Dissertations / Theses on the topic 'Integrated parametric design workflow'

The advancement of IT and technology has enabled the development of boundary breaking tools such as Parametric design and visual programming. Structural engineering has the potential to take the advantage of this development, by implementing visual programming which with the combination of optimization algorithms can explore design proposals. This opens up new possibilities to work closer with architects in the early stages of projects to create bolder architectural and structural designs. The task of the master thesis was to create a workflow using parametric design with visual programming and including an optimization algorithm. In the workflow, an optimization process should perform structural analysis and optimization operations to find suboptimal flat slab system designs. The idea was that the workflow should be implemented in the early stages of the structural design process, where an architectural model is used as a boundary to generate suboptimal flat slab systems based on user input. Thereafter, the different generated solutions need to be evaluated and verified by an engineer before proceeding further to the final design. The result obtained from the workflow was that an optimized flat slab system with column placements could be created through an optimization process with input data including geometry, loads and element properties. This led to an approach which exploited the capabilities of using parametric design and visual programming for structural design. This meant that, the user could alter the optimization process to narrow down the generated solutions to find the optimal flat slab system based on the requirements of the project. The results of the structural analysis in the workflow was not fully satisfactory, meaning it could not be used for final design without verification. The conclusion was that parametric design in combination with visual programming and optimization algorithms could generate multiple alternative designs. These alternatives could be used as inspiration for engineers to create new structural solutions in the early stages.
Framsteg inom IT och teknologi har möjliggjort utveckling av banbrytande verktyg som parametrisk design med visuell programmering. Konstruktörer har möjligheten att utnyttja denna utveckling genom att implementera visuell programmering, vilket i kombination med optimeringsalgoritmer kan generera alternativa konstruktionslösningar. Detta teknikskifte möjliggör ett närmare samarbete med arkitekter i tidiga skeden vilket kan resultera i mer vågade konstruktioner och arkitektur. Syftet med examensarbetet var att skapa ett arbetsflöde som utnyttjade parametrisk design och optimering i en visuell programmeringsmiljö som kunde utföra strukturanalys och optimering, vilket genererade optimala pelardäck med oväntade pelarplaceringar. Idén med detta var att arbetsflödet kunde implementeras i tidiga skeden med arkitekter, när den kan användas för att generera optimala pelardäck baserade på användarens indata. Därefter behöver de genererade lösningarna utvärderas och verifieras av en ingenjör, innan man fortsätter till nästa skede. Resultatet från arbetsflödet är att ett optimerat pelardäck med oväntade pelarplaceringar skapas genom en optimeringsprocess med indata innehållande geometri, laster, randvillkor och materialegenskaper. Detta arbetsflöde leder till ett angreppssätt som utnyttjar möjligheterna med parametrisk design och visuell programmering. Detta innebär att användaren kan påverka optimeringsprocessen för att smalna av resultatet för att hitta optimerade pelardäck baserade på projektets krav. Resultaten från strukturanalysen i arbetsflödet är inte helt tillförlitliga, vilket innebär att resultaten behöver verifieras. Sammanfattningsvis kan parametrisk design i kombination med visuell programmering och optimeringsalgoritmer skapa en mångfald av lösningar. Dessa alternativ kan inspirera ingenjörer att skapa nya konstruktionslösningar i tidiga skeden.

