Les fondations profondes sont largement utilisées et peuvent subir différent types de sollicitations axiales. Ces dernières peuvent avoir un caractère monotone ou cyclique. Le développement de méthodes de calcul pour prévoir le comportement des pieux sous ces deux types de chargements en termes de tassement et d’évolution de capacité portante, est d’une grande utilité pour l’ingénierie géotechnique. Le travail de cette thèse est une contribution à une meilleure prévision du comportement des pieux sous charges axiales monotones et cycliques. La thèse comporte deux parties principales. La première partie traite la problématique de la prévision du tassement d’un pieu sous charge axiale monotone. Un intérêt particulier est porté aux tassements car le dimensionnement d’un pieu requiert une estimation adéquate à la fois de la capacité portante et du déplacement en tête de pieu. La méthode utilisée dans cette partie est celle des courbes de transfert. L’étude commence tout d’abord par une analyse statistique du modèle habituellement utilisé pour le calcul de tassement nommé Frank et Zhao. Ensuite, deux nouveaux modèles de calcul de tassement sont proposés et analysés. Une analyse comparative entre les trois modèles de calcul de déplacement d’un pieu est réalisée dans cette étude. L’objectif de cette analyse est d’estimer la dispersion des modèles de calcul en comparant les tassements calculés aux tassements mesurés et répertoriés dans la base de données de l’IFSTTAR. La seconde partie s’intéresse à la problématique des chargements cycliques. En effet, plusieurs types de structures sont sollicités cycliquement et peuvent subir des désordres du fait de l’accumulation de déplacements en tête des fondations. L’objectif de cette partie de la thèse est de proposer un modèle permettant de rendre compte principalement de l’évolution de la capacité portante du pieu au cours des cycles, et aussi de proposer une estimation des déplacements. Le modèle développé se base sur la méthode des courbes de transfert, et est associé à deux procédures de dégradation du frottement axial limite. Il est tout d’abord présenté et ensuite appliqué à un exemple théorique. Afin de valider ce modèle, ses résultats sont comparés aux résultats d’essais de pieux en vraie grandeur / Different types of loads can be applied to deep foundations which are widely used nowadays. Piles can be subjected to monotonic or cyclic loads. Thus, geotechnical engineering needs the development of calculation methods to predict the behavior of piles under these types of loads in terms of displacements and bearing capacity. The study presented in this thesis aims to ameliorate the prediction of the behavior of piles under axial monotonic and cyclic loads. This thesis is divided into two parts. The issue of the first part is the prediction of the settlement of a pile submitted to monotonic and axial load. This part focuses on the estimation of settlements because a correct design of a pile requires a correct estimation of bearing capacity as well as displacements. The approach used in this part is the load transfer method. The study starts by a statistical analysis of the t-z model of Frank and Zhao which is generally used for the calculation of pile settlements. Two new t-z models of settlement calculation are presented and analyzed later. Moreover, a comparative analysis between the three t-z models is presented in this study. The aim of this analysis is to estimate the dispersion of the models using the comparison between calculated settlements and measured settlements listed in IFSTTAR’s database. The second part of the thesis deals with the issue of cyclic loads. Indeed, different structures can be subjected to cyclic loads and disorders can be noted due to the accumulation of displacements at the top of the pile. The aim of this part is to propose a calculation model allowing essentially the estimation the evolution of bearing capacity during cycles. It permits also the estimation of displacements. The model developed in this part is based on load transfer approach, and is associated to two calculation methods of degradation of shaft friction. This model is firstly presented, and then it is applied to a theoretical case. In order to validate this model, its numerical results are compared to experimental results of full scale pile tests

