Hexahedral refinement increases the density of an all-hexahedral mesh in a specified region, improving numerical accuracy. Previous research using solely sheet refinement theory made the implementation computationally expensive and unable to effectively handle multiply-connected transition elements and self-intersecting hexahedral sheets. The Selective Approach method is a new procedure that combines two diverse methodologies to create an efficient and robust algorithm able to handle the above stated problems. These two refinement methods are: 1) element by element refinement and 2) directional refinement. In element by element refinement, the three inherent directions of a hexahedron are refined in one step using one of seven templates. Because of its computational superiority over directional refinement, but its inability to handle multiply-connected transition elements, element by element refinement is used in all areas of the specified region except regions local to multiply-connected transition elements. The directional refinement scheme refines the three inherent directions of a hexahedron separately on a hexahedron by hexahedron basis. This differs from sheet refinement which refines hexahedra using hexahedral sheets. Directional refinement is able to correctly handle multiply-connected transition elements. A ranking system and propagation scheme allow directional refinement to work within the confines of the Selective Approach Algorithm.

Yes / A novel design strategy for microstrip harmonic-suppression antennas is presented. The computational method is based on an integral equation solver using adaptive surface meshing driven by a genetic algorithm. Two examples are illustrated, all involving design of coaxially-fed air-dielectric patch antennas implanted with shorting and folded walls. The characteristics of the antennas in terms of the impedance responses and far ¯eld radiation patterns are discussed theoretically and experimentally. The performances of all of the GA-optimised antennas were shown to be excellent and the presented examples show the capability of the proposed method in antenna design using GA. / MSCRC

Yes / The problem of parametric surface meshing for the purpose of design optimisation using finite element analysis is considered. Here the surface mesh is generated as a solution of a suitably posed boundary value problem implemented on a 2D parameter space. A robust meshing scheme is presented where an initial mesh is manipulated, with the aid of the 2D parameter space, so as to obtain a suitable surface triangulation. This meshing scheme can then be used to create suitable finite element meshes with which accurate design optimisations can be carried out.

A number of methods for adaptation of existing all-hexahedral grids by localized refinement have been developed; however, none ideally fit all refinement needs. This thesis presents the structure to a method of two-refinement developed for conformal, structured, all-hexahedral grids that offers flexibility beyond what has been offered to date. The method is fundamentally based on pillowing pairs of sheets of hexes. This thesis also suggests an implementation of the method, shows the results of examples refined using it and compares these results to results from implementing three-refinement on the same examples.

The design and analysis of large casted parts present significant challenges due to their complex geometry. Finite Element (FE) modeling is a vital tool for understanding the performance of casted components. However, the computational requirements associated with these parts often lead to excessive processing times and resource utilization. This thesis aims to enhance the efficiency of the mid-surface model creation by developing an FE modeling approach suited explicitly for large casted components. The study begins by exploring the background of casted parts and their applications. A comprehensive analysis of modeling and meshing techniques is conducted, emphasizing their application to large casted components. Building upon this knowledge, different ideas are examined, leading to the proposal of a methodology combining CAD strategies for design features, hybrid meshing techniques, and approaches aimed at reducing FE modeling time to streamline the overall process.To validate the proposed approach, a series of case studies involving casted parts with varying levels of complexity are undertaken. Real-world casting process parameters are considered, highlighting the advantages and limitations in each ideation phase. The proposed methodology is tested and show cased to expert engineers to evaluate its efficiency and feasibility. Furthermore, the efficiency of the new approach is quantitatively evaluated in terms of processing time. The developed methodology offers engineers and researchers a powerful tool to accelerate the design process and optimize FE modeling time while managing computational costs. As industries continue to push the boundaries of size and complexity in casted part design, the insights and techniques presented in this thesis offer a valuable resource for addressing the various engineering challenges inherent in future endeavors. / Utformningen och analysen av stora gjutna delar innebär betydande utmaningar på grund av deras komplexa geometri. Finita Element (FE)-modellering är ett viktigt verktyg för att förstå prestandan hos gjutna komponenter. De beräkningskrav som är förknippade med dessa delar leder dock ofta till alltför långa handläggningstider och resursutnyttjande. Detta examensarbete syftar till att förbättra effektiviteten av skapandet av mittyteta modeller genom att utveckla en FE-modelleringsmetod som är specifikt lämpad för stora gjutna komponenter. Studien börjar med att utforska bakgrunden till gjutna delar och deras tillämpningar. En omfattande analys av modellerings-och diskret iseringstekniker genomförs, med tonvikt på deras tillämpning på stora gjutna komponenter. Med utgångspunkt i denna kunskap undersöks olika idéer, vilket leder till förslaget om en metod som kombinerar CAD-strategier för designfunktioner, hybridmodelleringstekniker och tillvägagångssätt som syftar till att minska FE-modelleringstiden för att effektivisera den övergripande processen. För att validera det föreslagna tillvägagångssättet genomförs en serie fallstudier som involverar gjutna delar med varierande nivåer av komplexitet. Verkliga gjutprocessparametrar beaktas, vilket belyser fördelarna och begränsningarna i varje idéfas. Den föreslagna metoden testas och visas upp för expertingenjörer för att utvärdera dess effektivitet och genomförbarhet. Dessutom utvärderas effektiviteten av det nya tillvägagångssättet kvantitativt i termer av handläggningstid. Den utvecklade metoden erbjuder ingenjörer och forskare ett kraftfullt verktyg för att påskynda designprocessen och optimera FE-modelleringstiden samtidigt som de hanterar beräkningskostnader. När industrier fortsätter att tänja på gränserna för storlek och komplexitet idesign av gjutna delar, erbjuder de insikter och tekniker som presenteras i denna avhandling en värdefull resurs för att ta itu med de olika tekniska utmaningarna som är inneboende i framtida strävanden.

