No description available.

In order to understand biochemical networks we need to know not only how their parts work but also how they interact with each other. The goal of systems biology is to look at biological systems as a whole to understand how interactions of the parts can give rise to complex dynamics. In order to do this efficiently, new techniques have to be developed. This work shows how tools from mathematics are suitable to study problems in systems biology such as modeling, dynamics prediction, reverse engineering and many others. The advantage of using mathematical tools is that there is a large number of theory, algorithms and software available. This work focuses on how algebra can contribute to answer questions arising from systems biology. / Ph. D.

<p dir="ltr">Deep Neural Networks (DNNs) have been deployed on edge devices for numerous applications, ranging from computer vision, speech recognition, and anomaly detection. When deployed on edge devices, dedicated DNN compilers are used to compile DNNs into binaries to exploit instruction set architectures’ (ISAs’) features and hardware accelerators (e.g., NPU, GPU). These DNN binaries on edge devices process sensitive user information, conduct critical missions, and are considered confidential intellectual property.</p><p dir="ltr">From the security standpoint, the ability to reverse engineer such binaries (i.e., recovering the original, high-level representation of the implemented DNN) enables several applications, such as DNN models stealing, gray/white-box adversarial machine learning attacks and defenses, and backdoor detection. However, no existing reverse engineering technique can recover a high-level representation of a DNN model from its compiled binary code.</p><p dir="ltr">In this dissertation, we propose the following pioneering research for reverse engineering DNN on the edge device. (i) We design and implement the first compiler- and ISA-agnostic DNN decompiler, DnD, with the static analysis technique, capable of extracting DNN models from DNN binaries running on CPU-only devices without the hardware accelerator. We show that our decompiler can perfectly recover DNN models from different DNN binaries. Furthermore, it can extract DNN models used by real-world micro-controllers and enable white-box adversarial machine learning attacks against the DNN models. (ii) We design and implement a novel data-driven approach, NeuroScope, based on dynamic analysis and machine learning to reverse engineer DNN binaries. This compiler-independent and code-feature-free approach supports a larger variety of DNN binaries across different DNN compilers and hardware platforms. We demonstrate its capability by using it to reverse engineer DNN binaries unsupported by previous approaches with high accuracy. Moreover, we showcase how NeuroScope can be used to reverse engineer a proprietary DNN binary compiled with a closed-source compiler and enable gray-box adversarial machine learning attacks.</p>

Topologieoptimierte Geometrie wird vom Konstrukteur in einem Format benötigt, das im CAD verwendet werden kann. Die Herausforderung ist, dass man als Optimierungsergebnis eine STL-Geometrie bekommt, doch fürs CAD soll die Oberfläche möglichst glatt und vereinfacht sein. Die Software MSC Apex erzeugt aus einer STL-Geometrie eine echte, geglättete Parasolid-Geometrie. Eine spezielle Form der Topologieoptimierung ist die Topometrieoptimierung, welche die Rippenstruktur auf einem Blech vorschlägt. Auch hierfür ist Geometrierückführung möglich. Auch stark verformte Bauteile, z.B. Gummiteile, kann man als Geometrie im verformten Zustand ans CAD zurückgeben, um zu sehen, wie sie dann in die Baugruppe passen. / Topology optimized geometry must be provided to the designer in a format that can be used in CAD. The challenge is that the optimization result is typically an STL geometry, but for CAD the surface should be as smooth and simplified as possible. The MSC Apex software creates a real, smoothed parasolid geometry from an STL geometry. A special form of topology optimization is topometry optimization, which suggests the rib structure on a sheet. Reverse engineering is also possible for this. Even heavily deformed components, e.g. rubber parts, can be returned to CAD as geometry in the deformed state, to see how they fit into the assembly in this state.

