Dissertations / Theses on the topic 'Bottom of variable thickness'

Il principale obiettivo tecnologico di questo progetto di ricerca è lo sviluppo di una tecnologia di laminazione a freddo innovativa, tramite la distribuzione di aree di diverso spessore sul prodotto finito, in modo che il materiale possa essere distribuito al meglio dove è necessario, per soddisfare i requisiti di resistenza e/o rigidezza. Per quanto riguarda il processo di laminazione a freddo, l'obiettivo è quello di effettuare un upgrade della soluzione 1D, nota anche come profile- rolling, già presente in molte aree industriali: il processo di laminazione a zone denominato patch-rolling. Esso rappresenta un'innovazione assoluta consentendo di ottenere zone di maggiore spessore già in fase di laminazione a freddo, nel rispetto delle dimensioni richieste e del posizionamento risultante dalle mappe delle sollecitazioni. L'obiettivo è dunque quello di sviluppare uno strumento di supporto in grado non solo di indicare la fattibilità delle patch, come richiesto dal progetto, ma anche di suggerire eventuali modi che come la dimensione delle patch, variazione di spessore, raggio di raccordo ecc. La prima fase del progetto è stata la scelta del materiale dalle migliori prestazioni per i processi di laminazione e stampaggio. A tale scopo, diversi materiali sono stati caratterizzati mediante test mono-assiali e di formabilità. I primi hanno permesso di identificare le curve di flusso plastico e le principali caratteristiche meccaniche del materiale, mentre quelle di formabilità hanno permesso di valutare l'influenza dell'anisotropia e tracciare i diagrammi limite di formabilità grazie a metodi ottici avanzati. La fase successiva, quindi, è stata lo sviluppo di un modello agli elementi finiti del processo di laminazione a patch in cui venivano richieste solo le proprietà meccaniche del materiale. Dopo che il modello è stato con figurato e convalidato con successo, sono state studiate diverse configurazioni di patch con il modello FEM già descritto. A causa della deformazione non bilanciata nel senso della lunghezza del laminato, è stato sviluppato anche un modello analitico più semplice e veloce in MatLAB R per fornire un'anteprima della forma del laminato con il solo ne di valutare la qualità della distribuzione delle patch. Quindi, lo studio di come i principali fattori (variabili principali) coinvolti nel processo di laminazione influenzino la fattibilità del processo stesso, è stato eseguito attraverso una sperimentazione FEM organizzata secondo le tecniche del Design of Experiments (DoE). L'analisi dei risultati del DoE ha permesso di sviluppare e analizzare un'applicazione di casi reali, a partire dalla distribuzione delle patch no alla realizzazione del prototipo.<br>The main technological aim of this research project is the development of an innovative cold rolling technology with the distribution of different thickness areas on the finished product, so the material can be best distributed where it is needed, to satisfy strength and/or stiffness requirements. Regarding the cold rolling process, the aim is to achieve a better optimisation of the 1D solution, also known as profile-rolling, already seen in many industrial areas: the multi-thickness rolling process named patch-rolling. It represents an absolute innovation by allowing the higher thickness zones already used in the process of cold rolling to be obtained, with respect to the requested sizes and positioning resulting from the stress map. To this purpose the goal is to develop a support tool able not only to indicate the feasibility of the patches, as requested by the project, but also which suggests any amendments needed such as dimension of the patches, thickness variation, fillet radius etc. The first stage of the project was the choice of the material with best performances in both rolling and stamping processes. To this purpose, several possible challengers were deeply investigated through uni-axial and formability tests. First ones were led to identify the flow plastic curves and principal mechanical characteristics of the material while the formability ones allowed the anisotropy influence and the Forming Limit Diagrams to be evaluated thanks to advanced optical methods. The next stage, indeed, was the development of a finite element method model of the patch-rolling process in which only mechanical properties of the material were requested. After the model was successfully set-up and validated, several patch-configurations were investigated with the FEM model already described. Due to the unbalanced length-wise direction deformation of the work-piece a more simple and fast analytical model was also developed in MatLAB R to provide a pre-view of the shape of the laminated work-piece only to evaluate the quality of the patch distribution. Then the study of how the principal factors (main variables) involved in the rolling process may affect the feasibility of the process itself, was performed through a FEM experimentation organized according to the Design of Experiments (DoE) techniques. The analysis of the results of the DoE allowed a real case application to be developed and analysed, starting from the patch distribution until the prototype realisation.

Cette thèse aborde la problématique de la conception optimale de structures composites stratifiées d’épaisseur variable. Les variables d’empilement définissent un problème d’optimisation combinatoire et des espaces de décisions de grande taille et potentiellement multimodaux. Les algorithmes d’optimisation stochastiques permettent de traiter ce type de problème et de tirer profit des performances et de l’anisotropie des plis composites pour l’allègement des structures composites stratifiées. Le but de cette étude est double : (i) développer un algorithme d’optimisation dédié aux composites stratifiés d’épaisseur variable et (ii) estimer le potentiel des composites stratifiés pour la maîtrise des performances aérodynamiques d’une pale de CROR composite.Dans la première partie de cette thèse, un algorithme évolutionnaire est spécialisé pour l’optimisation de tables de drapage et la gestion d’un ensemble de règles de conception représentatif des pratiques de l’industrie. Pour se faire, un encodage spécifique des solutions est proposé et des opérateurs de variations spécialisés sont développés.Dans la deuxième partie, l’algorithme est enrichi d’une technique de guidage basée sur l’exploitation d’un espace auxiliaire afin d'accroître son efficacité et d’intégrer davantage de connaissances des composites dans la résolution du problème.Finalement, la méthode est appliquée pour la conception d’une pale de CROR composite à l’échelle de la maquette de soufflerie. Au préalable, des processus itératifs de mise à froid et mise à chaud de la pale sont mis en place afin d’estimer la forme de la pale au repos et l’état de contraintes dans la pale en fonctionnement<br>This thesis deals with the optimal design of variable-thickness laminated composite structures. The stacking variables define a combinatorial optimization problem and large decision spaces which are potentially multimodal. Stochastic optimization algorithms allow solving this type of problem and allow taking advantage from the performance and the anisotropic nature of unidirectional composite plies to lighten laminated composite structures.The purpose of this study is twofold: (i) developing an optimization algorithm dedicated to variable-thickness laminated composites and (ii) assessing the potential of laminated composites in influencing the aerodynamic performances of a composite CROR blade.Firstly, an evolutionary algorithm is specialized in order to optimize layup tables and handle a set of design guidelines which is representative of industrial practices. In this purpose, a specific encoding of the solutions is suggested and specialized variation operators are developed.Secondly, the algorithm is enriched with a guiding technique based on the exploitation of an auxiliary space in order to improve its efficiency and to include further composites-related knowledge for the resolution of the problem.Finally, the method is applied for the design of a reduced-scale composite CROR blade intended for wind-tunnel testing. Beforehand, iterative processes are implemented to estimate the shape of the non-operating blade and the stress state within the operating blade

Thesis: S.M., Massachusetts Institute of Technology, Department of Architecture, 2015.<br>Cataloged from PDF version of thesis. Page 61 blank.<br>Includes bibliographical references (pages 58-59).<br>Much of architecture's earliest material palettes and construction methods are often referred to today as legacy materials - those primarily consisting of various types of stone and masonry construction. While these materials are often conceptually thought of as being solid, monolithic, and even homogeneous, in actuality they rely on logics of assembly more akin to contemporary sandwich structures, which are laminar assemblies typically composed of two or more stressed skins and either a solid or cellular core that binds them together. While it is still common to use ancient materials in contemporary architecture, the construction methods and techniques used several hundred years ago are no longer appropriate for today's buildings. This thesis however, argues for a newfound relevance of their influence on contemporary and even future material selections and methods. Specifically, this thesis explores the potentials of composite sandwiches varying in thickness and material in search of architectural possibilities whose structural, formal, and aesthetic implications are a result of tuning multiple influences. Variable thickness is used here as a strategy for enabling a range of architectural and tectonic conditions, all within the same heterogeneous but integrated laminar assemblies. While most commercial products in the realm of composite sandwiches are of uniform thickness in section, this thesis suggests a method for constructing sandwiched elements with variable thickness. This is done primarily through a process of infill and backfill using expanding urethane foam as a medium which creates the so called "core" of the sandwich between two skins. This investigation works through a series of small scale prototypes, each of which focus on a particular tectonic, spatial, or structural condition. These mock ups are meant to serve as didactic artifacts, providing feedback with which to incorporate and speculate upon larger architectural propositions through drawing and representation. The end result is a set of architectural proposals which suggest the beginnings of new design methodologies.<br>by Nazareth Ekmekjian.<br>S.M.