Thesis: S.M., Massachusetts Institute of Technology, School of Architecture and Planning, Program in Media Arts and Sciences, 2016.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 131-136).
Biodesign is a growing field that harnesses the re-engineering capabilities of synthetic biology and the principles of design to create functional products on a variety of scales. It is now possible to precisely modify and program living organisms to create products useful for medicine, fabrication, and more. These capabilities are today inspiring designers to consider, and design for opportunities associated with, the incorporation of biological and otherwise living matter into the built environment. Standard Computer Aided Design (CAD) software used in design and engineering often does not have resolution required for living systems, whereas many known bioCAD software do not allow for larger scales. In addition, simulations and animations are usually limited to a short timescale, and do not allow for predictive models over days or weeks. For creating environments in which living materials or systems, from bacterial biofilms to functional swarms, can intertwine with synthetically fabricated constructs, rapid prototyping software must be developed that can allow for both design and simulation in different conditions over time. This thesis is an attempt at creating a design methodology and finding a software platform for inclusion of living material systems in manufacturing of products on multiple scales. Existing biological CAD software is surveyed and applied to two case study projects engaging multiple scales (i.e. human scale, and architectural scale) for which the digital fabrication of living materials provides additional functionality and augments the biological or ecological environment. In the process, novel work is presented in the areas of apiary management and 3D printing with biology. Additionally, several computational approaches, including rule based and agent-based techniques, are applied to both projects and evaluated for accuracy and usability. This research took place in Mediated Matters newly constructed BL2 Wet Lab, and serves as a demonstration of research that lies at the intersection of additive manufacturing and synthetic biology.
by Sunanda Sharma.
S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Architecture, 2019
"June 2019." Cataloged from PDF version of thesis.
Includes bibliographical references (pages 106-108).
The Urban Heat Island (UHI) effect is a well-studied phenomenon broadly attributed to human activities that transform open terrain into cityscapes. Among global 21st-century concerns, projected trends in population growth, urbanization, and regional climate change could exacerbate the warming in cities and intensify the UHI effect. Yet, microclimate analysis essential to assessing UHI intensity is often neglected, resulting in poor planning practices with adverse effects on health, comfort and energy use within cities. With buildings responsible for substantial quantities of global energy consumption and carbon emissions, this context demands climate-responsive design to achieve better-performing cities. The UHI effect presents an urban design challenge, but only recently has there been a platform for design workflow integration.
Despite existing engines that accurately evaluate UHI intensity in urban environments, architects, designers, and urban planners have often not incorporated such simulation into microclimate studies due to prohibitively expensive computational costs, disconnected workflows within unintuitive or unfamiliar platforms, and uncertainty about difficult-to-obtain urban climatology parameters. These hindrances cause impactful delay within the design feedback loop and often generate a lack of confidence in the simulation process and output. This thesis proposes a Computer-Aided-Design integrated graphical user interface for the Urban Weather Generator (UWG), an urban-scale climate prediction tool developed by Bruno Bueno to simulate microclimatic conditions of urban sites using operational weather station data.
The goal is to make the powerful and computationally cheap engine accessible to design workflows by incorporating it as a plugin within the conventional design software Rhinoceros-3D, and by coupling it with the Local Climate Zone classification scheme developed by urban climate experts lain Stewart and Timothy Oke to standardize quantitative physical descriptions of cities. The proposed update automates geometric parameter extraction and implements a reliable means of urban morphological parameter estimation. As a case study, an iterative urban-scale design exploration is analyzed for selected climates.
by Hellen Rose Anyango Awino.
S.M.
S.M. Massachusetts Institute of Technology, Department of Architecture

The demand for complex structures and the urge to perform more detailed structural analyses in an early stage of the project design phase has increased the use of parametric design in the construction sector, especially among architects and structural engineers. Also, an increasing demand for sustainable structures is creating pressure on engineers and architects to design optimized structures that consume as little resources as possible. Keeping these demands in mind, this thesis tries to uncover the benefits of parametric design and optimization by applying these processes to industrial roof truss structures.The primary objective of the thesis is to investigate the feasibility and reliability of parametric design and optimization processes in real-life designs. For this purpose, a parametric algorithm has been developed in the visual programming software Grasshopper 3D. The workflow performs structural analysis and design verification on a parametric FE-model using the FEA software for parametric engineering, Karamba 3D in combination with Python where standards for design verification were scripted. These procedures were developed to be applied on both steel and timber truss structures. The workflow then performs a constrained cross-sectional and geometrical optimization of the truss structures. For the optimization process, the plug-in Galapagos have been used which uses evolutionary and simulated annealing techniques.After analyses of different cases and on comparison of the results from the model response verification, the resulting models showed that the workflow and analysis procedure was capable of obtaining a solution that is more effective and as reliable as the traditional structural analysis procedures and thus can be used for real case. When used during preliminary design, the parametric design procedure displayed great potential in saving time, thus saving resources and cost which paves a promising path for implementations in this sector.

