Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Cement-based composite materials are widely used in engineering applications. The strength and damage patterns of such materials depend upon the properties of the constituent components as well as the microstructure. Three scale levels are generally recognized in the analysis of the mechanical behaviour of composites, namely, macro-scale, meso-scale, and nano- or atomistic scale. Modelling of the mechanical properties at the meso-level provides a powerful means for the understanding of the physical processes underlying the macroscopic strength and failure behaviour of the composite materials under various loading conditions. This thesis endeavours to develop effective and efficient mesoscale models for cement-based composites, especially concrete, with a focus on dynamic analysis applications and in a three-dimensional stress-strain environment. These models are subsequently applied to investigate the intrinsic microscopic mechanisms governing the behaviour of such material under complex and high rate loadings, such as those due to shock, impact and blast. To cater to the needs of dynamic analysis under complex stress conditions, a general 2-dimensinal (2D) mesoscale modelling framework is further developed with the incorporation of the 3-D effect. This framework integrates the capabilities of MATLAB programming for the generation of the mesoscale geometric structure, ANSYS-CAE for finite element mesh generation, and the hydrocode LS-DYNA for solving the dynamic response of the model. The 3D effect is incorporated via a novel pseudo-3D modelling scheme such that the crucial lateral confinement effect during the transient dynamic response can be realistically represented. With the above mesoscale model a comprehensive investigation is conducted on the dynamic increase factor (DIF) in the concrete strength under compression, with particular focus on the variation trend at different strain rate regimes, and the key influencing factors. The wave propagation effect under high strain rate is scrutinised from a strip-by-strip perspective, and the correlation between the externally measured stress-strain quantities and the actual processes within the specimen is examined. The contribution of the material heterogeneity, as well as the structural effect (inertia), in the dynamic strength enhancement is evaluated. The classical Brazilian (splitting) test for the dynamic tensile behaviour of concrete is also investigated with the aid of the mesoscale model. Of particular interest here is the validity of such an indirect setup in reproducing the tensile behaviour of the specimen under high strain rates, as well as the effect of the heterogeneity in the dynamic tensile strength. Complications are found to arise as the loading rate increases. The change of the damage patterns with increase of the loading rate and the implications on the interpretation of the results are discussed. As an ideal solution to modelling of the 3-D effects, a methodology for the creation of a complex real 3-dimensional mesoscale model is put forward in the last part of the thesis. A geometric concept, called convex hull, is adopted for the representation of aggregates, and this makes it possible to utilize the relevant algorithms in computational geometry for the present purpose of generation of random 3-D aggregates. A take-and-place procedure is employed to facilitate the generation of the complete 3-D meso-structure. Associated techniques are developed for fast detection of particle inclusion-intersection. An example 3D mesoscale model is presented and representative numerical simulations are carried out to demonstrate the performance of the 3-D mesoscale modelling scheme.

Inquiry into fairness of a test has been recognized as an important research activity to direct efforts to reduce bias and discrimination against certain groups of test takers, create equal opportunities for test takers to demonstrate their knowledge and skills, and promote social justice. Given the importance of fairness in testing, my research examined the fairness of one large-scale high-stakes test in China—the Graduate School Entrance English Exam (GSEEE). To guide this doctoral research, I first drew on four sources of conceptual and empirical work to identify key issues encompassing test fairness. Informed by Willingham’s conceptual framework, I investigated the fairness of the GSEEE in two studies. In Study 1, I examined whether the GSEEE test items functioned differentially and brought potential bias towards test taker groups based on gender and academic background. In Study 2, I investigated perceptions of the fairness of the GSEEE as expressed by program administrators, teachers, and test takers. In conclusion, this research offers empirical information with regard to the fairness of the GSEEE from psychometric and stakeholder perspectives. The research also provides evidence that the conceptualization of test fairness is mediated by contextualized beliefs and traditions. Whether a test is perceived as fair or not is derived from considerations in both the testing process and the broad socio-cultural context. / Thesis (Ph.D, Education) -- Queen's University, 2014-06-10 10:33:32.484