Early R-functions theory [1] combined with variational methods have been applied to linear [2] and nonlinear vibration problems [3,4] of the shallow shells theory of the constant thickness. In the present study, we first apply R-functions theory in order to investigate the geometrically nonlinear vibrations of orthotropic shallow shells of complex shape with variable thickness. Mathematical formulation is made in the framework of classical geometrically nonlinear theory of thin shallow shells. For a discretization of the original system in time, approximation of unknown functions is carried out by using a single mode approach. In order to construct a system of basic functions, the proposed algorithm includes sequence of the linear problems such as finding eigen functions of the linear vibrations of shallow shells with variable thickness and auxiliary tasks of the elasticity theory. The linear problems are solved by the R-functions method. The developed approach allows reducing the original problem to the corresponding problem of solving nonlinear ordinary differential equations (ODEs), whose coefficients are presented in analytical form. In order to solve the obtained system of ODEs the Bubnov-Galerkin method is applied. The proposed algorithm is implemented within an automated system POLE-RL [1]. Numerical examples of large-amplitude flexible vibrations of shallow orthotropic shells with complex shape and variable thickness are introduced demonstrating merits and advantages of the R-functions method. Comparison of the obtained results regarding shells with rectangular plans with the other methods confirms the reliability of the proposed method.

One of the recognized weaknesses of land surface models as used in weather and climate models is the assumption of constant soil thickness because of the lack of global estimates of bedrock depth. Using a 30-arc-s global dataset for the thickness of relatively porous, unconsolidated sediments over bedrock, spatial variation in soil thickness is included here in version 4.5 of the Community Land Model (CLM4.5). The number of soil layers for each grid cell is determined from the average soil depth for each 0.9 degrees latitude x 1.25 degrees longitude grid cell. The greatest changes in the simulation with variable soil thickness are to baseflow, with the annual minimum generally occurring earlier. Smaller changes are seen in latent heat flux and surface runoff primarily as a result of an increase in the annual cycle amplitude. These changes are related to soil moisture changes that are most substantial in locations with shallow bedrock. Total water storage (TWS) anomalies are not strongly affected over most river basins since most basins contain mostly deep soils, but TWS anomalies are substantially different for a river basin with more mountainous terrain. Additionally, the annual cycle in soil temperature is partially affected by including realistic soil thicknesses resulting from changes in the vertical profile of heat capacity and thermal conductivity. However, the largest changes to soil temperature are introduced by the soil moisture changes in the variable soil thickness simulation. This implementation of variable soil thickness represents a step forward in land surface model development.

With the rapid development of computer programming and systems, complex analytical shell problems are being tackled more than ever before. Finite element analysis has been the most widely used numerical method in modelling shell structures. In this thesis a more reliable and accurate method is demonstrated, in which the dynamic stiffness is derived from the exact differential equations, resulting in fewer elements and reduced computing time. Free undamped vibration analysis for constant and variable thickness cylindrical shells is carried out by deriving the dynamic stiffness matrix directly from the three differential equilibrium equations after transforming them to three ordinary differential equations and using the boundary value solver COLSYS. The natural frequencies and mode shapes are determined by using the Wittrick-Williams algorithm. A method of handling the effect of fixed end member frequencies below a limiting trial frequency is presented. Numerical examples for beams with constant and variable thicknesses and shells with constant and variable thicknesses are shown, and the method is validated by convergence studies and the handling of fixed end frequencies. A comparison of shell problem results with those of other authors is given, and a parametric study is carried out by varying the length to radius, radius to thickness, and thickness variation ratios. Results are obtained for different circumferential and axial harmonic numbers. Mode shapes are explained, for different shell thickness ratios and end conditions, and significant conclusions are drawn. From all the study and results obtained in this thesis it is ascertained that using the Wittrick-Williams algorithm in combination with COLSYS is a successful approach, and good agreements are effectively achieved for free undamped vibration of cylindrical shells with constant and variable thickness.

Le simulazioni numeriche rappresentano uno degli strumenti più rilevanti per la valutazione delle performance di compressori ed espansori di tipo scroll. I modelli monodimensionali sono una soluzione importante per ottenere informazioni veloci ed affidabili sul funzionamento generale di queste macchine, mentre le analisi CFD forniscono dati più accurati sul loro comportamento transitorio, in particolare durante i processi di aspirazione e scarica. Sfortunatamente, la disponibilità di strumenti open-source per condurre analisi di basso ordine su macchine volumetriche è ancora limitata. In questa tesi, l'autore ha realizzato sia analisi monodimensionali che simulazioni di fluidodinamica computazionale su macchine di tipo scroll. Il software scelto per le simulazioni a basso ordine è PDSim, un codice open-source rilasciato di recente e utilizzato per le analisi quasistatiche di macchine volumetriche. In alternativa, il ben noto software OpenFOAM è stato adottato per realizzare analisi tridimensionali. Contribuendo all'espansione delle possibilità di modellazione open-source di macchine scroll, sono stati realizzate due serie di librerie software per OpenFOAM. La prima, CoolFOAM, è un wrapper che connette OpenFOAM con la libreria termofisica CoolProp. La seconda, scrollFOAM, estende le possibilità del software permettendo di realizzare simulazioni di scroll totalmente tridimensionali. Tutti gli strumenti numerici utilizzati in questa tesi sono stati validati utilizzando dati sperimentali e modelli recuperati in letteratura. In particolare, i risultati di PDSim e OpenFOAM sono stati confrontati tra di loro e con una serie di misurazioni effettuate su un compressore scroll di origine industriale. Le librerie sviluppate sono state impiegate nell'ambito di diverse analisi, tra cui uno studio degli effetti di gas reale sulle simulazioni CFD di un espansore scroll e un'analisi multi-componente di una porzione di un ciclo ORC. Dal punto di vista del compressore, è stato studiato il comportamento della macchina in condizioni fuori progetto. In aggiunta, è stata effettuata un'analisi di sensibilità sull'effetto dei trafilamenti all'interno della macchina. In seguito, sia scrollFOAM che PDSim sono stati adattati per supportare una nuova formulazione geometrica per geometrie scroll a spessore variable. L'adozione di spirali a spessore variabile è teoricamente associata con diversi vantaggi, tra cui una riduzione dei costi e un aumento dei rendimenti. Infine, PDSim e OpenFOAM sono stati accoppiati per formare una catena open-source per la progettazione e ottimizzazione di scroll a spessore costante e variabile. Questa catena è stata testata realizzando un'ottimizzazione di tipo kriging centrata su un compressore a spessore variabile operante in condizioni di under-compression. Partendo da un riferimento iniziale a spessore costante, la geometria ottimizzata ha dimostrato di raggiungere prestazioni comparabili associate a una consistente riduzione dei costi di produzione.<br>Numerical analyses are one of the most important tools to evaluate the performance of scroll compressors and expanders. 0-D models are an important solution to obtain fast and reliable information on the general performance of these machines, while CFD simulations give more accurate data on the transient behaviour of the compressors, especially during the suction and discharge process. Unfortunately, the availability of open-source tools for low-order and CFD simulations of positive displacement machines is still limited. In this thesis, the author has performed both 0-D and CFD analyses of scroll machines. The numerical tool selected for low-order simulations is a recently published open-source tool for the quasi-static analysis of positive displacement machines, PDSim. On the other hand, the well known OpenFOAM software has been adopted for 3-D numerical analyses. Contributing to the expansion of the open-source modelling capabilities of scrolls, two sets of libraries for OpenFOAM have been developed. The first, CoolFOAM, is a wrapper that couples OpenFOAM with the thermophysical library CoolProp. The second, scrollFOAM, extends the software capabilities allowing to perform full 3-D simulations of scrolls. All the numerical tools employed in this thesis have been validated against experimental references and test cases retrieved in the literature. In particular, the results of PDSim and OpenFOAM simulations have been compared to each other and to a set of measurements taken from an industrial scroll compressor. The libraries have been employed for different analyses, including the investigation of real gas effects on CFD simulations of a scroll expander and a multi-component numerical investigation of a portion of an ORC. From the compressor point of view, the behaviour in off-design conditions and a sensitivity analysis of gap widths have been objects of additional studies. Lastly, both scrollFOAM and PDSim have been adapted to support a new geometrical formulation for variable thickness scroll geometries. The adoption of variable thickness wraps is theoretically associated with several advantages, including costs reduction and increased efficiencies. Finally, PDSim and OpenFOAM have been coupled to form an open-source tool-chain for the design and optimization of constant and variable thickness scrolls. This tool-chain has been tested by performing a kriging optimization process of a variable thickness compressor working in under-compression. Starting from a constant thickness reference, the optimized geometry reached comparable performance with a consistent reduction of the manufacturing costs.