The automotive engineering process is characterized by a long and complex design process which starts with the first sketches in the preliminary design phase and proceeds to the final detailed CAD and physical models. In this process, every design phase includes different process steps and tasks which are closely interconnected with each other. Therefore the different design stages demand capable Computer Aided Design (CAD) systems which are able to handle the different kinds of design information created and manipulated in the process. Currently in automotive practice, parametric and associative (PA) CAD systems are widely applied in the product development process. Such systems allow design knowledge to be embedded in CAD models by means of rules and formulae. In addition, CAD parts and assemblies can be generated faster and easier by modification of design parameters and therefore there is a possibility to create different CAD model variants which are based on the same CAD model. The four key element of the following work are (a) to identify the problems during the design process with parametric and associative (PA) methods during a three year of study and also the analysis of the literature survey. Furthermore (b) in this study the author will develop and implement a newly developed PA design approach (PARAMASS) in a ―real‖ industrial context. Beside this the following work will (c) discuss the issues which are important during the implementation of the developed PA approach in an industrial surrounding. The last key element (d) is to develop an evaluation approach for the PARAMASS approach during the application in an industrial context. In this case the author will be able to do action research in the industry and get first hand information during the accomplishment of these key elements. This thesis presents the results of a research programme carried out using the design research methodology of Blessing and Chakrabarti, aimed at understanding the difficulties and challenges faced by designers in using PA CAD systems and then developing and evaluating an integrated approach to the creation of PA CAD models in an automotive power train design context. Firstly, this thesis presents a review of the state of the art in PA design methods and approaches and also reviews previous research on the development of methodologies for the construction of PA CAD models. It then presents results of a descriptive study of the use of PA CAD tools and methods in vehicle power train design in an automotive original equipment manufacturer and in companies in its supply chain using questionnaires, interviews, tests and other field studies with a number of practising engineers. This study identified a number of issues faced by designers in the use of PA CAD tools and allowed the requirements for improved methods for the use of PA CAD tools to be formulated and indicators identified for their evaluation. Based on the results of the descriptive study a new integrated parametric associative (PA) approach for the design process of power train components was created in a prescriptive study stage. The approach, called PARAMASS, allows designers to construct and modify models in a methodical way based on three main phases: a specification phase to prepare the relevant parameters and associative relationships, a structuring phase that allows part and assembly structures to be created and a modification phase in which the created parametric and associative information can be modified and changed. The method makes extensive use of predefined structures matrix approaches adapted from the Design Structure Matrix. The prescriptive study phase of the research was followed by a second descriptive study to evaluate and investigate in both a qualitative and quantitative way the changes achieved by the PARAMASS approach. The qualitative evaluation was based on the Goal Question Metric approach and showed that there are advantages related to the reusability aspects like learning, application and acceptance of the developed integrated approach. The quantitative evaluation was based on the Use Case approach and demonstrated good advantages in applying the developed approach, but dependent on the complexity of the created parts and assemblies.

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2017.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 79-81).
Additive Manufacturing (AM), with its bottom-up approach enables the fabrication of highly complex geometries, the combination of multiple materials to achieve specific physical properties, and the consolidation of assemblies into single parts. Although many of the prominent processes were invented 20-30 years ago, we are now seeing rapid performance improvement and the synthesis of innovative processes that greatly expand the applications of AM. That being said, there are significant barriers to entry, one being the low throughput of additive processes when compared with more conventional methods, another being the high cost of introducing AM processes into current supply chains. Identifying key use cases where AM can be of maximum benefit is pivotal to exploit its potential. This research aims to provide methodology and tools to enable rapid estimation for part suitability in AM with physics-based and predictive models. A method is presented to estimate build time for parts manufactured via the Selective Laser Sintering process, utilizing specific metrics for part geometry and machine specific parameters. Key elements of the AM cost function are identified using the estimation model, paying close attention to assessing the impact of machine cost on the final part. In summary, these tools can be expanded to allow for not only part selection in AM, but also to act as decision tools when designing equipment..
by Talha Hasan.
S.M. in Engineering and Management

This thesis proposes a multifaceted production test and post-silicon yield enhancement framework for RF systems. The three main components of the proposed framework are the design, production test, and post-test phase of the overall integrated circuit (IC) development cycle. First, a circuit-sizing method is presented for incorporating test considerations into algorithms for automatic circuit synthesis/device resizing. The sizing problem is solved by using a cost metric that can be incorporated at minimal computational cost into existing optimization tools for manufacturing yield enhancement. Along with the circuit-sizing method introduced in the design phase, a low-cost test and diagnosis method is presented for multi-parametric faults in wireless systems. This test and diagnosis method allows accurate prediction of the end-to-end specifications as well as for the specifications of all the embedded modules. The procedure is based on application of optimized test stimulus and the use of a simple diode-based envelope detector to extract the transient test response envelope at RF signal nodes. This eliminates the need to make RF measurements using expensive standard testers. To further improve the parametric yield of RF circuits, a performance drift-aware adaptation scheme is proposed that automatically compensates for the loss of circuit performance in the presence of process variations. This work includes a diagnosis algorithm to identify faulty circuits within the system and a compensation process that adjusts tunable components to reduce the effects of performance variations. As a result, all the mentioned components contribute to producing a low-cost production test and to enhancing post-silicon parametric yield.

