Les travaux de cette thèse portent sur la méthodologie de conception des Machines Synchrones à Aimants Permanents (MSAP) avec pour application le véhicule électrique. Dans une première partie, nous avons abordé du contexte de l'étude et de la problématique du dimensionnement. A l'occasion, nous avons montré qu'il est nécessaire d'adopter une méthodologie de design appropriée en fonction de l'avancement de chaque projet dans le cadre du processus de conception. Nous avons alors proposé différents niveaux de modélisation afin de repérer la machine optimale face au cahier des charges pour ensuite la caractériser plus finement et l'optimiser. La seconde partie du rapport traite donc de la modélisation et de l'optimisation de la MSAP. En premier lieu, un modèle analytique permet d'évaluer les performances globales des centaines de machines très rapidement. Ce premier calcul permet de sélectionner les meilleures machines à l'application pour ensuite les optimiser. Puis, le second niveau de modélisation se base sur les réseaux de réluctances. Ce niveau plus fin permet d'une part de retrouver les performances affinées des MSAP et également de procéder à l'optimisation. Enfin, la dernière partie du rapport est consacrée à l'optimisation de la MSAP en vue d'améliorer grandement les performances et de satisfaire au mieux au cahier des charges. / The works of this thesis concern the design methodology of Permanent Magnet Synchronous Machine (PMSM) for an electric vehicle application. The first part of this report introduces the context of the study and the problem of design. Then, we have shown that it is necessary to adopt an appropriate design methodology based on the progress of each project in the design process. Thus, we proposed different levels of modeling to identify the optimal machine against the specifications, and then characterize it more finely and optimize. The second part of the report deals with the modeling and optimization of the PMSM. First, an analytical model is employed to evaluate the overall performance of hundreds of machines very quickly. The first calculation is used to select the best machine for the application and then optimize them. Then, the second level of modeling is based on reluctance networks. This model allows finer resolution of the PMSM and also permits to perform optimization. The last part of the report is devoted to the optimization of the PMSM in order to greatly improve performances and meet the specifications.

Gastrointestinal (GI) septa are pathological entities whereby a wall of tissue is present in the GItract, resulting in symptoms such as dysphagia and regurgitation. They can be associated withconditions such as esophageal diverticula or upper gastrointestinal duplication, or post-surgicalcomplications such as candy cane syndrome. Current treatments involve interventions by eitherhighly skilled endoscopists or invasive surgery. We suggest the use of compression anastomosis toachieve endoscopic septotomy, relieving the patients of their symptoms. Compression anastomosisusing rings, clips, magnets, and wires or rubber bands (though not named as such for these last two)has previously been described in the literature. We propose the use of a combination of MagneticCompression Anastomosis (MCA) and what we have defined as Wire Compression Cutting (WCC),both involving progressive pressure application to induce ischemia, necrosis, inflammation, andfibrosis with regeneration. This PhD thesis describes the development of a novel medical devicedesigned to achieve the aforementioned process, from initial conception to detailed constructionby a specialized company, based on clinical and technical requirements defined in collaborationwith physicians and from latest regulations. This device, the MAgnetic Gastrointestinal UniversalSeptotome (MAGUS), consists of two magnetic boxes linked by a self-retractable wire. The wire isactivated by a spring coil system located inside the two magnetic boxes. This MAGUS MagneticDevice (MMD) is designed to be used with the MAGUS Delivery System (MDS), a catheter to whichit is attached, enabling the physician to mobilize each magnet independently during the procedure.To assess that the MAGUS meets the technical requirements of the procedure, and to mitigateagainst all possible risks that were identified through Failure Mode and Effects Analysis, verificationtests were performed in combination with animal testing. Clinical trials started in February 2020at Erasme Hospital with good preliminary results. Finally, to make the procedure more accessiblefor less experienced physicians, future improvements on the delivery catheter and procedure areproposed. / En gastroentérologie, un septum est défini comme une paroi de tissu séparant deux cavités. Ce typede structure peut être la cause de symptômes tels que de la dysphasie ou des vomissements. Il estcausé soit par des pathologies, comme les diverticules ou duplications de l’oesophage, soit par descomplications chirurgicales, comme le syndrome dit du "candy cane". Le traitement classique estla chirurgie mais celle-ci est liée à un haut taux de mortalité et morbidité. Ces dernières années,de nouvelles techniques endoscopiques ont été proposées, mais la plupart de celles-ci nécessitentl’intervention de médecins très expérimentés. Le travail développé dans cette thèse propose d’utiliserle concept d’anastomose par compression pour réaliser une découpe de ce septum. L’anastomosepar compression a été abondamment traitée dans la littérature. Cette compression est réalisée pardes anneaux, des clips, des aimants, des élastiques, ou du fil. Afin de réaliser cette septotomie,nous proposons donc de combiner l’anastomose par compression magnétique et ce que nous avonsappelé la découpe par compression par fil (Wire Compression Cutting), qui consiste à appliquer unepression, pour induire de l’ischémie, de la nécrose, de l’inflammation, et enfin de la régénérationcombinée à de la fibrose. Deux cahiers des charges, l’un clinique et l’autre technique, ont été réalisés;ceux-ci se basent sur la littérature sur le sujet, des échanges avec des médecins, et les normeset régulation actuellement d’application. Dans la suite du travail, le développement techniquedu MAGUS (MAgnetic Gastrointestinal Universal Septotome) est décrit, de l’idée à la premièreutilisation sur patients. Ce dispositif permet de réaliser cette compression, à l’aide de deux aimantsreliés par un fil rétractable. Cette solution épurée permet ainsi de découper différentes tailles deseptum, en une seule endoscopie, et sans incisions, réduisant le risque de perforation. Le "MAGUSMagnetic Device" a été conçu avec son outil dédié, le "MAGUS Delivery System", permettant defaciliter la mise en place du dispositif par endoscopie. Afin de vérifier que les caractéristiquestechniques définies au préalable sont bien remplies, et de réduire les dangers identifiés par uneanalyse de risque, des tests de vérifications ont été réalisés, en parallèle d’essais de faisabilité sur descochons. Les premières études cliniques ont commencé en février 2020 à l’Hôpital Erasme, donnantdes résultats préliminaires très encourageants. Ces premiers traitements ont notamment permisde prouver la faisabilité de la découpe par compression par fil. Ce travail est conclu en proposantdes améliorations pour le dispositif de pose, afin de faciliter sa prise en main, ainsi que de futuresnouvelles applications cliniques. / Doctorat en Sciences de l'ingénieur et technologie / info:eu-repo/semantics/nonPublished