Geometrically nonlinear vibrations of laminated shallow shells with layers of variable thickness are studied. Nonlinear equations of motion for shells based on the first order shear deformation and classical shells theories are considered. In order to solve this problem we use the numerically-analytical method proposed in work [1]. Accordingly to this approach the initial problem is reduced to consequences of some linear problems including linear vibrations problem, special elasticity ones and nonlinear system of ordinary differential equations in time. The linear problems are solved by the variational Ritz’ method and Bubnov-Galerkin procedure combined with the R-functions theory [2]. To construct the basic functions that satisfy all boundary conditions in case of simply-supported shells we propose new solutions structures. The proposed method is used to solve both test problems and new ones.

A rather thorough and novel buckling analysis of an axially-loaded orthotropic circular cylindrical shell is formulated. The analysis assumes prebuckling rotations are negligible and uses a unique re-defining of the orthotropic material properties in terms of a so-called geometric mean isotropic (GMI) material. Closed-form expressions for the buckling stress in terms of cylinder geometry and orthotropic material properties are presented, the particular closed form depending on the specific character of the orthotropic material relative to the GMI material. With the formulation, the specific character of the buckling deformations - e.g., axisymmetric or nonaxisymmetric, the number of axial and circumferential waves - can be established. By using the maximum radius of curvature of an elliptical cross section in this formulation, the analysis is used to demonstrate the detrimental effects of an elliptical cross section on axial buckling capacity when compared to a circular cross section with the same circumference. Using the circumferentially-varying radius of curvature of an elliptical cross section, the analysis is then further used as the basis for developing two methods for improving the axial buckling capacity of elliptical cylinders. The first approach involves varying the wall thickness of an isotropic elliptical cylinder with circumferential position. Uniformly stable elliptical cross sections which preserve the same critical stress, critical load, or volume of an axially loaded circular cylinder of the same circumference are designed with the formulation. The second approach involves maintaining a uniform wall thickness but varying the orthotropic material properties with circumferential position. This approach is applied to a cylindrical lattice structure where it is assumed that the ribs are dense enough to be able to describe the lattice structure by means of an equivalent homogenized material. The orthotropic properties of the homogenized material are varied by varying the lattice rib angle with circumferential position. Considerable recovery of the axial buckling capacity of the variable-rib-angle design elliptical cylinder compared to the same cylinder constructed in isogrid fashion is demonstrated. In fact, recovery relative to an isogrid circular cylinder of the same circumference is demonstrated. For both approaches confirming finite element models are used to verify the findings. The two different approaches are compared, and finally the two approaches are recognized as special cases of a more general design philosophy.<br>Ph. D.

Cold-flow tests were performed using a simulated Air Augmented Rocket (AAR) operating as a mixer-ejector in order to investigate the effects of varied primary nozzle lip thickness on mixing and entrainment. The simulated primary rocket ejector was supplied with nitrogen at a maximum chamber stagnation pressure of 1712 psi, and maximum flow rate of 1.67 lbm/s. Secondary air was entrained from a plenum, producing pressures as low as 6.8 psi and yielding maximum stagnation pressure ratios as high as 160. The primary ejector nozzles each had an area ratio of approximately 20, yielding average primary exit Mach numbers between 4.34 and 4.57. The primary flow was ejected into an 18.75 inch-long mixing duct with a rectangular cross-sectional area of 2.10 in2. The secondary flow was entrained into the mixing duct through a total cross section of 0.94 in2. Two mixing duct configurations were used, one with plexiglass upper and lower surfaces for flow visualization and one with pressure ports along the lower surface for primary plume measurements. Shadowgraph images were used to characterize the mixing duct flow field, while pressure and temperature instrumentation allowed for calculation of various ejector performance characteristics. Experimentally-calculated performance characteristics were compared to inviscid theoretical predictions. Varying degrees of flow field asymmetry were observed with each nozzle. Test repeatability was found to be excellent for all nozzles. Several distinct phenomena were observed in both the primary plume and secondary streams. The duration of secondary flow choking was found to be inversely proportional to nozzle lip thickness, due to the primary plume being physically closer to the secondary flow with a thinner nozzle lip. This indicated that the ejector’s ability to choke the secondary flow is primarily an inviscid phenomenon. Secondary flow blockage was demonstrated in two consecutive tests using the thickest nozzle lip. Only the left secondary duct became blocked in each case. Blockage was only demonstrated in the centerline pressure configuration, so no visual evidence was able to support the blocked flow theory. At every pressure ratio, entrainment ratio was shown to increase with nozzle lip thickness. The original conical nozzle produced the largest level of entrainment, indicating that the angle of primary flow impingement was the largest contributing factor to secondary entrainment. The increase in efficiency resulting from a bell-mouth nozzle was less than the increase in entrainment efficiency of a conical nozzle, indicating that the conical design was more efficient overall for air augmented rocket applications.

Water is essential to life on Earth. Since water exists in all three phases (solid, liquid, and gas) on Earth, it exists in various reservoirs throughout the planet that compose the hydrologic cycle, and its movement through these reservoirs requires energy. Thus, water is a key component of the energy balance of the Earth. Despite its importance, its representation in modeling and dataset generation is problematic. Here, the depiction of three phenomena, ocean surface turbulent fluxes, humidity inversions, and groundwater, are assessed, and suggestions for improvements of their representations are made. First, ocean surface turbulent fluxes, including those of moisture (latent heat flux), heat (sensible heat flux), and momentum (wind stress), from reanalysis, satellite-derived, and combined products which are commonly used to produce climatologies and to evaluate global climate models are compared to in situ observations from ship cruises to ascertain which products are the least problematic. The National Aeronautics and Space Administration’s reanalysis, the Modern Era Retrospective Analysis for Research and Applications, is the least problematic for all three fluxes, while a couple of others are the least problematic for only one of the three fluxes. Also, the product biases are disaggregated into uncertainties from the grid cell mean quantities, or bulk variables, used plus the residual uncertainties which includes the algorithm uncertainties due to the parameterization used to relate the small-scale turbulent processes to the large-scale bulk variables. The latter contribute the most to the majority of product latent heat fluxes, while both uncertainties can contribute the most to product sensible heat fluxes and wind stress. Thus, both algorithms and bulk variables need to be improved in ocean surface flux datasets. Second, humidity inversion climatologies in five reanalyses are evaluated. Humidity inversions, similar to its thermal counterpart, are layers in which specific humidity increases with height rather than the usual decrease with height. These are especially persistent in the polar regions in autumn and winter. However, Arctic inversions are the strongest in summer corresponding to the time of year that low cloud cover is the highest. Comparing the reanalysis inversions to radiosonde observations reveals some problems with the realization of humidity inversions in reanalyses including the misrepresentation of the diurnal cycle and of the overproduction of inversions in areas outside the polar regions. Finally, the simulation of groundwater in the Community Land Model (CLM) as used in the Community Earth System Model is made more realistic by including variable soil thickness. Because the bottom of the model soil column is placed at effectively bedrock, the unconfined aquifer model currently used in CLM is removed and a zero bottom water flux is put in place. The removal of the unconfined aquifer allows the simulation of groundwater to not be treated separately from soil moisture. The model is most affected where the number of soil layers is reduced from the original constant 10 layers and largely unaffected where the number of soil layers is increased except for baseflow where the mean annual range in rainfall is large.