This PhD Thesis contributes to the domain of Educational Technologies, and more specifically to the Learning Design (LD) research field, which focuses on supporting teachers in the creation of effective computer-supported learning activities considering the needs of their educational contexts. Research in LD has provided a myriad of tools and methods. Yet, existing tools lack collaboration support for communities of teachers engaged in learning (co-)design. Moreover, scope of tools is varied in terms of representations used, pedagogical approaches supported, and design phases targeted (from conceptualization to authoring and implementation). This diversity of tools contrasts with lack of articulation of their synergies to offer meaningful, manageable and integrated LD ecosystems for teachers and communities of teachers. This Thesis is framed in this problem area. Its guiding research question is: How can community platform management mechanisms support teachers in integrated learning design ecosystems? This question is addressed by more specific investigation towards addressing four specific research objectives. The first objective is explorative, focused on understanding needs for management mechanisms in LD community platforms. The resulting contribution includes participation in building and evaluating LD community platforms (LdShake, Learning design Sharing and co-edition, and ILDE, Integrated Learning Design Environment) in the context of Spanish and European projects, and the identification of needs tackled in the following three research objectives. The second objective deals with enabling flexible management of learning (co-)design processes that involve use of several LD tools. The associated contribution is a model and implementation for LD Workflows, which shape orchestrated uses of selected LD tools that can be applied to LD Projects. The third objective focuses on supporting management of multiple learning design versions in scenarios of reuse and co-design. The contribution is a model and visualization strategy based on a family tree metaphor. The fourth objective concerns the need for interoperability between co-(design) tools and platforms, and in particular focuses on design patterns as structured LD representations of special interest because they collect repeatable good teaching practices. The contribution is a pattern ontology for computationally representing a pattern language (working case of design patterns in Computer-Supported Collaborative Learning) and a derived model together with an architecture for interoperable management of patterns across LD tooling. Contributions have been implemented in LdShake and ILDE community platforms, showing feasibility, enabling proofof-concept in significant scenarios and user studies involving teachers.
Las contribuciones de esta Tesis Doctoral se enmarcan en el ámbito de las Tecnologías Educativas, y más concretamente en el campo de investigación del Diseño de Aprendizaje (LD acrónimo en inglés). Este campo se centra en dar apoyo a los profesores en la creación de actividades educativas apoyadas por ordenador teniendo en consideración sus contextos educativos. La investigación en el campo de LD ha proporcionado gran cantidad de herramientas y métodos. Sin embargo, estas herramientas todavía carecen de mecanismos que posibiliten la colaboración en comunidades de profesores involucradas en el (co-)diseño de aprendizaje. Además, el alcance de las herramientas es muy variado en cuanto a las representaciones utilizadas, los enfoques pedagógicos utilizados, y fases de diseño a las que van dirigidas (desde la conceptualización, hasta la autoría y hasta la implementación). Esta diversidad de herramientas contrasta con la falta de articulación de sus sinergias para ofrecer ecosistemas LD significativos, manejables e integrados para profesores y comunidades de profesores. Esta problemática motiva la investigación realizada en esta Tesis. La pregunta de investigación que la guía es: ¿Cómo pueden apoyar los mecanismos de gestión de plataformas comunitarias dar soporte en ecosistemas de diseño de aprendizaje integrado? Esta cuestión se aborda en la investigación más concreta de cuatro objetivos específicos. El primer objetivo es exploratorio, se centra en la comprensión de las necesidades de mecanismos de gestión en plataformas para comunidades en LD. La contribución resultante incluye la participación en la implementación y evaluación de las plataformas para comunidades en LD (LdShake, acrónimo en inglés de Learning design Sharing and co-edition, e ILDE, acrónimo en inglés de Integrated Learning Design Environment) en el contexto de proyectos españoles y europeos, así como la identificación de las necesidades abordadas en los tres siguientes objetivos de la investigación. El segundo objetivo busca permitir una gestión flexible de los procesos de (co-)diseño de aprendizaje que implique el uso de varias herramientas de LD. La contribución asociada es un modelo e implementación de los flujos de trabajo de LD (LD Workflows en inglés). Los LD Workflows se definen para permitir la representación de las herramientas de LD seleccionadas que se pueden aplicar a proyectos de LD (LD Projects, en inglés). El tercer objetivo se centra en el apoyo a la gestión de múltiples versiones de diseño de aprendizaje en escenarios de reutilización y (co-)diseño. La contribución es un modelo y una visualización basada en una metáfora del árbol familiar (family tree, en inglés). El cuarto objetivo trata la necesidad de interoperabilidad entre herramientas de (co-)diseño y plataformas de LD, y en particular, se centra en los patrones de diseño como representaciones LD estructuradas de especial interés ya que recogen buenas prácticas docentes repetibles. La contribución es una ontología de patrones que representa computacionalmente un lenguaje de patrones (centrándose en los patrones de CSCL, del inglés:Computer-Supported Collaborative Learning) y un modelo derivado junto con una arquitectura para la gestión interoperable de patrones a través de herramientas de LD. Las contribuciones se han implementado en las plataformas de comunidades de LD LdShake e ILDE mostrando su viabilidad, ofreciendo la prueba de conceptos en escenarios significativos y estudios con profesores en entornos reales.