This thesis examines several problems related to singly-constrained Monotropic Network Flow Problems. In the first part, a linear time algorithm that reduces the solution of a monotropic network flow problem with an additional linear equality constraint to the solution of lower dimensional subproblems is presented. Of the subproblems, at most one is a singly-constrained monotropic network flow problem while the others are unconstrained. If none of the subproblems is constrained, the algorithm provides a linear-time transformation of constrained to unconstrained monotropic network flow problems. Extensions to nonlinear and inequality constraints are given. In the second part the qualitative theory of sensitivity analysis for Unconstrained Minimum-Cost Flow Problems presented by Granot and Veinott [GV85] is extended to Minimum-Cost Flow Problems with one additional linear constraint. The departure from the unconstrained network structure is shown to have a profound effect on computational issues. Two natural extensions of the "less-dependent-on" partial ordering of the arcs given in [GV85] are presented. One is decidable in linear time while the other yields more information but is NP-complete in general. The Ripple Theorem gives upper bounds on the absolute value of optimal-flow variations as a function of variations in the problem parameter. Moreover, it shows how changes may "ripple down" throughout the network, decreasing in magnitude as one gets "further away" from the arc whose parameter initiated the change. The Theory of Substitutes and Complements presents necessary and sufficient conditions for optimal-flow changes to consistently have the same (or the opposite) sign in two given arcs. The complexity of determining Substitutes and Complements is shown to be NP-complete in general. However, for all intractable problems, families of cases arise from easily recognizable graph structures and can be computed in linear time. The Monotonicity Theory links the changes in the value of the parameters to the change in the optimal arc-flows. Bounds on the rates of changes are discussed. We further provide a number of practical situations where our theory may apply. We discuss some Multi-Period Multi-Product Inventory-Production models that can be formulated as nonlinear parametric network flow problems with one additional linear constraint. We then apply our theory to help decision makers understand qualitatively how to respond to changes in the environment such as machine breakdown, strike or variations in inventory carrying costs without additional computation. In a second example, we show how a Cash-Flow Management model can be formulated as a nonlinear parametric network flow problem with one additional linear constraint. The theory is then recommended as a method by which a decision maker could understand qualitatively how to respond to changes in the environment such as variations in interest rates, taxes or asset prices without any additional computation. / Business, Sauder School of / Graduate

京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第22222号 / 経博第610号 / 新制||経||292(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 関口 格, 教授 原 千秋, 准教授 陳 珈惠 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DGAM

We study two classes of problems within sublinear algorithms: data structures for approximate nearest neighbor search, and property testing of Boolean functions. We develop algorithmic and analytical tools for proving upper and lower bounds on the complexity of these problems, and obtain the following results: * We give data structures for approximate nearest neighbor search achieving state-of-the-art approximations for various high-dimensional normed spaces. For example, our data structure for 𝘢𝘳𝘣𝘪𝘵𝘳𝘢𝘳𝘺 normed spaces over R𝘥 answers queries in sublinear time while using nearly linear space and achieves approximation which is sub-polynomial in the dimension. * We prove query complexity lower bounds for property testing of three fundamental properties: 𝘬-juntas, monotonicity, and unateness. Our lower bounds for non-adaptive junta testing and adaptive unateness testing are nearly optimal, and the lower bound for adaptive monotonicity testing is the best that is currently known. * We give an algorithm for testing unateness with nearly optimal query complexity. The algorithm is crucially adaptive and based on a novel analysis of binary search over long paths of the hypercube.

This study investigates the monotonic and cyclic response of light-frame wood shear walls with and without openings. Effects of overturning restraint in the form of tie-down anchors and corner segments on light-frame shear walls with and without door and window openings were quantified. While the results are useful to refine a design methodology for shear walls containing openings, they also provide important knowledge that is needed to accurately quantify anchorage requirements for shear wall design, and assess remaining load and ductility capacity of wood frame buildings after earthquakes or hurricanes. Sixteen full-scale wall specimens were tested using monotonic and sequential phased displacement (SPD) patterns. A total of five different wall configurations, five anchorage, and two loading conditions were used. All walls were eight feet (2.4m) high. Straight wall specimens were forty feet (12.2m) long, whereas corner walls measured twelve feet (3.7m) in length. The analysis includes data from a previous investigation in order to further expand the scope of this study. Results reveal that ultimate capacity and stiffness increase with increasing overturning restraint. A shift in failure mode was observed when overturning restraints were omitted. Accumulated damage experienced by the wall specimens tested cyclically was fairly uniform, regardless of the amount of overturning restraint or size of openings present. / Master of Science

No / This paper presents a numerical modelling method to predict the shear behaviour of typical wall-stud cold-formed steel wall panels subjected to in-plane monotonic loads. In the research presented in this paper, different material and mechanical properties for cold-formed steel sheets, self-drilling screwed connectors and wall fixing boundary conditions were considered to explore the influence of sheath sheeting, connectors and fixing boundary conditions on the structural behaviour of selected wall panels. After the FE model being validated against experimental results, a parametric study was conducted and the comparison and analysis highlight the effect of different sheaths, connectors on to the structural shear behaviour of typical wall-stud cold-formed steel wall panels.