Au cours de cette thèse, un modèle potentiel résolvant les équations d’Euler-Zakharov a été développé dans le but de simuler la propagation de vagues et d’états de mer irréguliers et multi-directionnels, du large jusqu’à la côte, sur des bathymétries variables. L’objectif est de représenter les effets non-linéaires et dispersifs le plus précisément possible pour des domainescôtiers bidimensionnels (dans le plan horizontal) de l’ordre de quelques kilomètres.La version 1DH initiale du modèle, résolvant le problème aux limites de Laplace à l’aide de schémas aux différences finies d’ordre élevé dans la direction horizontale combinés à une approche spectrale sur la verticale, a été améliorée et validée. L’implémentation de conditions aux limites de type Dirichlet et Neumann pour générer des vagues dans le domaine a été étudiée en détail. Dans la pratique, une zone de relaxation a été utilisée en complément deces conditions pour améliorer la stabilité du modèle.L’expression analytique de la relation de dispersion a été établie dans le cas d’un fond plat. Son analyse a montré que la représentation des effets dispersifs s’améliorait significativement avec l’augmentation de la résolution sur la direction verticale (i.e. avec le degré maximal de la basede polynômes de Tchebyshev utilisée pour projeter le potentiel des vitesses sur la verticale).Une étude de convergence menée pour des ondes solitaires modérément à fortement non-linéaires a confirmé la convergence exponentielle avec la résolution verticale grâce à l’approche spectrale, ainsi que les convergences algébriques en temps et en espace sur l’horizontale avec des ordres d’environ 4 (ou plus) en accord avec les schémas numériques utilisés.La comparaison des résultats du modèle à plusieurs jeux de données expérimentales a démontré les capacités du modèle à représenter les effets non-linéaires induits par les variations de bathymétrie, notamment les transferts d’énergie entre les composantes harmoniques, ainsi que la représentation précise des propriétés dispersives. Une formulation visco-potentielle a également été implémentée afin de prendre en compte les effets visqueux induits par la dissipation interne et le frottement sur le fond. Cette formulation a été validée dans le cas d’une faible viscosité avec un fond plat ou présentant une faible pente.Dans le but de représenter des champs de vagues 2DH, le modèle a été étendu en utilisant une discrétisation non-structurée (par nuage de points) du plan horizontal. Les dérivées horizontales ont été estimées à l’aide de la méthode RBF-FD (Radial Basis Function-Finite Difference), en conservant l’approche spectrale sur la verticale. Une étude numérique de sensibilité a été menée afin d’évaluer la robustesse de la méthode RBF-FD, en comparant différents types de RBFs, avec ou sans paramètre de forme et l’ajout éventuel d’un polynôme. La version 2DH du modèle a été utilisée pour simuler deux expériences en bassin, validant ainsi l’approche choisie et démontrant son applicabilité pour simuler la propagation 3D des vagues faisant intervenir des effets non-linéaires. Dans le but de réduire le temps de calcul et de pouvoir appliquer le code à des simulations sur de grands domaines, le code a été modifié pour utiliser le solveur linéaire direct en mode parallèle<br>In this work, a potential flow model based on the Euler-Zakharov equations was developed with the objective of simulating the propagation of irregular and multidirectional sea states from deep water conditions to the coast over variable bathymetry. A highly accurate representation of nonlinear and dispersive effects for bidimensional (2DH) nearshore and coastal domains on the order of kilometers is targeted.The preexisting 1DH version of the model, resolving the Laplace Boundary Value problem using a combination of high-order finite difference schemes in the horizontal direction and a spectral approach in the vertical direction, was improved and validated. The generation of incident waves through the implementation of specific Dirichlet and Neumann boundary conditions was studied in detail. In practice, these conditions were used in combination witha relaxation zone to improve the stability of the model.The linear dispersion relation of the model was derived analytically for the flat bottom case. Its analysis showed that the accuracy of the representation of dispersive effects improves significantly by increasing the vertical resolution (i.e. the maximum degree of the Chebyshev polynomial basis used to project the potential in the vertical). A convergence study conducted for moderate to highly nonlinear solitary waves confirmed the exponential convergence in the vertical dimension owing to the spectral approach, and the algebraic convergence in time and in space (horizontal dimension) with orders of about 4 (or higher) in agreement with the numerical schemes used.The capability of the model to represent nonlinear effects induced by variable bathymetry, such as the transfer of energy between harmonic components, as well as the accurate representation of dispersive properties, were demonstrated with comparisons to several experimental data sets. A visco-potential flow formulation was also implemented to take into account viscous effects induced by bulk viscosity and bottom friction. This formulation was validated inthe limit of small viscosity for mild slope bathymetries.To represent 2DH wave fields in complex nearshore domains, the model was extended using an unstructured discretization (scattered nodes) in the horizontal plane. The horizontal derivatives were estimated using the RBF-FD (Radial Basis Function - Finite Difference) method, while the spectral approach in the vertical remained unchanged. A series of sensitivity tests were conducted to evaluate numerically the robustness of the RBF-FD method, including a comparison of a variety of RBFs with or without shape factors and augmented polynomials. The 2DH version of the model was used to simulate two wave basin experiments, validating the approach and demonstrating the applicability of this method for 3D wave propagation, including nonlinear effects. As an initial attempt to improve the computational efficiency ofthe model for running simulations of large spatial domains, the code was adapted to use a parallelized direct linear solver

The current clinical management of abdominal aortic aneurysm (AAA) disease is based on measuring the aneurysm maximum diameter to decide when timely intervention can be recommended to a patient. However, other parameters may also play a role in causing or predisposing the AAA to either an early or delayed rupture relative to its size. Therefore, patient-specific assessment of rupture risk based on physical principles such as individualized biomechanics can be conducive to the development of a vascular tool with translational potential. To that end, the present doctoral research materialized into a framework for image based patient-specific vascular biomechanics assessment. A robust generalized approach is described herein for image-based volume mesh generation of complex multidomain bifurcated vascular trees with the capability of incorporating regionally varying wall thickness. The developed framework is assessed for geometrical accuracy, mesh quality, and optimal computational performance. The relative influence of the shape and the constitutive wall material property on the AAA wall mechanics was explored. This study resulted in statistically insignificant differences in peak wall stress among 28 AAA geometries of similar maximum diameter (in the 50 – 55 mm range) when modeled with five different hyperelastic isotropic constitutive equations. Relative influence of regionally varying vs. uniform wall thickness distribution on the AAA wall mechanics was also assessed to find statistically significant differences in spatial maxima of wall stresses, strains, and strain energy densities among the same 28 AAA geometries modeled with patient-specific non-uniform wall thickness and two uniform wall thickness assumptions. Finally, the feasibility of estimating in vivo wall strains from individual clinical images was evaluated. Such study resulted in a framework for in vivo 3D strain distributions based on ECG gated, unenhanced, dynamic magnetic resonance images acquired for 20 phases in the cardiac cycle. Future efforts should be focused on further development of the framework for in vivo estimation of regionally varying hyperelastic, anisotropic constitutive material models with active mechanics components and the integration of such framework with an open source finite element solver with the goal of increasing the translational potential of these tools for individualized prediction of AAA rupture risk in the clinic.

Residual state of stress in structural materials affect positively or negatively behaviour of component parts. The goal of this scope is not to deal with possible process of creating residual stresses neither about elimination of residual stress, but is focused how to determine magnitude of residual stress by hole-drilling method. We need to know magnitude and direction (angular orientation) of principal stresses to determine how residual state of stress affects behaviour of specimen. The most widely used modern technique for measuring residual stresses is hole-drilling strain-gage method. Hole-drilling method is in scope of this paper and is restricted only for measuring uniform residual stresses of steel specimens with finite dimensions. Structural, linear, elastic and isotropic material model is used with material properties =0,3 and E=2,1[10]^5 MPa. For correct application of this method we need to determine calibration coefficients “a“ and “b“ first. These coefficients are used to determine magnitude and direction of residual stresses in specific depth and diameter of drilled hole for materials with finite dimensions. Geometry and shape of model is simply represented by block with planar faces. Note that numerical determination of calibration coefficients is useful only for one type of strain gauge rosette RY 61 S. Main goals of this thesis are motivation and request to clearly report effectiveness, accuracy and applicability of calibration coefficients in relation to thickness of specimen, dimensions of drilled hole, condition of “through” or “blind” hole and number of drilled increments. High quality and accuracy of created numerical model is necessary too. Numerical simulation of residual stresses by MKP needs to be done to obtain requested results. All results are presented by 3D, 2D graphs and tables and compared with analytical results or results from other authors. Although is this publication focused on numerical modeling using FEM, hole-drilling method has many significant restrictions. The most substantial of them is influence of eccentricity of drilled hole, creation of stress concentration near drilled area and subsequent plastification, influence of geometrical inaccuracy of hole, etc. All these aspects have significant influence of determining calibration coefficients and can not be included into numerical simulation. These problems are closely discussed in background research. All obtained results should be helpful for practical use of calculated calibration coefficients to determine uniform residual stresses of specimens with various thickness and drilled hole. All these results are also applicable only for one type of strain gauge rosette, which is RY 61 S.