Dans cette thèse, nous abordons la conception des composants d'électronique de puissance, intégrés sur semi-conducteur. Dans cette large problématique, nous nous intéressons plus particulièrement aux méthodes et outils logiciels et numériques pour le dimensionnement technologique et géométrique. Ainsi, nous abordons le dimensionnement en faisant des compromis d'intégration entre la technologie du composant de puissance et les fonctions électriques de ses composants annexes, en prenant en compte la fiabilité de la réalisation technologique en salle blanche et les impacts de l'environnement électronique. Pour cela, nous avons proposé des démarches, méthodes et outils pour repousser les limites existantes de la conception, visant à offrir le support correspondant en terme de « design kit ». Finalement, nous appliquons les méthodes et les démarches choisies et développées, au dimensionnement d'un MOSFET de puissance (VDMOS), pour différents cahiers des charges
The thesis deals with the design of integrated power electronics components. In this large problematic, the thesis focuses on the methods and numerical and software tools for the geometrical and technological sizing. So, the thesis deals with the sizing carrying out compromises between the technology of a power component and the electrical characteristics of its auxiliary components, taking into account the reliability of the technological making and the impacts of the electronic environment. In this way, approaches, methods and tools are proposed to push away the existing design limits, to offer the corresponding elements for the design kit. Finally, the developed and chosen methods and approaches are applied to the sizing of a power MOSFET (VDMOS) according to several cases of specifications

Under de senaste 10-20 åren har användandet av BIM-modellering i byggbranschen ökat. En BIMmodellbyggs upp med hjälp av BIM-objekt. Dessa BIM-objekt kan skapas av projektören men dekan också skapas av leverantören av den produkt som BIM-objektet representerar. En samling avBIM-objekt med liknande egenskaper eller utseende kallas för en objektfamilj. En objektfamilj kant.ex. vara en dörr eller ett fönster. En fördel med objektfamiljer generellt är att de endast behövermodelleras en gång, men kan sedan användas i många olika projekt.Syftet med denna studie är att underlätta leverantörens arbete med utvecklingen av deras BIMfamiljer.Studien är utförd i Autodesks BIM-plattform Revit.Arbetet inleddes med en litteraturstudie som utgjorde grunden för arbetet. Att undersöka ochutforska befintliga objektfamiljer var även det en stor del av förarbetet. Detta gjorde att vi skaffadeoss goda kunskaper om objektfamiljer som vi sedan utvecklade genom att vi själva modelleradeobjektfamiljer. Modelleringen gjorde att vi fick en bättre förståelse om det vi tidigare endast lästom.Resultatet blev en generell arbetsgång som kan användas vid skapandet av många olika typer avBIM-objekt. Arbetsgången förutsätter att man förstår de olika delarna som en familj är uppbyggdav, vilket beskrivs i rapporten.En slutsats är att modellering av objektfamiljer kan göras på flera olika sätt och att det är viktigt attfortlöpande testa sin modell. Vidare bör inte objektfamiljerna innehålla mer information ännödvändigt, eftersom det kan göra familjerna onödigt komplicerade och mindre användarvänliga.Objektfamiljer med mindre mängd information föredras därför framför stora och detaljrikaobjektfamiljer. Därmed undviks också onödig belastning på den programvara som används.Studien genomfördes i samarbete med Preconal som har för avsikt att ta fram Revitfamiljer försina produkter. Vi använde oss av deras produkter som underlag till modelleringen
Within the last 10-20 years, the usage of BIM-modeling in the construction sector has increased.BIM-modeling is used daily in different stages of the design process in Sweden. During the designprocess BIM-objects is used which is created by either the manufacturers or the planner. Theseobjects are created by a real product and is made to represent that in the program. The increase ofBIM usage has led to more disciplines that didn’t work with BIM has to adapt and start with that.The purpose of this study is to increase the understanding about the creation of object families formanufacturers. Besides the literature study about the subject we also describe the whole modellingprocess in Revit step by step. This because to give an example how it could be made.The result became a general workflow that is usable for most kinds of objects. The workflowrequires the reader to understand the different parts of an object family which we describe in thethesis.A conclusion is that a family could be built in different ways and it’s important to regularly testyour model. Furthermore, it should not contain more information than what is necessary. BIMobjectswith less information is often preferred over heavy and detailed BIM-objects. This will alsoprevent unnecessary load on the software.

The present master thesis is an exploratory qualitative study in academic libraries regarding the transition from the integrated library systems to the next generation integrated library systems or library services platforms and the potential implications in their internal workflows. Nowadays, libraries all over the world are facing up with a number of challenges in terms of acquiring, describing and making available to the public all the resources, both printed and electronic, they manage. In particular, the academic libraries have more reasons to wish to fulfill their users’ needs since the majority of them use the library sources more and more for scientific research and educational purposes.In this study we attempt to explore the phenomenon in the globe using the available literature and to identify the implications in libraries’ workflows and the possible future developments. Moreover, through observation and semi-structured interviews we try to identify the current developments in the Greek context regarding the adoption of next ILS and possible implications in their workflows. Finally, we attempt a comparison between the Greek situation and the international one.