Yes / In the simulation of a chain of manufacturing processes, several finite element packages can be employed and for each process or package a different mesh density or element type may be the most suitable. Therefore, there is a need for transferring finite element analysis (FEA) data among packages and mapping it between meshes. This paper presents efficient algorithms for mapping FEA data between meshes with different densities and element types. An in-core spatial index is created on the mesh from which FEA data is transferred. The index is represented by a dynamic grid partitioning the underlying space from which nodes and elements are drawn into equal-sized cells. Buckets containing references to the nodes indexed are associated with the cells in a many-to-one correspondence. Such an index makes nearest neighbour searches of nodes and elements much faster than sequential scans. An experimental evaluation of the mapping techniques using the index is conducted. The algorithms have been implemented in the open source finite element data exchange system FEDES.

Personalisierung ist ein aktuelles Thema im Bereich der Online-Lehre. Lernende sind divers, sie haben unterschiedliche Vorkenntnisse, adressieren verschiedene Lernziele und variieren in ihren (Lern-)Präferenzen. Aufgrund dieser Diversität besteht die Notwendigkeit, Onlinekurse an die Bedürfnisse der Lernenden anzupassen. Dabei wird das Ziel verfolgt, Lernende so zu motivieren, dass sie ein bestmögliches Lernergebnis erzielen können. Im Rahmen der Dissertation wurde die Domäne des Sprachenlernens gewählt, auf welche Methoden zur Personalisierung angewandt wurden. Eine initiale Analyse kommerzieller Sprachlern-Apps hat gezeigt, dass Personalisierung bislang nur rudimentär als Sequenzierungsstrategie umgesetzt wurde. Dabei beschränkt sich die Anwendung hauptsächlich auf das Lernen mit Karteikarten. Die Dissertation beschäftigt sich mit der übergeordneten Forschungsfrage wie Personalisierung in automatisch generierten Online-Sprachlernkursen angewandt und mit fundamentalen und experimentellen Sequenzierungsstrategien kombiniert werden kann. Hierzu setzt die vorliegende Arbeit eine modifizierte Version der "Meshing-Hypothese" exemplarisch als Sequenzierungsstrategie um. Die Idee dieser Strategie besteht darin, dass Lernende besser lernen, wenn die Lehrmethoden mit den Präferenzen der Lernenden übereinstimmen. Diese Strategie wird in der Forschung seit Jahrzehnten kontrovers diskutiert. In mehreren Schritten wurden verschiedene Perspektiven zur Anwendung der Personalisierung mit dieser Strategie untersucht. Zunächst wurde ein Instrument zur Erfassung von Präferenzen entwickelt. Die exemplarische Auswahl beschränkt sich auf jene Präferenzen, welche durch konkrete Lehrmethoden abgebildet werden können. Anschließend wurde ein Experiment durchgeführt, um Leistungsunterschiede zwischen Lernenden mit unterschiedlichen Präferenzen zu ermitteln. Dazu wurden Lehrmethoden in einen konkreten Sprachlernkurs mit einer vordefinierten Lernprogression imitiert. Im nächsten Schritt wurde die technische Basis für adaptive Kurse vorbereitet. Zu diesem Zweck wurde die Lernplattform Moodle funktional so angepasst, dass sie für Experimente zur Personalisierung von Onlinekursen genutzt werden kann. Danach wurde die inhaltliche Basis vorbereitet, welche die Generierung von Sprachlerneinheiten unter Verwendung modernster generativer Modelle ermöglicht. Die dabei entstehenden interaktiven Lernmaterialien wurden um Imitationen zu einer Auswahl von Lehrmethoden ergänzt. Das so generierte Lernmaterial musste anschließend in eine optimale Reihenfolge gebracht werden. Ein neuer Ansatz zur Kombination verschiedener Sequenzierungsstrategien wird unter der Verwendung generativer neuronaler Netze eingeführt. Darauffolgend wird ein Konzept vorgestellt, das es ermöglicht, Sequenzierungsstrategien zu verstärken, wenn sie einen positiven