A microstructure-based plasticity-damage model is used to predict the mechanical behavior of commercially available AISI 4140 steel. Monotonic tension, compression and torsion tests were performed to obtain the set of plasticity and damage constants required for model calibration. Then, tension tests on Bridgman notched specimens were undertaken to study the damage-triaxiality dependence. Three different notch radii generated different levels of triaxiality at the notch. The modeled triaxiality-damage correlation was validated with SEM fracture surface analysis. Stress-strain correlations under different strain rate and temperature testing conditions were also studied. Little influence of the strain rate was observed. A preliminary study in high-porosity LENS materials was later performed, with satisfactory stress-strain correlation at two different temperatures on tension tests. Finally, a multistage fatigue model was used to predict life in AISI 4140 steel. The goal was to create a baseline for future application of these mathematical models into LENS manufactured materials in component design

No description available.

The effect of cyclic moisture infiltration on connections in light-frame wood buildings has received limited research attention. Specifically, the connections between wood-based sheathing materials (OSB, plywood) and solid wood studs are of interest. A comprehensive understanding of connection performance will enhance structure and material design, thereby improving the overall integrity and robustness of light-frame structures. The focus of this research project was to evaluate the strength and stiffness of wood-frame connections exposed to cyclic humidity conditioning. Nailed sheathing/stud connection samples were tested for lateral resistance following various periods of moisture exposure. Elastic stiffness, 5% offset yield load, maximum yield load, and failure yield were computed and analyzed using the data collected. The parameters were compared among connection specimens receiving either 0, 1, 5, 10, 15, 25, or 40 periods of cyclic moisture conditioning. In addition, the bearing resistances of the materials were investigated for application to the general dowel equations for calculating lateral connection values, the current basis for design of single dowel-type fastener connections between wood-based members. An x-ray density profilometer was used to observe the de-densification processes within the composite sheathing materials throughout the moisture conditioning regime. Results indicated moderate to extreme changes in the performance of cycled connections involving lower density sheathing materials. Higher density sheathing materials performed favorably at each cycle test period. Comparisons to the yield model were similar to the control results, but usually differed as cycling increased. Analysis of connection performance following cyclic moisture loading is a vital component in developing a holistic model for service-life prediction of nailed connections in light-frame residential construction. / Master of Science

This research has been aimed to introduce larger moment carrying connections for any type of buildings, in particular the metal building industry. A total of four connection configurations, namely eight-bolt extended four wide, eight-bolt extended stiffened, six bolt flush unstiffened and twelve bolt extended unstiffened, have been investigated. The last two configurations are proposed whereas the first two configurations have been tested before, but the design procedures need to be validated against the test results. Design procedures, namely yield line analysis and bolt force models, were proposed to calculate moment capacity for end-plate yielding, moment capacity at bolt rupture with prying action and moment capacity at bolt rupture without prying action. The calculated values from these procedures were compared with the values obtained from the experimental test data available (whether from the literature or from this testing program). The experimental data from already tested configurations, eight-bolt extended four wide and eight-bolt extended stiffened, was analyzed. It was concluded that for the eight-bolt extended four wide configuration, the experimental values matched with the calculated values. For the eight-bolt extended stiffened configuration reasonable match was found between the experimentally obtained data and theoretically calculated values only for shallower depths. Hence, it was concluded that two deeper tests need to be performed for this configuration. A full-scale testing program was conducted for ten specimens covering three configurations. The two new configurations (six bolt flush unstiffened and twelve bolt multiple row extended unstiffened) were designed for a shallow and deep beam depth and the behavior of each depth observed for a thin end-plate and a thick end-plate respectively (four tests for each configuration). Also, two deep beam tests, one each for thick and thin plate behavior, were done for the eight-bolt extended stiffened configuration. Based on the comparison, it was determined whether the predicted values were in reasonable agreement with the experimental values or not. The design procedures for both the new configurations appear to be validated for a range of design parameters. The calculated moment capacities for bolt rupture, based on the nominal material properties, were found to be safe when compared with the experimentally obtained moments. The calculations for end-plate yield moments was within ±10% of the experimental yield moment. Also, for the deep tests for eight-bolt extended stiffened the yield line analysis seems to be a valid model and the bolt force model appears to be safe in comparison to the experimental values. / Master of Science