Dimensions of a polytomous employee attitude survey were examined for the presence of differential item functioning (DIF) and differential test functioning (DTF) utilizing Raju, van der Linden, & Fleer's (1995) differential functioning of items and tests (DFIT) framework. Comparisons were made between managers and non-managers on the 'Management' dimension and between medical staff and nurse staff employees on both the 'Management' and 'Quality of Care and Service' dimensions. 2 out of 21 items from the manager/non-manager comparison were found to have significant DIF, supporting the generalizability of Lynch, Barnes-Farell, and Kulikowich (1998). No items from the medical staff/nurse staff comparisons were found to have DIF. The DTF results indicated that in two out of the three comparisons 1 item could be removed to create dimensions free from DTF. Based on the current findings implications and future research are discussed. / Master of Science

No description available.

This simulation study focused on determining the effect of unequal sample sizes on statistical power of SIBTEST and Mantel-Haenszel procedures for detection of DIF of moderate and large magnitudes. Item parameters were estimated by, and generated with the 2PLM using WinGen2 (Han, 2006). MULTISIM was used to simulate ability estimates and to generate response data that were analyzed by SIBTEST. The SIBTEST procedure with regression correction was used to calculate the DIF statistics, namely the DIF effect size and the statistical significance of the bias. The older SIBTEST was used to calculate the DIF statistics for the M-H procedure. SAS provided the environment in which the ability parameters were simulated; response data generated and DIF analyses conducted. Test items were observed to determine if a priori manipulated items demonstrated DIF. The study results indicated that with unequal samples in any ratio, M-H had better Type I error rate control than SIBTEST. The results also indicated that not only the ratios, but also the sample size and the magnitude of DIF influenced the behavior of SIBTEST and M-H with regard to their error rate behavior. With small samples and moderate DIF magnitude, Type II errors were committed by both M-H and SIBTEST when the reference to focal group sample size ratio was 1:.10 due to low observed statistical power and inflated Type I error rates. / Ph. D.

XerC et XerD, deux recombinases impliquées dans la recombinaison site spécifique, résolvent les multimères d’ADN en monomères. Cette réaction se produit au niveau du site dif du chromosome, et nécessite le domaine C-terminale de la protéine de division cellulaire FtsK. Caulobacter crescentus est une bactérie aquatique de type Gram-négative qui se retrouve dans plusieurs environnements. Elle présente un cycle cellulaire asymétrique avec deux types de cellules distinctes. Cette propriété peut être utilisée pour synchroniser la croissance d’une population bactérienne pour permettre l’étude de l’expression de gènes à travers le temps et les liens entre le cycle cellulaire et le développement de la bactérie. La liaison à l’ADN et la capacité de former des complexes covalents (phosphotyrosyl) avec le site dif de C. crescentus (ccdif) ont été testé pour les recombinases de C. crescentus (ccXerC et ccXerD). Les deux recombinases ont eu une meilleure liaison au demi-site gauche de ccdif et sont incapable d’effectuer une liaison coopérative, contrairement à ce qui se produit au niveau du site dif de E. coli. La formation de complexes covalents a été testé en utilisant des «substrats suicides avec bris» marqués à la fluorescence ainsi que des protéines de fusion (marquées ou non à la fluorescence). Des complexes ADN-protéines résistants à la chaleur et au SDS ont été observé lors de la réaction de ccXerC et ccXerD de type sauvage avec ccdif, mais pas lors de la réaction de mutants avec le même ADN. Des complexes covalents phosphotyrosine sont formés de façon plus efficace sur les substrats suicides avec un bris au niveau du brin supérieur que ceux ayant un bris au niveau du brin inférieur. Dans les deux cas, c’est ccXerC qui est resté lié de façon covalente à l’ADN de ccdif. / In most bacteria, the chromosomal dimer resolution process is mediated by two tyrosine recombinases, XerC and XerD, which bind cooperatively and perform the recombination reaction at the dif site near the terminus of replication. This reaction also requires the C-terminal domain of the cell division protein FtsK. Caulobacter crescentus is an aquatic Gram-negative bacterium found in various environments. This bacterium has an asymmetric cell cycle which can be used to synchronize cell growth in order to study the temporal expression of a gene and the interconnection between the cell cycle and development. The binding activity and the formation of phosphotyrosyl complex of the C. crescentus recombinases, ccXerC and ccXerD, were tested on the C. crescentus dif (ccdif) site. Both ccXerC and ccXerD bound preferentially to the left half-site of ccdif and showed reduced cooperative binding, unlike what was found with the E. coli dif site. Covalent complex formation activity was tested by using fluorescently labelled linear “nicked suicide substrates” and labelled proteins. Heat and SDS-resistant protein-DNA complexes were formed when both wild-type ccXerC and ccXerD reacted with ccdif but not in the presence of active-site tyrosine mutant proteins. Phosphotyrosine complexes formed on the top-nicked suicide substrate were found to be more efficient than on the bottom-nicked suicide substrates and surprisingly ccXerC remained bound to both top and bottom-nicked ccdif suicide substrates.