碩士
國立成功大學
建築學系碩博士班
101
The term - ”Parametricism” was proposed on 11th Venice Biennale in 2008, which then became a wide spreading trend in architecture domain that influence the thinking of modern design. Today, a growing number of designers widely adapt the Parametricism Design in architecture and industrial design. By using programing logic, designers can generate various Parametric Designs rapidly. However, two critical issues accompanied with the design, they are: 1) How to evaluate the design?, and 2) How to integrate with construction? In the past, design thinking normally seems as a black box. Designers' personal aesthetic or design experience are used to judge the final results, but may lead to incorrect evaluation that cause consuming works on fixing design. Moreover, the lack of integrating design may also create numerous problems on construction. In this research, we defined “Design Criteria” by setting up “Fitness Function” accordingly. Then, we fix and review the performance of Parametric Design by cooperating “Feedback Loops” with performance evaluation and simulation software, and calculate numerous optimized solutions via computers. Finally, we determine the final design via human brain. The whole process aims to transparentize the black box of design thinking, and involve both aesthetic and performance into design. To the perspective of integrating construction, BIM is an integrating platform for building information. The platform integrates information of each stages in building life cycles, which may reduce errors while transforming building information between each stages, and minimize human documentation errors. In addition, building evaluation can also adapted in early stage of design to achieve the goal of sustainable and fix the design errors in advance. In recent years, together with Parametric Design and BIM techniques is a coming trend on solving problems in digital design. In the design context of integrating BIM and Parametric Design, the major issues we are going to discuss in this research includes: 1) Develop optimized design results via Parametric Design process, and transparentize the black box of design thinking. 2)Use BIM as an integration platform for building information, which form as an integrated performance oriented building design process. We discuss the decision and feedback mechanics between human brain and computer. 3)Discuss the method of developing optimized architecture design and Parametric Design through numerous design criteria and physical environment performance evaluation principles. 4)Discuss the problems need to be solved on BIM design process. This research target the project - “Continuation of digital design building in College of Planning and Design, NCKU” as the design proposition, and process the Parametric Design to find the optimized design. The used software includes: Autodesk Vasari, Revit, NavisWorks, Rhinoceros, Grasshopper, Galapagos, Geco, Ecotect and Geometry Gym.