Organisationer genomgår ständigt förändring för att utveckla nya produkter och för att förbli relevanta på marknaden. Ett tillvägagångssätt som organisationer kan tillämpa för att genomföra förändringar av produkter är metoder inom Design Thinking. En sådan metod kan vara Design Sprint, som ursprungligen skapades för att optimera samarbete i organisationer, för att lösa stora problem och testa nya idéer snabbt. Design Sprint är en relativt ny metod, som är framtagen för produktutveckling. Under de senaste åren har några forskningsstudier publicerats där studierna testar att använda Design Sprint. Det här skapade nyfikenhet kring vad som krävs av organisationer för att tillämpa Design Sprint. Metoden Design Sprint innehåller flera egenskaper som överlappar med många av de förutsättningar som krävs för att genomföra en förändring av en produkt i en organisation. Syftet med studien var att undersöka hur Design Sprint kan tillämpas i förändring av produkter i organisationer. Studien har gjort en djupdykning i ett specifikt fall som sedan har kompletterats med en sekundär datainsamling av ytterligare fallstudier. Studiens resultat visar på att det är möjligt för organisationer att tillämpa Design Sprint som metod i en förändring av produkter, men att det ställer krav på organisationen för att tillämpa metoden. / Organizations continuously undergo change to develop new products and to remain relevant on the market. One approach that organizations can apply when implementing change in products are methods in design thinking. Such a method could be Design Sprint, which originally was created to optimize how organizations cooperate and also to solve problems test new ideas quickly. Design Sprint is a relatively new method, created for product development. Over the past couple of years a few research studies have been published where these studies have tested Design Sprint. The method Design Sprint contains several properties that overlap with several of the requirements to successfully create change of a product in an organisation. The purpose of this study was to examine how Design Sprint might be applied to change products in organizations. The study has made a deep dive into a specific case which then have been supported by secondary data collection of additional case studies. The result of the study shows that it is possible for organizations to apply Design Sprint as a method in change of products, but that there are certain requirements of the organization to do it.

Hybrid Energy Resource System (HERS) is studied and applied aroundworld in recent years. Control and monitor of them are quite important in realapplication. HERS also has the equirement to integral with power grid such asdistribution grid networks. Therefore, to design and implement the informationcommunication system following IEC 61850, which is most promising standard fordesign of substation communication and automation system, is necessary. This paperpresents the design of Information Communication Technology (ICT) architectureand Unified Modeling Language (UML) models and final implementation through LabVIEW programming for Smart Energy Container. Applying design following IEC61850 series standards allow the HERS can communicate and interoperate with other IEC61850 devices and SCADA systems. The implementation is applied to SmartEnergy Container which contains wind power, solar power, battery energy storagesystem, and hydrogen energy storage system. Verification and testing results shows thedesign is qualified to control and monitor Smart Energy Container. / Smart Energy Container

This is an action-oriented study aimed at designing a practical methodology for generating evidence backed solutions for practical problems by means of literature review. Three iterations of systematic review are applied which evaluates different search strategies and reporting structures to provide a framework for an ideation technique. Resulting in an adaptation of a previously used framework which can be deployed to different extents that appears to result in design propositions alongside individual interventions.