[ES] El cambio climático y la conciencia social sobre el impacto de las infraestructuras en el medio está llevando a la necesidad de diseñar diques en talud con cotas de coronación reducidas frente a eventos de rebase más extremos. Además, la mayoría de estos diques se construyen en zonas de profundidades reducidas, donde el oleaje rompe a causa de la limitación por fondo. Estudios recientes apuntan a la necesidad de considerar no sólo la caudal medio de rebase (q) sino también el máximo volumen individual de rebase (Vmax), el espesor de lámina de agua (OLT) y la velocidad del flujo de rebase (OFV) en el diseño de la cota de coronación de un dique en talud según criterios de rebase. No obstante, existen pocos estudios en la literatura científica centrados en Vmax en estructuras costeras sometidas a oleaje limitado por fondo. Además, estos estudios proporcionan resultados contradictorios en relación a la influencia de la limitación por fondo del oleaje sobre Vmax. En cuanto a OLT y OFV, no se han encontrado estudios en la literatura científica que permitan su predicción en diques en talud. En esta tesis doctoral, se han realizado ensayos físicos 2D en diques en talud rebasables (0.3≤Rc/Hm0≤2.5) sin espaldón y con tres mantos principales (Cubípodo®-1L, cubo-2L y escollera-2L) sobre dos pendientes de fondo suaves (m=2% and 4%) en condiciones de oleaje limitado por fondo (0.2≤Hm0/h≤0.9). Vmax junto con q son las variables más recomendadas en la literatura científica para diseñar la cota de coronación de diques en talud según criterios de rebase. En el presente estudio, los mejores resultados en la estimación de Vmax*=Vmax/(gHm0T012) se han obtenido empleando la función de distribución Weibull de dos parámetros con un coeficiente de determinación R2=0.833. Durante la fase de diseño de un dique en talud, es necesario predecir q para calcular Vmax cuando se emplean los métodos dados en la literatura científica. Por tanto, se debe estimar q con fines de diseño si no se dispone de observaciones directas. En caso de emplear la red neuronal CLASH NN para estimar q (R2=0.636), la bondad de ajuste de la función de distribución Weibull de dos parámetros propuesta en esta tesis para predecir Vmax* es R2=0.617. Así, el ratio entre Vmax* medido y estimado cae dentro del rango de 1/2 a 2 (banda de confianza del 90%) cuando se emplea q estimado con CLASH NN. Los nuevos estimadores desarrollados en la presente disertación proporcionan resultados satisfactorios en la predicción de Vmax* con un método más simple que aquellos propuestos en la literatura científica. No se ha encontrado una influencia significativa de la pendiente de fondo ni de la limitación por fondo del oleaje sobre Vmax* en este estudio. OLT y OFV están directamente relacionados con la estabilidad hidráulica de la coronación del dique y la seguridad peatonal frente a rebase. Por tanto, se requiere estimar OLT y OFV en la coronación del dique para diseñar apropiadamente su cota de coronación empleando criterios de rebase. En este estudio, se han empleado redes neuronales para desarrollar nuevos estimadores explícitos que permiten predecir OLT y OFV superados por el 2% del oleaje incidente con un alto coeficiente de determinación (0.866≤R2≤0.867). El número de cifras significativas apropiado para los coeficientes experimentales de dichos estimadores se ha determinado en base a su variabilidad. El punto óptimo en el que las características del oleaje deben ser estimadas para predecir OLT y OFV se ha identificado a una distancia de 3h desde el pie de la estructura (siendo h la profundidad a pie de dique). La pendiente de fondo tiene influencia sobre OLT y OFV. Los valores más extremos de OLT y OFV se han descrito empleando las distribuciones Exponencial de un parámetro y Rayleigh, respectivamente, con resultados satisfactorios (0.803≤R2≤0.812).<br>[CA] El canvi climàtic i la consciència social sobre l'impacte de les infraestructures al medi està portant a la necessitat de dissenyar dics en talús amb cotes de coronació reduïdes front a esdeveniments d'ultrapassament més extrems. A més, la majoria dels dics es construeixen en zones amb profunditats reduïdes on l'onatge es trenca a causa de la limitació per fons. Estudis recents apunten a la necessitat de considerar no solament el cabal mitjà de sobrepasse (q) sinó també el màxim volum individual de sobrepasse (Vmax), l'espessor de la làmina d'aigua (OLT) i la velocitat del flux de sobrepasse (OFV) pel disseny de la cota de coronació d'un dic en talús segons criteris de sobrepasse. No obstant, existeixen pocs estudis a la literatura científica centrats en Vmax en estructures costeres sotmeses a onatge limitat per fons. Addicionalment, aquests estudis proporcionen resultats contradictoris en relació a la influència de la limitació per fons de l'onatge sobre Vmax. Quant a OLT i OFV, no s'han trobat estudis a la literatura científica que permeten la seua predicció a dics en talús. En aquesta tesi doctoral, s'han realitzat assajos físics 2D amb dics en talús amb sobrepassos rellevants (0.3≤Rc/Hm0≤2.5) sense espatlló i amb tres elements al mantell principal (Cubípode-1L, cubs-2L i esculleres-2L) ubicats sobre pendents de fons suaus (m=2% i 4%) en condicions d'onatge limitat pel fons (0.2≤Hm0/h≤0.9). Vmax conjuntament amb q són les variables més recomanades a la literatura científica per dissenyar la cota de coronació en dics en talús segons criteris d'ultrapassament. Al present estudi, els millors resultats en l'estimació de Vmax*=Vmax/(gHm0T012) s'han obtingut utilitzant la funció de distribució Weibull de dos paràmetres amb un elevat coeficient de determinació R2=0.833. Durant la fase de disseny d'un dic en talús, és necessari predir q per calcular Vmax quan s'utilitzen els mètodes donats a la literatura científica. Per tant, es deu estimar q amb fins de disseny si no es disposa d'observacions directes. Si s'aplica la xarxa neuronal de CLASH NN per estimar q (R2=0.636), la bondat d'ajust de la funció de distribució Weibull de dos paràmetres proposada a aquesta tesi per predir Vmax* és R2=0.617. Així doncs, el ràtio entre el Vmax* mesurat i estimat es troba dins del rang de 1/2 a 2 (banda de confiança del 90%) quan s'usa q predit amb CLASH NN. Els nous estimadors desenvolupats a aquesta dissertació proporcionen resultats satisfactoris en la predicció de Vmax* amb un mètode més senzill que aquells proposats a la literatura científica. No s'ha trobat una influència significativa de la pendent de fons ni de la limitació de l'onatge per fons sobre Vmax* a aquest estudi. OLT i OFV estan directament relacionats amb l'estabilitat hidràulica de la coronació de dics i la seguretat de vianants front a ultrapassaments. Per tant, es requereix estimar OLT i OFV en la coronació de dics per dissenyar apropiadament la seua cota de coronació utilitzant criteris de sobrepasse. En aquest estudi, s'han usat xarxes neuronals per desenvolupar nous estimadors explícits que permeten predir OLT i OFV superats pel 2% de l'onatge incident amb un elevat coeficient de determinació (0.866≤R2≤0.867). El nombre de xifres significatives apropiat per als coeficients experimentals dels mencionats estimadors s'ha determinat basant-se en la seua variabilitat. El punt òptim on determinar les característiques de l'onatge deuen ser estimades per predir OLT i OFV s'ha identificat a una distància de 3h des del peu de l'estructura (on h és la profunditat a peu de dic). La pendent de fons té influència sobre OLT i OFV. Els valors més extrems de OLT i OFV s'han descrit amb les distribucions Exponencial d'un paràmetre i Rayleigh, respectivament, amb resultats satisfactoris (0.803≤R2≤0.812).<br>[EN] Climate change and the social concern about the impact of infrastructures is leading to mound breakwaters with reduced crest freeboards facing higher extreme overtopping events. In addition, most mound breakwaters are built in the surf zone where depth-limited wave breaking takes place. Recent studies point out the need of considering not only the mean wave overtopping discharge (q) but also the maximum individual wave overtopping volume (Vmax), the overtopping layer thickness (OLT) and the overtopping flow velocity (OFV) when designing mound breakwater crest elevation using overtopping criteria. However, few studies in the literature are focused on Vmax on coastal structures under depth-limited breaking wave conditions. In addition, those few studies report contradictory conclusions regarding the significance of depth-limited breaking waves on Vmax. With respect to OLT and OFV, no studies are found in the literature for their prediction on mound breakwaters. In this PhD thesis, 2D physical model tests were conducted on overtopped mound breakwaters (0.3≤Rc/Hm0≤2.5) without a crown wall armored with three armor layers (Cubipod®-1L, cube-2L and rock-2L) on two gentle bottom slopes (m=2% and 4%) in depth-limited breaking wave conditions (0.2≤Hm0/h≤0.9). Vmax together with q are the most recommended variables in the literature to design mound breakwater crest elevation based on overtopping criteria. In the present study, the 2-parameter Weibull distribution provides the best results when estimating Vmax*=Vmax/(gHm0T012) with coefficient of determination R2=0.833. During the design phase of a mound breakwater, q is needed to predict Vmax using methods given in the literature. Thus, q must be estimated for design purposes when direct observations are not available. If CLASH NN is used to estimate q (R2=0.636), the goodness-of-fit of the 2-parameter Weibull distribution proposed in this thesis to predict Vmax* is R2=0.617. Hence, the ratio between the estimated and measured Vmax* falls within the range 1/2 to 2 (90% error band) when q is predicted using CLASH NN. The new estimators derived in this study provide satisfactory estimations of Vmax* with a method simpler than those found in the literature. Neither the bottom slope nor the depth-induced wave breaking seem to significantly influence the dimensionless Vmax* in this study. OLT and OFV are directly related to the hydraulic stability of the armored crest and the pedestrian safety. Thus, OLT and OFV are required to properly design crest elevation using overtopping criteria. Neural Networks (NNs) are used in this study to develop new explicit unbiased estimators for the OLT and OFV exceeded by 2% of the incoming waves with a high coefficient of determination (0.866≤R2≤0.867). The appropriate number of significant figures of the empirical coefficients of such estimators is selected according to their variability. The optimum point where wave characteristics are determined to predict OLT and OFV was identified at a distance of 3h from the toe of the structure (where h is the water depth at the toe of the structure). The bottom slope does influence both OLT and OFV. The most extreme values of OLT and OFV are described with the 1-parameter Exponential and Rayleigh distribution functions, respectively, with satisfactory results (0.803≤R2≤0.812).<br>Al Ministerio de Educación, Cultura y Deporte, por la financiación brindada con el programa de Formación de Profesorado Universitario (FPU16/05081). Al Ministerio de Economía y Competitividad, por la financiación de los proyectos ESBECO (EStabilidad hidráulica del manto, BErmas y COronación de diques en talud con rebase y rotura por fondo, BIA2015-70436-R) y HOLOBREAK (Estabilidad Hidráulica y Transmisión de Diques Rompeolas Homogéneos de Baja Cota Diseñados a Rotura por Fondo, RTI2018-101073-B-I00-AR).<br>Mares Nasarre, P. (2021). Overtopping flow on mound breakwaters under depth-limited breaking wave conditions [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/163154<br>TESIS