Effekt auf das Lernergebnis haben oder sie zu vergessen, wenn der Effekt zu gering ist oder nicht existiert. Das Konzept wurde zunächst simuliert. Anschließend wurden die generierten Lernmaterialien, die technische Basis, die Sequenzierungsstrategien und die Präferenzen der Lernenden in einem finalen Experiment kombiniert, um die modifizierte Meshing-Hypothese als Fallbeispiel umzusetzen. Hierzu wurde das vorweg simulierte Konzept des Verstärkens oder Vergessens von Sequenzierungsstrategien in einem realen Online-Sprachlernkurs praktisch angewandt. Die im Rahmen der Dissertation durchgeführten Experimente erlauben Schlussfolgerungen für die Generierung und Personalisierung von Sprachlernkursen. Darüber hinaus geben mehrere Machbarkeitsstudien (Proof-of-Concepts) Einblicke in die praktische Anwendbarkeit der Ansätze. / Recently, personalization has become a topic of some interest in the field of education. Learners possess different levels of prior knowledge, along with various learning goals and preferences. This diversity underscores the necessity of adapting online courses to suit learners’ needs, with the ultimate goal of engaging learners for optimal learning outcomes. As a representative domain, the field of language learning was chosen, to which personalization has been applied. An initial analysis of commercial language learning apps revealed the scarce realization of personalization, mainly limited to flashcard learning as an item sequencing strategy. Subsequently, the dissertation focused on the overall research question: How can personalization in auto-generated online language learning courses be utilized and combined with fundamental and experimental item sequencing strategies? Therefore, the dissertation employed a modified version of the “Meshing Hypothesis” as an example, proposing that students learn more effectively when instructional teaching methods align with their preference levels. Its original strategy has been controversially discussed in research for decades. The research unfolded in several stages to utilize personalization, commencing with the development of an instrument to collect preferences linked to instructional teaching methods. An experiment was conducted to discern performance differences among learners with varying preference levels. This involved imitating scenarios for some instructional teaching methods within an online language learning course that follows a fixed learning progression. The subsequent phase concentrated on preparing the technical groundwork for personalized courses. To this end, the Moodle learning platform was functionally adapted to facilitate experiments in personalizing online courses. Simultaneously, a contextual foundation was established, enabling the generation of template-based language learning units through cutting-edge generative models. Applied imitations of a selection of instructional methods enhanced the resulting interactive learning materials. In order to optimize the arrangement of the generated learning materials, a novel approach was introduced, combining multiple item sequencing strategies using generative neural networks. A conceptual framework was introduced, allowing for the amplification of sequencing strategies that positively impact learning outcomes or their removal if the effect is negligible. First, the concept was simulated. Then, the technical foundation, generated learning materials, item sequencing strategies, and learner preferences were combined in a final experiment to employ the modified Meshing Hypothesis and to utilize the idea of “unlearning” item sequencing strategies in a real online language learning course. The experiments conducted in this dissertation allow conclusions for generating and personalizing online language learning courses. Furthermore, several proof-of-concepts provide insights into the applicability of the approaches to be drawn for practice.