The design and adjustment of convolutional neural network architectures is an opaque and mostly trial and error-driven process. The main reason for this is the lack of proper paradigms beyond general conventions for the development of neural networks architectures and lacking effective insights into the models that can be propagated back to design decision. In order for the task-specific design of deep learning solutions to become more efficient and goal-oriented, novel design strategies need to be developed that are founded on an understanding of convolutional neural network models. This work develops tools for the analysis of the inference process in trained neural network models. Based on these tools, characteristics of convolutional neural network models are identified that can be linked to inefficiencies in predictive and computational performance. Based on these insights, this work presents methods for effectively diagnosing these design faults before and during training with little computational overhead. These findings are empirically tested and demonstrated on architectures with sequential and multi-pathway structures, covering all the common types of convolutional neural network architectures used for classification. Furthermore, this work proposes simple optimization strategies that allow for goal-oriented and informed adjustment of the neural architecture, opening the potential for a less trial-and-error-driven design process.

Until now infrequent terms have been neglected in searching in order to save time and memory. With the help of a cascaded index and the introduced algorithms, such considerations are no longer necessary. A fast and efficient method was developed in order to find all terms in the largest freely available corpus of texts in the German language by exact search, part-word-search and fuzzy search. The process can be extended to include transliterated passages. In addition, documents that contain the term with a modified spelling, can also be found by a fuzzy search. Time and memory requirements are determined and fall considerably below the requests of common search engines.

This thesis addresses designing Product Lines (PLs) of Graphical Editors (GEs). It provides a feasible top-down design approach specialized on such Graphical Editor Product Lines (GEPLs), which can be configured dynamically. Furthermore, the end product's features are implemented modular, which has numerous positive effects on the development and maintenance processes for the family. These effects reach from decreasing the complexity of big PLs, allowing to delegate split up development tasks onto multiple isolated working teams, easier debugging and flexibility to extend or specialize a family of products as well as being able to use functionalities developed by third-party vendors. While design methods avoiding monolithic architectures and implementations exist for many PL domains, there are none known for GEPLs. Yet, the domain of those offers many challenges as GEPLs are actually comprised of Software Product Lines (SPLs) and Language Product Lines (LPLs), which is a combination untackled by any modular design approach known to me. Additionally, products in the domain require to implement multiple distinct and specific concerns, leading to artifacts which differ significantly but have to be located and managed in a single component. Overall, this justifies the need for specialized design approaches for the GEPL domain. In regard to this need, this thesis gives an overview of the existing landscape of approaches to design PLs, analyzing solutions offered by other researchers. Furthermore, a requirement analysis for the GEPL domain is conducted. Its results are the foundation for the presentation of a top-down design approach for dynamically configurable GEPLs, which are implemented feature modularly. Finally, a case study documenting the development of such a family of GEs is providing a proof of its feasibility.:1 Introduction 1.1 Motivation 1.2 Problem Definition 1.3 Outline 1.4 Terminology 2 Survey on Software and Language Product Line Design 2.1 Classification Scheme 2.1.1 Domain 2.1.2 Configuration 2.1.3 Design Method 2.1.4 Modularity 2.2 Overview 2.3 Discussion 2.3.1 Evaluation 2.3.2 Results 3 Requirements of Graphical Editor Product Lines 32 3.1 Functional Requirements 3.1.1 Edit Concerns 3.1.2 Language Family Concerns 3.2 Non-Functional Requirements 3.2.1 User Requirements 3.2.2 Development Requirements 4 Design of Graphical Editor Product Lines 4.1 Characteristics 4.2 Design Approach 4.2.1 Edit Concerns 4.2.2 Language Family Concerns 4.3 Discussion 4.3.1 Techniques 4.3.2 Evaluation 5 Case study: Modularization of a Family of Graphical Editors 5.1 Background 5.1.1 Compartment Role Object Model 5.1.2 Full-fledged Role Modeling Editor 5.1.3 Reusable Technology 5.2 Realization 5.2.1 Edit Concerns 5.2.2 Language Family Concerns 5.3 Discussion 5.3.1 Requirements 5.3.2 Limitations of the Modularization 5.3.3 Results 6 Conclusion 6.1 Summary 6.1.1 Desired Properties 6.1.2 Feasibility 6.2 Contributions 6.3 Future Work 6.3.1 Bottom-Up Design Method 6.3.2 Requirements 6.3.3 Modularization