碩士<br>國立中興大學<br>應用數學系<br>85<br>The problem of rectangular plates with variable thickness along two directions is solved by using the method of equivalent systems. The original differential equation for plates with variable thickness is difficult to solve due to the non-constant flexural rigidity. By using the method of equivalent systems, we permit the replacement of the original variable flexural rigidity with uniform flexural rigidity to reduce the mathematical complexity of the variable thickness problem. The equivalent systems of uniform flexural rigidity are derived and proved mathematically. To facilitate the mathematical computations, an accurate approximation is named after the simplified equivalent systems by simplifying the equivalent systems which is developed for eliminating complexity of the high order series expansion. The variable rectangular plate can be analyzed in both equivalent systems and simplified equivalent systems for fear of failure, thus permitting observation of progressive deterioration of the plate's ability to resist load, stress, and deformation, and establishing useful, practical, critical limits in engineering application.

碩士<br>國立中興大學<br>奈米科學研究所<br>99<br>This work introduces the method of variable thickness bridges (VTB) for fabricating high-Tc Josephson junctions and DC SQUIDs. Focused Ion Beam (FIB) is used to control the thickness of the junction under a nanometer. The region that links the two electrodes exhibits the weak link effect. YBa2Cu3O7-X (YBCO) VTBs were fabricated with widths of 1 - 3 μm and a thickness of 150 nm using the standard techniques of photolithography and Ar+ ion milling. The thickness of the junctions was maintained between 40 and 90 nm. When the thickness of the junction was close to 80 nm, the voltage-current curve revealed RSJ-like behavior at 77 K. The critical current vs. temperature curve satisfies the general equation, Ic∝(1-T/Tc)α, with α=2 near Tc and 1 at low temperature. Shapiro steps were observed in a single junction at frequencies from 3.02 to 14.64 GHz. The widths of the current steps revealed oscillatory behavior with irradiation intensities. The characteristics of DC SQUID magnetometer were also measured. The peak-to-peak voltage swing is about 1.5 μV for a single SQUID at 70 K.

碩士<br>大同大學<br>機械工程學系(所)<br>95<br>Natural frequency of orthotropic annular plates with particular thickness variation is investigated based on Mindlin’s thick plate theory in this study. Also two outer restrained edges: simply -supported and clamped conditions are considered. In other words, the effects of shear deformation and rotary inertia are considered. In order to solve the governing eigenvalue problem, boundary characteristic orthonormal polynomials are applied in Rayleigh-Ritz method. Frequency parameters of the plates are evaluated for different values variables, such as ratio of radii, thickness ratio, thickness constant, taper parameter, orthotropic rigidity ratio, and vibration modes. The numerical convergence of the method is tested and a comparison of results for special cases of isotropic annular plates shows good accuracy. The effects of plate geometry and material properties on frequency parameter are demonstrated in forms of tables and figures. Therefore, in this study, the dynamic characteristics of plates can be predicted accurately by considering all effects of variables for different vibration modes.

碩士<br>國立中興大學<br>應用數學系<br>89<br>In this paper, analysis of the natural frequencies and mode shapes for free vibration of annular plates with variable thickness are presented. The governing differential equations for free vibration of annular plates with varying mass density and flexural rigidity are derived. They can be reduced to Bessel’s equations or Euler’s equation by selecting suitable expressions, such as power functions, for the distributions of flexural rigidity and mass along the radii of the plates. The general solutions of annular plates with variable thickness are obtained. Therefore, frequencies and mode shapes of free vibration of plates with variable thickness are computed and compared for different values of radii ratio and different boundary conditions. Aside from the frequencies and mode shapes for design purposes, these data can be used as test cases for assessing the accuracy of various approximate methods of solution.

碩士<br>中華科技大學<br>飛機系統工程研究所<br>99<br>Thermoelastic problems involving rotating disks have been popular in engineering applications, such as brakes, circular saws, gears, rotors etc. Theoretically, they are boundary value problems involving thermal and mechanical loadings. Therefore, the present thesis investigates the thermoelastic problems of functionally graded variable-thickness rotating discs subject to heat source by using the finite-element computer code ABAQUS with user subroutines. In addition, the corresponding analytical solutions are derived and provided by the computing software Mathematica. Different and interesting thermoelastic behavior is revealed by various parametric studies. Finally, the present research results can be extended further to other applications, e.g. thermal/mechanical analysis, laser machining, fracture mechanics, vibrations and acoustics etc.