XerC et XerD, deux recombinases impliquées dans la recombinaison site spécifique, résolvent les multimères d’ADN en monomères. Cette réaction se produit au niveau du site dif du chromosome, et nécessite le domaine C-terminale de la protéine de division cellulaire FtsK. Caulobacter crescentus est une bactérie aquatique de type Gram-négative qui se retrouve dans plusieurs environnements. Elle présente un cycle cellulaire asymétrique avec deux types de cellules distinctes. Cette propriété peut être utilisée pour synchroniser la croissance d’une population bactérienne pour permettre l’étude de l’expression de gènes à travers le temps et les liens entre le cycle cellulaire et le développement de la bactérie. La liaison à l’ADN et la capacité de former des complexes covalents (phosphotyrosyl) avec le site dif de C. crescentus (ccdif) ont été testé pour les recombinases de C. crescentus (ccXerC et ccXerD). Les deux recombinases ont eu une meilleure liaison au demi-site gauche de ccdif et sont incapable d’effectuer une liaison coopérative, contrairement à ce qui se produit au niveau du site dif de E. coli. La formation de complexes covalents a été testé en utilisant des «substrats suicides avec bris» marqués à la fluorescence ainsi que des protéines de fusion (marquées ou non à la fluorescence). Des complexes ADN-protéines résistants à la chaleur et au SDS ont été observé lors de la réaction de ccXerC et ccXerD de type sauvage avec ccdif, mais pas lors de la réaction de mutants avec le même ADN. Des complexes covalents phosphotyrosine sont formés de façon plus efficace sur les substrats suicides avec un bris au niveau du brin supérieur que ceux ayant un bris au niveau du brin inférieur. Dans les deux cas, c’est ccXerC qui est resté lié de façon covalente à l’ADN de ccdif. / In most bacteria, the chromosomal dimer resolution process is mediated by two tyrosine recombinases, XerC and XerD, which bind cooperatively and perform the recombination reaction at the dif site near the terminus of replication. This reaction also requires the C-terminal domain of the cell division protein FtsK. Caulobacter crescentus is an aquatic Gram-negative bacterium found in various environments. This bacterium has an asymmetric cell cycle which can be used to synchronize cell growth in order to study the temporal expression of a gene and the interconnection between the cell cycle and development. The binding activity and the formation of phosphotyrosyl complex of the C. crescentus recombinases, ccXerC and ccXerD, were tested on the C. crescentus dif (ccdif) site. Both ccXerC and ccXerD bound preferentially to the left half-site of ccdif and showed reduced cooperative binding, unlike what was found with the E. coli dif site. Covalent complex formation activity was tested by using fluorescently labelled linear “nicked suicide substrates” and labelled proteins. Heat and SDS-resistant protein-DNA complexes were formed when both wild-type ccXerC and ccXerD reacted with ccdif but not in the presence of active-site tyrosine mutant proteins. Phosphotyrosine complexes formed on the top-nicked suicide substrate were found to be more efficient than on the bottom-nicked suicide substrates and surprisingly ccXerC remained bound to both top and bottom-nicked ccdif suicide substrates.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / La investigaciÃn tuvo como objetivo detectar el funcionamiento diferencial del Ãtem (DIF) en dos tests de rendimiento - PortuguÃs y MatemÃticas - a travÃs del empleo de cuatro distintos procedimientos estadÃsticos: Mantel-Haenszel, CÃlculo del Ãrea entre las Curvas CaracterÃsticas de los Ãtems (CCIâs), ComparaciÃn del ParÃmetro b y el MÃtodo Estandarizado. Los resultados obtenidos indicaron que la cantidad de Ãtems con DIF sufrià un importante incremento cuando la variable âtipo de escuelaâ fue usada en la composiciÃn de los grupos estudiados, pero tambiÃn cuando las alternativas o distractores de los Ãtems estudiados poseian funcionamiento diferencial (DAF). AdemÃs, determinamos el grado de concordancia entre los mÃtodos en el proceso de diagnÃstico del DIF e identificamos sus principales limitaciones en la detecciÃn de los dos principales tipos de DIF - uniforme y no uniforme - en el Ãmbito de la TeorÃa de la Respuesta al Ãtem (TRI). AsÃ, sugerimos la adaptaciÃn del mÃtodo a las caracterÃsticas de las CCIâs de los Ãtems estudiados y, para ello, deberemos representar grÃficamente las CCIâs para los grupos comparados - el de referencia (GR) y el focal (GF) - y, posteriormente, eligir los mÃtodos mejor adaptados a las peculiaridades de las CCIâs, o sea, los mÃs eficaces para detectar el DIF.