博士
國立交通大學
電機與控制工程系
88
In general, motion control and interpolator are the most important parts in multi-axis motion systems. Therefore, six advanced motion control algorithms (Chapter 2~5) and three parametric curve interpolators (Chapter 6~8) are proposed to achieve motion accuracy under high speed operations for multi-axis motion systems in this dissertation. To reduce contouring errors due to the mismatched dynamics among multiple axes, the perfectly matched feedback control design is developed to achieve identical frequency responses among all axes by applying the stable pole-zero cancellation and complementary zeros. To decrease perturbations caused by external disturbances, a digital disturbance observer (DDOB) control and a practical nonlinear function feedback compensation algorithms are also proposed. Moreover, by applying the design of the optimal digital pre-filter (DPF) and perfectly matched feedforward control, the perfectly matched feedforward control (PMFFC) is proposed to increases bandwidth of multi-axis systems. Furthermore, the robust multi-axis cross-coupled control (CCC) is developed to improve contouring accuracy by applying estimation and compensation of the newly developed contouring error vector. The contouring error transfer function (CETF) for multi-axis motion systems is also derived to simplify the CCC design. Accordingly, the multi-axis integrated control which combines (1) feedback control, (2) feedforward control, and (3) multi-axis cross-coupled control is proposed to significantly improve both tracking and contouring accuracy. Also, systematic control design procedures are obtained by applying the error signals analysis. Furthermore, machining quality may be deteriorated by the incorrect position command generation. The second part of this thesis discusses the design of precise interpolation algorithms for parametric curve of non-uniform rational B-spline (NURBS). In the second part of this dissertation, the implementation complexity of NURBS interpolator is solved by the newly developed method which uses the common data registers to realize an efficient interpolation algorithm for NURBS. Also, based on the common data registers, the 1st and 2nd derivatives of NURBS are efficiently obtained. Moreover, the speed-controlled interpolation algorithm generates precise position commands with a constant curve speed by applying the trajectory dependent compensatory value. By estimating the curve speed corresponding to the specified chord errors, the chord-error-controlled interpolation algorithm improves the position accuracy between two interpolating points. Feasibility of all proposed motion controllers and interpolators has been verified on an industrial DYNA CNC machining center. ABSTRACT (CHINESE) i ABSTRACT (ENGLISH) iii ACKNOWLEDGMENT v TABLE OF CONTENTS vi LIST OF TABLES x LIST OF FIGURES xiii CHAPTER 1 INTRODUCTION 1 1.1 General review 1 1.2 Problem statement 5 1.3 Proposed approach 7 1.4 Experimental setup 9 1.5 Organization of thesis 10 CHAPTER 2 FEEDBACK CONTROL DESIGN FOR MULTI-AXIS MOTION SYSTEMS 12 2.1 Perfectly matched feedback control design for multiple axes 13 2.2 Digital disturbance observer (DDOB) 23 2.3 Nonlinear compensation 34 2.4 Summary 48 CHAPTER 3 PERFECTLY MATCHED FEEDFORWARD CONTROL 49 3.1 Optimal ZPETC (Yeh and Hsu, 1999c) 49 3.2 Perfectly matched feedforward control 68 3.3 Summary 76 CHAPTER 4 ROBUST CROSS-COUPLED CONTROL 78 4.1 Robust CCC (Yeh and Hsu, 1999a) 78 4.2 Cross-coupled control by using the contouring error vector 94 4.3 Multi-axis cross-coupled control 102 4.4 Summary 110 CHAPTER 5 INTEGRATED CONTROL FOR MULTI-AXIS SYSTEMS 112 5.1 Biaxial integrated control (Yeh and Hsu, 1999b) 112 5.2 Multi-axis integrated control 127 5.3 Summary 137 CHAPTER 6 IMPLEMENTATION OF NURBS INTERPOLATORS 139 6.1 Review of NURBS 139 6.2 Structure of the basis function 141 6.3 The moving data registers 144 6.4 The branch structure and the extended basis function 145 6.5 The implementation of the proposed algorithm 147 6.6 Simulation results 149 6.7 Summary 151 CHAPTER 7 CONSTANT SPEED INTERPOLATION ALGORITHM 152 7.1 Parametric curve formulation 152 7.2 The speed-controlled interpolation algorithm 154 7.3 Applications 160 7.4 Summary 168 CHAPTER 8 CHORD-ERROR-CONTROLLED INTERPOLATION ALGORITHM 169 8.1 Parametric curve formulation and interpolations 169 8.2 Chord error controlled interpolation algorithm 170 8.3 Simulation results 173 8.4 Summary 177 CHAPTER 9 CONCLUSIONS 179 REFERENCES 181 VITA 191 PUBLICATION LIST 194

As we move deep into nanometer regime of CMOS VLSI (45nm node and below), the device noise margin gets sharply eroded because of continuous lowering of device threshold voltage together with ever increasing rate of signal transitions driven by the consistent demand for higher performance. Sharp erosion of device noise margin vastly increases the likelihood of intermittent failures (also known as parametric failures) during device operation as opposed to permanent failures caused by physical defects introduced during manufacturing process. The major sources of intermittent failures are capacitive crosstalk between neighbor interconnects, abnormal drop in power supply voltage (also known as droop), localized thermal gradient, and soft errors caused by impact of high energy particles on semiconductor surface. In nanometer technology, these intermittent failures largely outnumber the permanent failures caused by physical defects. Therefore, it is of paramount importance to come up with efficient test generation and test application methods to accurately detect and characterize these classes of failures. Soft error rate (SER) is an important design metric used in semiconductor industry and represented by number of such errors encountered per Billion hours of device operation, known as Failure-In-Time (FIT) rate. Soft errors are rare events. Traditional techniques for SER characterization involve testing multiple devices in parallel, or testing the device while keeping it in a high energy neutron bombardment chamber to artificially accelerate the occurrence of single events. Motivated by the fact that measurement of SER incurs high time and cost overhead, in this thesis, we propose a two step approach: ⟨i⟩ a new filtering technique based on amplitude of the noise pulse, which significantly reduces the set of soft error susceptible nodes to be considered for a given design; followed by ⟨ii⟩ an Integer Linear Program (ILP)-based pattern generation technique that accelerates the SER characterization process by 1-2 orders of magnitude compared to the current state-of-the-art. During test application, it is important to distinguish between an intermittent failure and a permanent failure. Motivated by the fact that most of the intermittent failures are temporally sparse in nature, we present a novel design-for-testability (DFT) architecture which facilitates application of the same test vector twice in a row. The underlying assumption here is that a soft fail will not manifest its effect in two consecutive test cycles whereas the error caused by a physical defect will produce an identically corrupt output signature in both test cycles. Therefore, comparing the output signature for two consecutive applications of the same test vector will accurately distinguish between a soft fail and a hard fail. We show application of this DFT technique in measuring soft error rate as well as other circuit marginality related parametric failures, such as thermal hot-spot induced delay failures. A major contribution of this thesis lies on investigating the effect of multiple sources of noise acting together in exacerbating the noise effect even further. The existing literature on signal integrity verification and test falls short of taking the combined noise effects into account. We particularly focus on capacitive crosstalk on long signal nets. A typical long net is capacitively coupled with multiple aggressors and also tend to have multiple fanout gates. Gate leakage current that originates in fanout receivers, flows backward and terminates in the driver causing a shift in driver output voltage. This effect becomes more prominent as gate oxide is scaled more aggressively. In this thesis, we first present a dynamic simulation-based study to establish the significance of the problem, followed by proposing an automatic test pattern generation (ATPG) solution which uses 0-1 Integer Linear Program (ILP) to maximize the cumulative voltage noise at a given victim net due to crosstalk and gate leakage loading in conjunction with propagating the fault effect to an observation point. Pattern pairs generated by this technique are useful for both manufacturing test application as well as signal integrity verification for nanometer designs. This research opens up a new direction for studying nanometer noise effects and motivates us to extend the study to other noise sources in tandem including voltage drop and temperature effects.