碩士<br>大同大學<br>機械工程學系(所)<br>98<br>The purpose of this study was to analysis natural frequency of orthotropic annular plates with particular thickness variation subjected to noisothermal loads is investigated based on Mindlin’s thick plate theory. We consider different boundary conditions restrained with inner and outer edges: inner fix - outer free、inner free - outer fix and inner fix-outer fix. And the effects of shear deformation and rotary inertia are considered. In order to solve the governing eigenvalue problem and get the nondimension natural frequencies, the boundary characteristic orthonormal polynomials were applied in Rayleigh-Ritz method. The natural frequency of the plates subjected to the thermal of critical buckling loads are evaluated for different values variables, such as ratio of radii, thickness ratio, noisothermal constants, thickness constants, orthotropic rigidity ratio, and vibration modes etc. In this study, we present the comparison with references results for special cases of isotropic annular plates of uniform thickness, it shows good accuracy with the value of the nondimension thermal buckling loads, the results showed in tables and figures.

碩士<br>國立中興大學<br>土木工程學系所<br>98<br>The purpose of this thesis is to investigate the application of the stress wave techniques to concrete plate thickness measurement. Five measurement methods are used. The first one makes use of the frequency response combined with the far distance reflection wave method (Method one). The second one uses an impact device able to record the start time of impact and a receiver sensing the vertical surface displacement (Method two). The third method adopts an impact device able to record the start time of impact and a receiver sensing the horizontal surface displacement (Method three). The forth method uses two vertical receivers according to ASTM C 1383 (Method four). The final one employs two horizontal receivers and follows the ASTM C 1383 test procedure (Method five). A comparison of the in-place measurement results among these five methods is made and the difference between the measurement values and the known thicknesses is discussed. For the efficiency on measurement of concrete thickness, this thesis uses two sets of programs which are presented by graphic interface, called C1383-Modified, FF Plate Thickness Test Method for Method one and SEC_NDT_02.2_C1383 Plate Thickness Test Method for Method two 〜 five. The program C1383-Modified, FF Plate Thickness Test Method (Method one) adopts the travel time of the P-wave propagating from the impact point through the bottom reflection to the receiver at a point far away from impact and the thickness frequency of the plate and after measurement the velocity of internal P-wave and the plate thickness can be obtained simultaneously. The program SEC_NDT_02.2_C1383 Plate Thickness Test Method (Method two 〜 Method five) measures the surface P-wave velocity and the thickness frequency of the plate to estimate the thickness of concrete plates. Experimental results show that the accuracy of the thickness measurement depends on the accuracy of measurements on the thickness frequency and the wave velocity. Use of multiple measurement methods is of great help in making the measurement results reliable.

碩士<br>國立中興大學<br>機械工程學系<br>92<br>Abstract In this study, a computational fluid dynamics (CFD) package of Flotherm was used to investigate the thermal analysis of plate fins with variable thickness and constant height for an electronic module in laminar flow. The simulated parameters are two kinds of thermal conductivity (k=180W/m-K and 385 W/m-K) , two different aspect ratio of plate fins, and plate sharpness(a=0.1~1) so there are the analysis results of 40 sets shown in present analysis. At the same Reynolds number and aspect ratio, the calculated results at various plate sharpness parameters(a=0.1~1) are shown. In the present study, We can find the optimum heat dissipation of base fin plate. At the same Reynolds number, k=180W/m-K, at the aspect ratio H/L=1/3, the heat dissipation of plate fin at plate sharpness a=0.8 is best; At the same Reynolds number, k=385W/m-K, at the aspect ratio H/L=1/3, the heat dissipation of plate fin at plate sharpness a=0.7 is best.At the same Reynolds number and material, at the aspect ratio H/L=1/3, the heat dissipation of the plate fin with variable thickness and constant height is superior to the plate fins with uniform thickness. At the same Reynolds number and material, at the aspect ratio H/L=1/3, the heat dissipation of plate fin at plate sharpness a=1 is better than the others sharpness a=0.1~0.9.

碩士<br>國立中興大學<br>物理學系所<br>100<br>In this paper, we controlled the thickness about the link region produced by the microbridges of high-temperature superconducting YBCO (YB2C3O7-δ). In this thesis, Focus Ion Beam (FIB) was used to control different thickness of YBCO microbridges. And when the thickness of the microbridge was controlled under the nanometer scale, it showed weak link behavior. YBCO microbridges were fabricated with widths of 5-8 μm and thickness 150 nm using standard techniques of photolithography and Ar+ ion milling. To control the thickness of the micro-bridge was maintained between 40 and 90 nm. Using field-emission transmission electron microscope (FE-TEM) to observe the structure of the Variable Thickness Bridge (VTB), Ga+ damages horizontally and vertically into the amorphous structure about 20-30 nm. When the thickness was below 40 nm, there was little complete part of YBCO structure in the bottom. As the thickness of the micro-bridge decreased, critical temperature (Tc) and critical current (Ic) became smaller. The successful recovery of oxygen content during the annealing process, the superconductivity of the microbridge became better.

The behavior of variable thickness laminated composite beams has not so far been fully understood. In the present thesis, a finite element formulation is established for uniform and variable thickness composite beams (externally and mid-plane tapered composite beams). First the conventional formulation is used to establish the stiffness, geometric stiffness (for constant axial load, uniformly distributed axial load, and non-uniformly distributed axial load), and mass matrices. Second a new formulation (advanced formulation) is established, which considers not only the geometric boundary conditions, but also the natural boundary conditions. This means that at each node there will be four degrees of freedom, that are deflection, slope, bending moment, and shear force, such that all physical parameters that can be encountered in any practical situation can be included in the element formulation. The new stiffness, geometric stiffness, and mass matrices corresponding to the new formulation are set up. These matrices are provided into the MATLAB ® environment to obtain the natural frequencies and the critical buckling load.

碩士<br>大同大學<br>機械工程學系(所)<br>97<br>Natural frequency of orthotropic annular plates with particular thickness variation subjected to a torque and thermal loads is investigated based on Mindlin’s thick plate theory in this study. And three boundary conditions restrained with inner and outer edges: inner fix - outer free、inner free - outer fix、inner fix-outer fix conditions are considered. In other words, the effects of shear deformation and rotary inertia are considered. In order to solve the governing eigenvalue problem to get the nondimension natural frequencies, boundary characteristic orthonormal polynomials are applied in Rayleigh-Ritz method. The natural frequency of the plates subjected to a buckling torque and thermal of critical buckling loads are evaluated for different values variables, such as ratio of radii, thickness ratio, thickness constant, orthotropic rigidity ratio, and vibration modes etc. Therefore, in this study, the comparison with references results for special cases of isotropic annular plates of uniform thickness shows good accuracy with the value of the nondimension buckling torque and the value of the nondimension thermal buckling loads then demonstrated in forms of tables and figures.

碩士<br>國立成功大學<br>機械工程學系<br>103<br>In a general hydrostatic guideway rotary working table system, the rotary driving system is usually not a direct type, so there are some disadvantages like backlash, the gear matching, lubrication, slow response and other shortcomings. When external load applied on the table, the oil film thickness will be decreased, so the rotary speed will be affected. Moreover, the initial oil film thickness of each chamber are not the same due to manufacture precision of hydrostatic guideway, unequal oil from spilt-flow dividers and uneven pressure of each restrictor. This study will improve the lacks of traditional hydrostatic guideway rotary system. One will use relief valves to regulate the initial oil film thickness and let it be the same. For the rotary driving system, one will use a variable displacement hydraulic motor as a direct type driving system to improve the lacks of a general non-direct one. By regulating the displacement of a hydraulic motor with a self-tuning fuzzy controller, one can control the rotary speed of working table and resist external load. Moreover, when external load applying to the table, it will decrease oil film thickness. With speed control compensation, it will improve the change of oil film thickness. The experimental results show that the variable hydraulic motor can resist external load and let the working table rotate at constant speed. When an eccentric load applies on the table, the variable motor will compensate the load torque and decrease the change oil film thickness.