Fibre reinforced polymer (FRP) composites have been used for strengthening concrete structures since early 1990s. More recently, FRP has been used for retrofitting concrete structures for high energy events such as impact and blast. Debonding at the FRP-to-concrete interface is one of the predominant failure modes for both static and dynamic loading. Although extensive research has been conducted on the static bond behaviour, the bond-slip mechanics under high strain rates is not well understood yet. This thesis is mainly concerned with the FRP-to-concrete bond behaviour under dynamic loading. Because debonding mostly occurs in the concrete adjacent to the FRP, the behaviour of concrete is of crucial importance for the FRP-to-concrete bond behaviour. The early emphasis of this thesis is thus on the meso-scale concrete modelling of concrete with appropriate consideration of static and dynamic properties. Issues related to FE modelling of tensile and compressive localization of concrete are first investigated in detail under static condition using the K&C concrete damage model in LS-DYNA. It is discovered for the first time that dilation of concrete plays an important role in the FRP-to-concrete bond behaviour. This has led to the development of a model relating the shear dilation factor to the concrete strength based on the modelling of a large number of static FRP-to-concrete shear tests, forming the basis for dynamic modelling. Concrete dynamic increasing factor (DIF) has been a subject of extensive investigation and debate for many years, but it is for the first time discovered in this study that mesh objectivity cannot be achieved in meso-scale modelling of concrete under high strain rate deformation. This has led to the development of a mesh and strain rate dependent concrete tension DIF model. This DIF model shall have wide applications in meso-scale modelling of concrete, not limited to the topic in this thesis. Based on a detailed numerical investigation of the FRP-to-concrete bond shear test under different loading rates, taking on the above issues into careful consideration, a slip rate dependent FRP-to-concrete dynamic bond-slip model is finally proposed for the first time. The FE predictions deploring this proposed bond-slip model are compaed with test results of a set of FRP-to-concrete bonded specimens under impact loading, and a FRP plated slab under blast loading, validating the model.