The thesis presents the design and analysis of a Twin Tensioner for a Belt-driven Integrated Starter-generator (B-ISG) system. The B-ISG is an emerging hybrid transmission closely resembling conventional serpentine belt drives. Models of the B-ISG system's geometric properties and dynamic and static states are derived and simulated. The objective is to reduce the magnitudes of static tension in the belt for the ISG-driving phase. A literature review of hybrid systems, serpentine belt drive modeling and automotive tensioners is included. A parametric study evaluates tensioner parameters with respect to their impact on static tensions. Design variables are selected from these for an optimization study. The optimization uses a genetic algorithm (GA) and a hybrid GA. Results of the optimization indicate the optimal system contains spans with static tensions that are significantly lower in magnitude than that of the original design. Implications of the research on future work are discussed in closing.

This thesis departs from the conviction that spaces that can change their formal configuration through movement may endow buildings of bigger versatility. Through kinetic architecture may be possible to generate adaptable buildings able to respond to different functional solicitations in terms of the used spaces. The research proposes the exploration of rigidly folding origami surfaces as the means to materialize reconfigurable spaces through motion. This specific kind of tessellated surfaces are the result of the transformation of a flat element, without any special structural skill, into a self-supporting element through folds in the material, which gives them the aptitude to undertake various configurations depending on the crease pattern design and welldefined rules for folding according to rigid kinematics. The research follows a methodology based on multidisciplinary, practical experiments supported on digital tools for formal exploration and simulation. The developed experiments allow to propose a workflow, from concept to fabrication, of kinetic structures made through rigidly folding regular origami surfaces. The workflow is a step-by-step process that allows to take a logical path which passes through the main involved areas, namely origami geometry and parameterization, materials and digital fabrication and mechanisms and control. The investigation demonstrates that rigidly folding origami surfaces can be used as dynamic structures to materialize reconfigurable spaces at different scales and also that the use of pantographic systems as a mechanism associated to specific parts of the origami surface permits the achievement of synchronized motion and possibility of locking the structure at specific stages of the folding.
A presente tese parte da convicção de que os espaços que são capazes de mudar a sua configuração formal através de movimento podem dotar os edifícios de maior versatilidade. Através da arquitectura cinética pode ser possível a geração de edifícios adaptáveis, capazes de responder a diferentes solicitações funcionais, em termos do espaço utilizado. Esta investigação propõe a exploração de superfícies de origami, dobráveis de forma rígida, como meio de materialização de espaços reconfiguráveis através de movimento. Este tipo de superfícies tesseladas são o resultado da transformação de um elemento plano, sem capacidade estrutural que, através de dobras no material, ganha propriedades de auto-suporte. Dependendo do padrão de dobragem e segundo regras de dobragem bem definidas de acordo com uma cinemática rígida, a superfície ganha a capacidade de assumir diferentes configurações. A investigação segue uma metodologia baseada em experiências práticas e multidisciplinares apoiada em ferramentas digitais para a exploração formal e simulação. Através das experiências desenvolvidas é proposto um processo de trabalho, desde a conceptualização à construção, de estruturas cinéticas baseadas em superfícies dobráveis de origami rígido de padrão regular. O processo de trabalho proposto corresponde a um procedimento passo-apasso que permite seguir um percurso lógico que atravessa as principais áreas envolvidas, nomeadamente geometria do origami e parametrização, materiais e fabricação digital e ainda mecanismos e controle. A dissertação demonstra que as superfícies de origami dobradas de forma rígida podem ser utilizadas como estruturas dinâmicas para materializar espaços reconfiguráveis a diferentes escalas. Demonstra ainda que a utilização de sistemas pantográficos como mecanismos associados a partes específicas da superfície permite atingir um movimento sincronizado e a possibilidade de bloquear o movimento em estados específicos da dobragem.