碩士<br>國立臺灣大學<br>機械工程學研究所<br>107<br>The purpose of this study is to establish a deep learning intelligent control system that can combine processing efficiency, stability and kerf consistency in variable thickness processing. In the past, the wire electrical discharge machining Control for variable thickness workpiece, and focus on improving processing efficiency and stability. The research on the problem that the kerf difference affects the accuracy of the variable thickness workpiece is relatively lacking. The true straightness is not good after processing, and it may lead to error fitting of future workpieces. Deep Learning is used to automatically extract features that are representative of data characteristics by transforming data through multiple nonlinear hidden layers, that is, from a variety of discharge signals that are highly correlated with the thickness of the workpiece. The recessive feature function has strong robustness and is not sensitive to noise or some of extreme values, so the thickness of the workpiece can be accurately known. The second stage is the setting of processing parameters. In the past, it relied on experienced operators. Therefore, the deep learning technique was used again to normalize the hidden features of the input data, so that the model output is like a experience operator to determination what the parameters should be setting. Intelligent control strategy not only reduce the operator&apos;&apos;s technical threshold, but also achieve a goal of uniform material removal rate during the trim cutting and reduce the occurrence of line marks. The experimental results show that the deep learning workpiece thickness online estimation system established in this study which can accurately estimate the thickness of the workpiece, and the processing parameter intelligent control system can adjust the discharge TOFF and TAOFF according to different workpiece thicknesses. The discharge frequency is controlled within the desired value, and the processing efficiency can be effectively improved when the thickness of the workpiece is thin to thick; when the thickness of the workpiece is thick to thin, the feed rate will be accelerated and the wire is broken by discharge concentration phenomenon. On the other hand, the strategy of this study effectively balances the kerf uniformity, that is, the average value and variation of the kerf of each class thickness can be effectively maintained within a stable value when processing different workpiece thicknesses.

碩士<br>中原大學<br>機械工程研究所<br>98<br>英文摘要 Weld line in the plastic products no mater or in the injection molding products, the are often seriously Defect and intensity appearance of adverse. 4C in the era of portable electronic products, emphasis on short, small, light, thin, quality, safety, product appearance and comfort requirements, has reached the ultimate attention. This can see a thin piece of plastic intensity, quality requirements and their associated impact factors will be more future more attention, so suture tensile intensity again become a major subject of injection molding. This study will address the most commonly used electronic products PC + ABS engineering plastic of the shell pieces when the object, take PC-540 into the future conduct of the injection molding. Use of different thickness, different process conditions and non-symmetrical mold temperature, by placing the sutures to be consistent with the results of standard ASTM D 638 of the pull-stretching experiments to study and compare information on their mechanical nature, quality and intensity in impact. The results showed that raise material temperature, mold temperature, injection molding parameters for the combination of speed lines have significantly reduced the depth, and with the combination of narrow-line depth has also advanced to enhance the anti-tensile strength of specimen. The use of induction heating heated to 130 ℃ in comparison with the traditional, in line with the thickness of 1.0mm depth of the decline in the more traditional 34% while the anti-tensile strength was raised by 3%, in line with the thickness of 1.5mm depth of the decline in the more traditional 80% of the anti-tensile strength was raised by 26%, in line with the thickness of 2.0mm of the decline in the depth of the more traditional 22% while the anti-tensile strength was raised by 1.78%, with the combination of the depth of the decline in line also makes anti-tensile strength can be increase. In the induction heating of the manufacturing process can also be evident that as the specimen thickness is thinner lines for the public mode of the combination of the depth of the impact rate as well. From this study, can not fully understand the depth and integrated line of anti-impact tensile strength of the quality of access and the establishment of the relevant reference data for future combination of lines can be further understanding of mechanical properties Key word ：dynamic mold temperature control、non-symmetry mold temperature、depth of weld line、tensile strength

碩士<br>國立清華大學<br>化學系<br>101<br>CHAPTER 1 One−Pot Synthesis of Silver Nanocubes and Their Morphological Transformation We have used one-pot reaction to synthesize cubic silver nanoparticles with average sizes of 62 to 80 nm in aqueous solution at 50 ºC for 6 hours. The reagents used here are AgNO3, cetyltrimethylammonium chloride (CTAC), ascorbic acid, and NaOH. In this method, silver nanoparticles are obtained by using ascorbic acid as reducing agent to reduce AgCl(s) in the presence of CTAC. NaOH is added to increase the reducing ability of ascorbic acid. Silver nanocubes with sizes varying from 62 to 80 nm are obtained by increasing the amount of AgNO3. Different amounts of NaOH also influence the morphology of silver nanoparticles. By increasing the amount of NaOH, the formation of {111} facets is enhanced and the shapes of silver nanoparticles change from cubes to truncated cubes and cuboctahedra due to the increased reaction rate. In addition, reaction rate is also increased when CTA+NO3─ serve as the surfactant. Furthermore, silver cuboctahedra are obtained and evolve into truncated octahedra by increasing the reaction temperature. CHAPTER 2 Synthesis and Plasmonic Properties of Au−Pd Core−Shell Heterostructures with Variable Shapes and Shell Thickness In this work, we report the investigation of plasmonic properties of Au−Pd core−shell heterostructures with different shapes, including cubes, cuboctahedra, truncated octahedra, and octahedra. Here, we have used a seed-mediated growth method to synthesize Au−Pd core−shell heterostructures with 35, 45, and 74 nm gold octahedra as cores. Au−Pd core−shell heterostructures with various shapes and sizes are prepared by mixing cetyltrimethylammonium bromide (CTAB), octahedral gold cores, H2PdCl4, and ascorbic acid at 50 ºC in less than 2 hours. The uniform shape of these nanocrystals and the ability to tune the shell thickness allow us to investigate their localized surface plasmon resonance (LSPR) properties. When the shell thickness become thin enough, blue shift of Au LSPR absorption band of Au−Pd core−shell heterostructures is observed. The Au LSPR absorption band red-shifts and become more obvious when the shape of core−shell nanostructures transforms from cubes to octahedra due to the shell thickness variation induced by change in the mole ratios of Pd : Au in a nanoparticle. Comparing to previous studies, this is the first time the plasmonic properties of bimetallic core−shell heterostructures with various shapes and tunable sizes is investigated. For Au−Pd core−shell octahedra, their Au LSPR absorption band is more pronounced and they may have the potential to be applied for plasmonic sensing such as hydrogen sensing.

Electrodialysis performance strongly depends on the boundary layer near ion exchange membranes. The thickness of the boundary layer has not been clearly evaluated due to its substantial fluctuation around the spacer geometry. In this study, the boundary layer thickness was defined with three statistical parameters: the mean, standard deviation, and correlation coefficient between the two boundary layers facing across the spacer. The relationship between the current and potential under conditions of the competitive transport between mono- and di-valent cations was used to estimate the statistical parameters. An uncertainty model was developed for the steady-state ionic transport in a two-dimensional cell pair. Faster ionic separations were achieved with smaller means, greater standard deviations, and more positive correlation coefficients. With the increasing flow velocity from 1.06 to 4.24 cm/s in the bench-scale electrodialyzer, the best fit values for the mean thickness reduced from 40 to less than 10 μm, and the standard deviation was in the same order of magnitude as the mean. For the partitioning of mono- and di-valent cations, a CMV membrane was examined in various KCl and CaCl₂ mixtures. The equivalent fraction correlation and separation factor responded sensitively to the composition of the mixture; however, the selectivity coefficient was consistent over the range of aqueous-phase ionic contents between 5 and 100 mN and the range of equivalent fractions of each cation between 0.2 and 0.8. It was shown that small analytic errors in measuring the concentration of the mono-valent cation are amplified when estimating the selectivity coefficient. To minimize the effects of such error propagation, a novel method employing the least square fitting was proposed to determine the selectivity coefficient. Each of thermodynamic factors, such as the aqueous- and membrane-phase activity coefficients, water activity, and standard state, was found to affect the magnitude of the selectivity coefficient. The overlimiting current, occurring beyond the electroneutral limit, has not been clearly explained because of the difficulty in solving the singularly perturbed Nernst-Planck-Poisson equations. The steady-state Nernst-Planck-Poisson equations were converted into the Painlevé equation of the second kind (P[subscript II] equation). The converted model domain is explicitly divided into the space charge and electroneutral regions. Given this property, two mathematical formulae were proposed for the limiting current and the width of the space charge region. The Airy solution of the P[subscript II] equation described the ionic transport in the space charge region. By using a hybrid numerical scheme including the fixed point iteration and Newton Raphson methods, the P[subscript II] equation was successfully solved for the ionic transport in the space charge and electroneutral regions as well as their transition zone. Above the limiting current, the sum of the ionic charge in the aqueous-phase electric double layer and in the space charge region remains stationary. Thus, growth of the space charge region involves shrinkage of the aqueous-phase electric double layer. Based on this observation, a repetitive mechanism of expansion and shrinkage of the aqueous-phase electric double layer was suggested to explain additional current above the limiting current.<br>text