How might we encourage playful yet provocative attitudes to explore unfamiliar methods? This thesis investigates how interaction designers can use unusual and thought-provoking methods in their design process. It focuses on understanding the challenges, limitations and opportunities of these types of approaches. The thesis explores the use of an abusive game probe as a way to spark conversations between participants and designers. The abusive game probe has the potential to generate meaningful and insightful outcomes. The proposed abusive game probe provided difficulties, being a unique and untested method. The topic of abuse presents challenges such as ethical consideration, which still requires further research. While determining its effectiveness can be complex, the main purpose is to encourage designers to embrace unexpected and unknown elements. This thesis provides guidance for future designers to incorporate these methods into their research, continue on this project or venture out and create new and alternate methods as contributions to the field of interaction design.

This Master Thesis consisted in realizing a tool able to design, estimate the cost and calculate annual yields of a molten salt solar power tower. Such tool was made for a company providing CSP equipment and plant solutions for engineering, engineering and procurement or also EPC of a solar power tower. The Company wishes to propose competitive plant configurations presenting a good trade-off between cost and revenues. The Company can oversee the EPC of a whole power plant or/and supply some components of the molten salt cycle and of the water/steam cycle. The tool models a large scale solar power tower with a thermal energy storage system on EBSILON®Professional 12.04, a thermodynamic software.   When launching a simulation, the tool sizes the components of the molten salt cycle (design phase) according to user’s inputs, the other components have their characteristics based on a reference project. Depending on the size of the components, the total cost is determined and the revenues over a year of operation are calculated (annual performance). When performing several simulations with different configurations, the Company can judge about the economic viability of plant configurations by comparing their LCOEs and NPVs.   The present document describes the result of the Master Thesis, that is to say the tool itself, what it contains and how it works. The methodology adopted to design the components is presented in depth, the way costs were calculated is exposed. The document explains the annual performance calculations and the simple operation strategy implemented. Finally, a technical and cost validation was carried out but it would require some further work to be complete. The design and cost calculations are performed in few seconds, the annual calculations take around 2-3h.   One main contribution of the Master Thesis is to show that designing, estimating costs and calculating annual performances is feasible in a single tool operating at a high level of detail. Using the tool during a solar power tower project could considerably facilitate the current process in place at the Company. It can also allow to compare an important number of configurations to determine a good techno-economic solution. / Denna uppsats bestod i att genomföra ett verktyg som kan utforma, beräkna kostnaden och beräkna årliga avkastningar på ett smält salt soltorn. Ett sådant verktyg gjordes för ett företag inom soltornsteknik, upphandling och konstruktion (SUK) som vill föreslå konkurrenskraftiga anläggningskonfigurationer som presenterar en bra avvägning mellan kostnad och intäkter. Företaget, samtidigt som det övervakar SUK för en hel kraftverk, levererar det också vissa komponenter i den smälta saltcykeln. Verktyget modellerar ett storskaligt soltorn med ett värmeenergilagringssystem på EBSILON®Professional 12.04, en termodynamisk programvara.   När en simulering startas, ritar verktyget komponenterna i den smälta saltcykeln (designfas) enligt användarens inmatningar, de andra komponenterna är baserade på ett referensprojekt. Beroende på komponenternas storlek bestäms den totala kostnaden och intäkterna över ett verksamhetsår beräknas (årlig prestation). När flera simuleringar görs med olika konfigurationer kan företaget bedöma sin ekonomiska lönsamhet genom att jämföra sina LCOE och NPV.   Det här dokumentet beskriver resultatet av masterprojektet, det vill säga själva verktyget, vad det innehåller och hur det fungerar. Den metod som antagits för att designa komponenterna presenteras grundligt samt hur kostnaderna beräknades. Dokumentet förklarar de årliga prestationsberäkningarna och den enkla operationsstrategin som implementerats. Slutligen genomfördes en teknisk och kostnadsvalidering, men det skulle kräva ytterligare insats för att göra arbetet fullständigt. Konstruktionen och kostnadsberäkningarna utförs på få sekunder, de årliga beräkningarna tar cirka 2-3 timmar.   Ett huvudbidrag av examensarbetet är att visa att utformning, uppskattning av kostnader och beräkning av årliga prestanda är möjlig i ett enda verktyg som arbetar på en detaljrik nivå. Att använda verktyget under ett soltornsprojekt kan betydligt underlätta den nuvarande processen på plats hos företaget. Det kan också göra det möjligt att jämföra ett viktigt antal konfigurationer för att bestämma en bra tekno-ekonomisk lösning.