Master of Science / Department of Statistics / Shie-Shien Yang / There have been considerable debates over the robust parameter design. As a result, there have been many approaches presented that are suited to the robust parameter design. In my report, I illustrate and present Taguchi's robust parameter design, response surface approach and semi-parameter design. Considerable attention has been placed on the semi-parameter design. This approach is new technology that was introduced to Picke, Robinson, Birch and Anderson-Cook (2006). The method is a combined parametric and nonparametric technique to improve the estimates of both the mean and the variance of the response.

This thesis aims to tune the control parameters of a bio-inspired guidance system designed to confer a tactical behavior to unmanned aerial vehicles (UAVs). This bio-inspired guidance system is capable of reducing exposure to threats, while traversing previously uncharted, and potentially hostile territories. UAVs employing this guidance system may exhibit a more or less tactical behavior by tuning 9 user-defined parameters within specified intervals. Although the UAV's behavior can be easily forecasted whenever all parameters are set to exhibit the most cautious behavior or the most reckless behavior, it is difficult to devise a taxonomy of flight behavior whenever these parameters are not set at the boundaries of their admissible intervals. The scope of this thesis is to analyze and forecast the UAV's behavior as a function of these user-defined parameters. To this goal, the Taguchi analysis method is employed to deduce those parameters that affect the UAV's behavior more than others. Successively, 81 software-in-the-loop simulations have been performed to analyze the UAV's behavior as a function of the most influential user-defined parameters. Finally, 10 flight tests were performed to validate the numerical results. / Master of Science / This thesis aims to tune the control parameters of a bio-inspired guidance system designed to confer a tactical behavior to unmanned aerial vehicles (UAVs). This bio-inspired guidance system is capable of reducing exposure to threats, while traversing previously uncharted, and potentially hostile territories. UAVs employing this guidance system may exhibit a more or less tactical behavior by tuning 9 user-defined parameters within specified intervals. Although the UAV's behavior can be easily forecasted whenever all parameters are set to exhibit the most cautious behavior or the most reckless behavior, it is difficult to devise a taxonomy of flight behavior whenever these parameters are not set at the boundaries of their admissible intervals. The scope of this thesis is to analyze and forecast the UAV's behavior as a function of these user-defined parameters. To this goal, the Taguchi analysis method is employed to deduce those parameters that affect the UAV's behavior more than others. Successively, 81 software-in-the-loop simulations have been performed to analyze the UAV's behavior as a function of the most influential user-defined parameters. Finally, 10 flight tests were performed to validate the numerical results.

The ever-increasing complexity of software systems makes them hard to comprehend, predict and tune due to emergent properties and non-deterministic behaviour. Complexity arises from the size of software systems and the wide variety of possible operating environments: the increasing choice of platforms and communication policies leads to ever more complex performance characteristics. In addition, software systems exhibit different behaviour under different workloads. Many software systems are designed to be configurable so that policies (e.g. communication, concurrency and recovery strategies) can be chosen to meet the needs of various stakeholders. For complex software systems it can be difficult to accurately predict the effects of a change and to know which configuration is most appropriate. This thesis demonstrates that it is useful to run automated experiments that measure a selection of system configurations. Experiments can find configurations that meet the stakeholders' needs, find interesting behavioural characteristics, and help produce predictive models of the system's behaviour. The design and use of ACT (Automated Configuration Tool) for running such experiments is described, in combination a number of search strategies for deciding on the configurations to measure. Design Of Experiments (DOE) is discussed, with emphasis on Taguchi Methods. These statistical methods have been used extensively in manufacturing, but have not previously been used for configuring software systems. The novel contribution here is an industrial case study, applying the combination of ACT and Taguchi Methods to DC-Directory, a product from Data Connection Ltd (DCL). The case study investigated the applicability of Taguchi Methods for configuring complex software systems. Taguchi Methods were found to be useful for modelling and configuring DC-Directory, making them a valuable addition to the techniques available to system administrators and developers.