Thesis (MScEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The main components of a micro gas turbine engine are a centrifugal or mixed-flow compressor, a combustion chamber and a single stage axial-flow or radial-flow turbine. The goal of this thesis is to formulate a design methodology for small axial-flow turbines. This goal is pursued by developing five design-related capabilities and applying them to develop a turbine for an existing micro gas turbine engine. Firstly, a reverse engineering procedure for producing digital three-dimensional models of existing turbines is developed. Secondly, a procedure for generating candidate turbine designs from performance requirement information is presented. The third capability is to use independent analysis procedures to analyse the performance of a turbine design. The fourth capability is to perform structural analysis to investigate the behavior of a turbine design under static and dynamic loading. Lastly, a manufacturing process for prototypes of a feasible turbine design is developed. The reverse engineering procedure employs point cloud data from a coordinate measuring machine and a CT-scanner to generate a three-dimensional model of the turbine in an existing micro gas turbine engine. The design generation capability is used to design three new turbines to match the performance of the turbine in the existing micro gas turbine engine. Independent empirical and numerical turbine performance analysis procedures are developed. They are applied to the four turbine designs and, for the new turbine designs, the predicted efficiency values differ by less than 5% between the two procedures. A finite element analysis is used to show that the stresses in the roots of the turbine rotor blades are sufficiently low and that the dominant excitation frequencies do not approach any of the blade natural frequencies. Finally prototypes of the three new turbine designs are manufactured through an investment casting process. Patterns made of an organic wax-like material and a polystyrene material are used, with the former yielding superior results. / AFRIKAANSE OPSOMMING: Mikro-gasturbiene-enjins bestaan uit 'n sentrifugaal- of ‘n gemende-vloeikompressor, 'n verbrandingsruim en 'n enkel-stadium-aksiaalvloei- of ‘n radiaalvloei-turbine. Die doel van hierdie tesis is om 'n ontwerpsmetodologie vir klein aksiaalvloei-turbines saam te stel. Hierdie doel word deur die ontwikkeling en toepassing van vyf ontwerpsverwante vermoëns nagestreef. Eerstens word 'n tru-waartse-ingenieursproses ontwikkel om drie-dimensionele rekenaarmodelle van die bestaande turbines te skep. Tweedens word 'n metode om kandidaatturbineontwerpe vanaf werkverrigtingsvereistes te verkry, voorgestel. Die derde ontwerpsvermoë is om die werksverrigting van 'n turbineontwerp met onafhanklike analises te evalueer. Die vierde ontwerpsvermoë is om die struktuur van 'n turbinelem te analiseer sodat die effek van statiese en dinamiese belastings ondersoek kan word. Laastens word 'n vervaardigingsproses vir prototipes van geskikte turbineontwerpe ontwikkel. Die tru-waartse-ingenieursproses maak gebruik van 'n koördinaat-meet-masjien en 'n CT-skandeerder om puntewolkdata vanaf die turbine in 'n bestaande mikro-gasturbiene-enjin te verkry. Die data word dan gebruik om 'n drie-dimensionele model van die turbine te skep. Die ontwerpskeppingsvermoë word dan gebruik om drie kandidaatturbineontwerpe vir die bestaande mikro-gasturbiene-enjin te skep. Onafhanklike empiriese en numeriese prosedures om die werkverrigting van 'n turbineontwerp te analiseer word ontwikkel. Beide prosedures word op die vier turbineontwerpe toegepas. Daar word gevind dat die voorspelde benuttingsgraadwaardes van die nuwe ontwerpe met minder as 5% verskil vir die twee prosedures. 'n Eindige-element-analise word dan gebruik om te wys dat die spannings in die wortels van die turbinelemme laag genoeg is, asook dat die dominante opwekkingsfrekwensies nie die lem se natuurlike frekwensies nader nie. Laastens word prototipes van die drie nuwe turbineontwerpe deur 'n beleggingsgietproses vervaardig. In die vervaardigingproses word die effektiwiteit van twee materiale vir die gietpatrone getoets, naamlik 'n organiese wasagtige materiaal en 'n polistireen-materiaal. Daar word bevind dat die gebruik van die wasagtige gietpatrone tot beter resultate lei.

<p>Cilj istraživanja realizovanih u ovom radu, bio je je da se formalno opišu mogući pristupi transformacijama različitih opisa baza podataka iz jednog modela podataka u drugi i praktično provere kroz njihovu implementaciju u okviru jednog CASE alata, namenjenog projektovanju informacionih sistema i baza podataka. U skladu sa postavljenim ciljem definisan je metodološki pristup i softversko okruženje IIS*Ree koje omogućava visok nivo automatizacije procesa reinženjeringa informacionih sistema. Okruženje IIS*Ree, zasnovano na MDSD principima, kao krajnji rezultat generiše šemu baze podataka u izabranom ciljnom, konceptualnom ili implementacionom modelu podataka, kao i prototip aplikacije.</p> / <p>The goal of the research presented in this paper is to formally specify approaches to transformation of database specifications between different data models. The transformations are then to be implemented and tested using a CASE tool for modelling information systems and databases. Following this goal, a methodological approach is defined together with a software named IIS*Ree that provides a high level of automation of the information system reengineering process. The IIS*Ree software, developed in accordance to MDSD principles, generates database schemas specified in a target, conceptual or implementation data model, as well as application prototypes.</p>