This thesis presents experimental work and numerical analysis that has been undertaken to assess the bearing capacity of perforated offshore foundations. Perforated foundations may be used to support subsea infrastructure, including as mudmats into which a number of perforations have been made, or as grillages which consist of a series of structurally connected strip footings. Larger gravity base foundations, such as for offshore wind turbines or oil and gas platforms, may adopt a single central perforation. The advantages of using perforated foundations can include reduced material requirements and easier offshore handling as a result of smaller weight and lower hydrodynamic forces during deployment. Limited guidance currently exists for assessing the bearing capacity of these foundation types. Bearing capacity of perforated foundations has been examined in this thesis under conditions of combined vertical, horizontal and moment loading which is typical in offshore settings. Undrained soil conditions have been considered, except for the case of grillages in which drained conditions are often most relevant. Experimental work has included centrifuge testing of ring and square annular foundations on clay, and 1g testing of grillage foundations on sand. Finite element modelling has also been undertaken to assess perforated foundation capacity. A Tresca material subroutine (UMAT) and an adaptive meshing scheme have been developed to improve the accuracy of the finite element analysis carried out. The results showed that perforated foundations can be an efficient foundation solution for accommodating combined loading. As a ratio of their vertical load capacity, perforated foundations may be able to withstand higher moment and horizontal loads compared with unperforated foundations. The experimental and numerical results have been used to develop design expressions that could be employed by practitioners to estimate the vertical and combined load bearing capacity of these foundation types.

An understanding of the mechanical properties of viral capsids (protein assemblies forming shell containers) has become necessary as their perceived use as nano-materials for targeted drug delivery. In this thesis, a heterogeneous, spatially detailed model of the viral capsid is considered. This model takes into account the increased degrees of freedom between the capsomers (capsid sub-structures) and the interactions between them to better reflect their deformation properties. A spatially realistic finite element multi-domain decomposition of viral capsid shells is also generated from atomistic PDB (Protein Data Bank) information, and non-linear continuum elastic simulations are performed. These results are compared to homogeneous shell simulation re- sults to bring out the importance of non-homogenous material properties in determining the deformation of the capsid. Finally, multiscale methods in structural analysis are reviewed to study their potential application to the study of nanoindentation of viral capsids. / text

This dissertation delivers a new computational framework for the automatic generation of geometric feature patterns for industrial design and architectural facades on free-form surfaces. Such patterns include holes in a speaker grill, showerhead holes, protrusions on ceramics or bumpy textures on a panel. These patterns play a key role in making a designed object aesthetically pleasing as well as functional. Computer Aided Design (CAD) systems currently offer tools for generating simple patterns, such as uniformly spaced rectangular or radial patterns. However, they are not applicable to more general cases required in industrial design, including arbitrarily shaped target geometry and graded feature sizes. These tools are limited in several ways: (1) They cannot be applied to free-form geometries used in industrial design, (2) Patterning of these features happens within a single working plane and is not applicable to highly curved surfaces, and (3) Created features lack anisotropy and spatial variations, such as changes in the size and orientation of geometric features within a given region. This thesis proposes a new method of taking input for a target region along with sizing metrics. It will generate feature patterns automatically in three steps: (1) packing isotropic or anisotropic cells tightly in a target region, (2) scaling features according to the specified sizing metrics, and (3) adding features on the base geometry. This approach automatically generates complex patterns that conform to the boundary of any specified region. User input of a small number of geometric features (called “seed features”) of desired size and orientation in preferred locations also can be specified within the target domain. These geometric seed features are then transformed into tensors and used as boundary conditions to generate a Riemannian metric tensor field. A form of the Laplace heat equation is used to generate the field over the target domain, subject to specified boundary conditions. The field represents the anisotropic pattern of the geometric features. The system is implemented as a plugin module in a commercial CAD package to add geometric features to the target region of the model using two set operations, union and subtraction. This method facilitates the creation of a complex pattern of hundreds of geometric features in minutes. All the features are accessible from the CAD system and can be manipulated individually if required by the user. This allows the industrial designer or architect to explore more alternatives by avoiding the tedious and time-consuming manual generation of these geometric patterns.