in higher shear strength. The short creep tests (30 minutes of a constant load) showed that the deformation increased more for Sika PS than for Sikasil SG-500. However both adhesives had large deformations after only 30 minutes. There was also a difference in the average longitudinal shear strength, where it resulted in 1.06 MPa for Sikasil SG-500 and 2.02 MPa for Sika PS. This study indicates that Sika PS is more preferable in TCC structures than Sikasil SG-500.

Submitted by Aelson Maciera (aelsoncm@terra.com.br) on 2017-05-30T17:42:50Z No. of bitstreams: 1 DissBL.pdf: 8282671 bytes, checksum: 5dfca89e85bbea10d97a0617e91d1408 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-06-01T17:47:56Z (GMT) No. of bitstreams: 1 DissBL.pdf: 8282671 bytes, checksum: 5dfca89e85bbea10d97a0617e91d1408 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-06-01T17:48:03Z (GMT) No. of bitstreams: 1 DissBL.pdf: 8282671 bytes, checksum: 5dfca89e85bbea10d97a0617e91d1408 (MD5) / Made available in DSpace on 2017-06-01T18:20:58Z (GMT). No. of bitstreams: 1 DissBL.pdf: 8282671 bytes, checksum: 5dfca89e85bbea10d97a0617e91d1408 (MD5) Previous issue date: 2017-02-13 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / The study of ceramic composites assisted by computer simulation is well spread nowadays. It is an interest of the ceramic industry the development of materials models trustfully enough to reduce cost with prototypes and virtually explore an infinitude of materials compositions and thermomechanical loads. The mismatch of thermal and mechanical properties among the composite’s phases may induce decohesions or cracks after temperature variation. The computer modeling of this behavior could auxiliate the planning of the systems’ composition, as it becomes possible to evaluate the influence of the concentration of an specific constituent on the global behavior. The present dissertation aims to analyze the role of the geometrical feature, such as inclusion radius and volume fraction, in the composite behavior submitted to temperature variation. It was also analyzed the application of coesive elements to simulate the cracking phenomena in ceramic systems. The main results of this dissertation were not only the thermomechanical properties influence on the global behavior of ceramic systems, but also the construction of finite element models that might be usefull to others reseachers on the investigation of the thermomechancial behavior of distincts ceramic systems. / O estudo de compósitos cerâmicos assistido por simulação computacional encontra-se em plena expansão. É de grande interesse da indústria cerâmica o desenvolvimento de modelos confiáveis de materiais o suficiente para reduzir os custos com protótipos e explorar virtualmente uma infinidade de possibilidades de composições e solicitações termomecânicas. A diferença entre as propriedades termomecânicas dos constituintes de compósitos pode induzir o surgimento de defeitos quando estes são submetidos a variações de temperatura. A modelagem computacional adequada destes sistemas nestas condições pode auxiliar no planejamento da composição do sistema, uma vez que é possível avaliar a influência da concentração de determinado constituinte no comportamento global do compósito. O presente trabalho objetiva analisar desde a influência do raio e fração volumétrica de inclusões até o efeito das diferenças de propriedades termomecânicas entre as fases no comportamento de compósitos cerâmicos submetidos a variações de temperatura. Analisar também, a aplicação de elementos coesivos na simulação da fissuração de sistemas cerâmicos. Como resultados principais, apresenta-se não só a compreensão dos efeitos de determinadas propriedades termomecânicas, mas também a construção de modelos em elementos finitos que podem ser utilizados por terceiros na investigação do comportamento termomecânico de sistemas cerâmicos.

[pt] Este trabalho visa discutir tópicos ainda não totalmente compreendidos a respeito do comportamento de compósitos pultrudados de polímero reforçado com fibras de vidro, englobando os mecanismos de falha e danos, bem como a resposta à múltipla fissuração dos compósitos. Investigações experimentais e numéricas foram conduzidas, com foco nas junções mesa-alma extraídas de variados tipos de perfis e decks de pontes. Fatores que propiciam a delaminação, como a influência de defeitos e imperfeições locais e a consequente modificação das trajetórias de tensão são discutidas. Equações teóricas foram propostas para a determinação da rigidez rotacional das junções mesa-alma, resultando em máximas diferenças de 24 por cento e 38 por cento para seções I e C, respectivamente. Para contribuir para um banco de dados de parâmetros da fratura relacionados a esse material, adaptações de clássicos experimentos da mecânica da fratura, para ambos os Modos I e II, foram propostos e conduzidos. Nove métodos de regressão de dados foram analisados e comparados com modelos de elementos finitos. O método Modified Beam Theory (MBTASTM) apresentou os melhores resultados para propagação de fissuras no Modo I, enquanto para o Modo II, os métodos Corrected beam theory using effective crack length (CBTE) e Experimental compliance method (EMC) apresentaram os resultados mais próximos dos modelos numéricos. Finalmente, uma nova abordagem para a análise de múltipla fissuração é proposta. Curvas R e curvas compliance foram obtidas com base na soma dos comprimentos de fissuras e as respectivas taxas de liberação de energia elástica foram avaliadas separadamente. Os resultados da abordagem proposta foram comparados com modelo numérico desenvolvido com o auxílio de elementos coesivos e boas correlações em termos da resposta carga x deslocamento foram encontradas. / [en] This work aims to discuss the behavior of pultruded glass fiber-polymer composites related to issues not fully understood, comprising the damage and failure mechanisms, as well as the multi-crack composites response. Experimental and numerical investigations are conducted mainly on web-flange junctions (WFJ) extracted from varied profiles and bridge decks systems. Factors that favor the composites delamination, such as the influence of local defects/imperfections and the consequent modification on the junctions stresses trajectories are addressed. Theoretical equations were proposed to determine the WFJs rotational stiffness, resulting in maximum differences of 24 percent and 38 percent for I-sections and channels, respectively. To complement the scarcity of fracture parameter data related to the material, adaptations of classical fracture mechanic experiments for both Modes I and II were proposed and conducted. Nine data reduction methods were analyzed and compared with finite element models. The Modified Beam Theory method presented the best results for crack propagation in Mode I, whereas for Mode II, the Corrected Beam Theory using effective crack length and the Experimental Compliance Methods presented the closest results to numerical models. Finally, a novel approach for the multi-crack assessment is proposed. Compliance and R-curves are obtained based on the sum of the crack lengths and the strain energy release rate for each crack is evaluated separately. The results of this novel approach were compared with numerical analyses developed with the use of cohesive elements and a good agreement was found in terms of load vs. displacement response.