<p>Doktorska disertacija razmatra primenu savremenih aditivnih proizvodnih tehnologija u postupku preciznog livenja ortopedskih implantata i njihov uticaj na skraćenje vremena i smanjenje tro&scaron;kova izrade topljivih modela. Konvencionalni postupak preciznog livenja ortopedskih implantata zahteva značajno vreme i tro&scaron;kove za izradu kalupa za topljive modele. U disertaciji je razvijena metoda za &bdquo;brzo precizno livenje&ldquo; kojom se elimini&scaron;e potreba za izradom kalupa za topljive modele ortopedskih implantata. Potrebno je utvrditi da li se pomoću predložene metode &bdquo;brzog preciznog livenja&ldquo; koja predstavlja integraciju aditivnih proizvodnih tehnologija i reverznog inženjerstva sa konvencionalnim preciznim livenjem, može dati značajniji doprinos daljem razvoju u oblasti izrade prilagođenih ortopedskih implantata.</p> / <p>The doctoral thesis discusses the application of modern additive manufacturing technologies in investment casting of orthopaedic implants and their impact on time and cost savings in meltable wax models development. The conventional procedure of investment casting of orthopaedic implants demands considerable time and costs when developing moulds for meltable wax models. The thesis shows a method of &ldquo;rapid investment casting&rdquo; developed to avoid the making the moulds for meltable wax models of orthopaedic implants. It is necessary to establish whether the proposed method of &ldquo;rapid investment casting&rdquo;, which integrates additive manufacturing technologies and reverse engineering with conventional investment casting, may give a significant contribution to further development of manufacturing of customized orthopaedic implants</p>

Nous proposons une approche semi-automatique pour la rétro-ingénierie des diagrammes de séquence d’UML. Notre approche commence par un ensemble de traces d'exécution qui sont automatiquement alignées pour déterminer le comportement commun du système. Les diagrammes de séquence sont ensuite extraits avec l’aide d’une visualisation interactive, qui permet la navigation dans les traces d'exécution et la production des opérations d'extraction. Nous fournissons une illustration concrète de notre approche avec une étude de cas, et nous montrons en particulier que nos diagrammes de séquence générés sont plus significatifs et plus compacts que ceux qui sont obtenus par les méthodes automatisées. / We propose a semi-automated approach for the reverse engineering of UML sequence diagrams. Our approach starts with a set of execution traces that are automatically aligned to determine the common behavior. Sequence diagrams are then extracted with the help of an interactive visualization, which allows navigating though execution traces and performing extraction operations. We provide a concrete illustration of our approach with a case study, and show in particular that the resulting diagrams are more meaningful and more compact than those extracted by automated approaches.

Nous proposons une approche qui génère des scénarios de visualisation à partir des descriptions de tâches d'analyse de code. La dérivation de scénario est considérée comme un processus d'optimisation. Dans ce contexte, nous évaluons différentes possibilités d'utilisation d'un outil de visualisation donnée pour effectuer la tâche d'analyse, et sélectionnons le scénario qui nécessite le moins d'effort d'analyste. Notre approche a été appliquée avec succès à diverses tâches d'analyse telles que la détection des défauts de conception. / We propose an approach that derives interactive visualization scenarios from descriptions of code analysis tasks. The scenario derivation is treated as an optimization process. In this context, we evaluate different possibilities of using a given visualization tool to perform the analysis task, and select the scenario that requires the least effort from the analyst. Our approach was applied successfully to various analysis tasks such as design defect detection and feature location.

Continuous improvements are an integral part for the development of everyday life. These improvements do not only ascertain financial gain but also lessening the environmental impact. The purpose of this study is to gain a deeper understanding of the design process and the decisions required to achieve an optimal design with respect to weight reduction, while retaining the required strength. The study will also investigate the choice of material. This will be done through material and design studies, along with strength calculations of the product in question. The conclusion from this study is that it is possible to save a certain percentage of the material used, while keeping the strength, form, fit and function intact. The material study provided with a recommendation of the zinc alloy ZA-8. It is an alloy with a good combination of great strength, low density and price.