De nos jours, le domaine des implants est un des enjeux important pour notre civilisation pour permettre d’améliorer notre quotidien. Pour ce faire, une large offre de matériaux et de technologies existe offrant de nombreuses possibilités afin de répondre aux attentes chirurgicales. Plusieurs familles de matériaux coexistent : les polymères, les céramiques et les matériaux métalliques ainsi que différents procédés de mise en forme. Parmi ceux-ci, le procédé de micro fusion laser sur lit de poudre est un procédé prometteur permettant de réaliser des pièces de géométries complexes. C’est précisément cette technologie qui a été retenue. Pour cela, afin d’approfondir la connaissance du procédé et évaluer l’impact des paramètres sur les structures métallographiques, une orientation se tournant vers des matériaux métalliques tels que le CoCrMo et le titane T40 a été envisagée.Ainsi, les objectifs de la thèse ont été de générer un matériau possédant de bonnes caractéristiques mécaniques ainsi qu’en faciliter son intégration dans un milieu biologique (implants). Pour ce faire, une structure de porosité contrôlée (pour faciliter le développement tissulaire) avec des propriétés mécaniques adaptées aux sollicitations est nécessaire. Le premier travail effectué fut donc une recherche préliminaire afin d’approfondir la connaissance du procédé, en particulier d’un point de vue énergétique et thermique. Afin d’identifier et d’évaluer l’impact des paramètres sur le taux de porosité et donc les propriétés mécaniques de la pièce, une méthode statistique de type Taguchi a été utilisée. Au travers de cette analyse, il est apparu que 3 paramètres inhérents au procédé (la distance entre tache laser, temps d’exposition et le pas de balayage) expliquent prêt de 80% des résultats. De plus, il est mis en évidence que les propriétés mécaniques d’une structure (module de Young et résistance à la rupture) peuvent être maitrisées grâce au taux de porosité de cette dernière et permettre ainsi un rapprochement des propriétés mécaniques de l’os cortical. Pour ce faire, des caractérisations mécaniques ont donc été réalisées pour évaluer le module de Young et la résistance à la rupture des pièces avec différentes structures. Une maîtrise des propriétés peut donc être envisagée et peut même être adaptée en réalisant des structures mixtes alliant partie dense et partie poreuse. / Nowadays, the field of implants is one of the major challenges for our civilization to help improve our lives. To do this, a wide range of materials and technologies are offering many opportunities to meet the surgical needs. Several types of materials exist: polymers, ceramics and metal as well as different methods of shaping materials. Among them , the process of micro fusion laser powder bed is a promising method for producing parts with complex geometries. It is this technology that has been used. To do this, in order to deepen the knowledge of the process and evaluate the impact of parameters on the metallographic structures , guidance , turning to metallic materials such as CoCrMo and titanium T40 was considered.Thus, the objectives of the thesis were to generate a material with good mechanical properties as well as ease of integration in a biological medium ( implants) . To do this, a controlled pore structure (to facilitate tissue development ) with properties adapted to mechanical stress is required. The first work was therefore a preliminary research to deepen understanding of the process , particularly an energy and thermal point of view . To identify and assess the impact of parameters on the porosity and therefore the mechanical properties of the part , a statistical method of Taguchi type was used. Through this analysis, it appeared that three parameters inherent to the process ( the distance between laser spot exposure and no scanning time ) explain 80% loan results. Moreover, it is highlighted that the mechanical properties of a structure ( Young's modulus and tensile strength ) can be controlled through the porosity of the latter and thus permit reconciliation of the mechanical properties of cortical bone . To do this , mechanical characterizations were therefore conducted to evaluate the Young's modulus and tensile strength of parts with different structures. A control properties can be considered and can even be adapted by making composite structures combining dense part and porous part .

The objective of thermal unit commitment is to schedule the on or off status and the real power outputs of units and minimize the system production cost during the period while simultaneously satisfying operational constraints. In this thesis, the Real Genetic Algorithms (RGA) and the Hybrid Taguchi-Genetic Algorithm (HTGA) approaches are presented to solve the thermal unit commitment problem, and comparison with the results obtained using GA. Then this thesis applied the systematic reasoning ability of the Taguchi method operated after mutation can promote the RGA efficiency. The objective of Taguchi method is to improve the quality of offsprings by optimizing themselves to generate a better result, because the offsprings produced randomly by crossover and mutation process is not necessary better than the parents. This method can not only enhance the neighborhood search, but can also search the optimum solution quickly to advance convergence. Finally, it will be shown that the HTGA outperforms RGA by comparing simulation results of unit commitment.

Made available in DSpace on 2014-06-11T19:35:40Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-12-18Bitstream added on 2014-06-13T20:46:43Z : No. of bitstreams: 1 alves_lhd_dr_guara.pdf: 2379670 bytes, checksum: 1a59c87cd9293dbd93fa101676adf89e (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O grande número de variáveis envolvidas no processo de fundição de aço o torna muito susceptível a ocorrência de defeitos relacionados à solidificação como rechupes e poros, que elevam o custo de fabricação devido ao sucateamento e retrabalho de peças. Explorando oportunidades neste cenário, esta Tese de Doutorado, desenvolvida dentro do ambiente industrial de uma fundição de aço propõe um modelo matemático capaz de predizer a combinação de níveis de variáveis de processo que minimizam a ocorrência de rechupes e poros relacionados à solidificação em peças fundidas em aço ABNT 1030, moldadas em areia com aglomerante químico e vazamento por gravidade. Empregou-se, no desenvolvimento do trabalho o Planejamento Estatístico de Experimentos com fatorial fracionado na fase exploratória e o método de Taguchi para validação final e desenvolvimento do Modelo matemático. Na fase de validação, utilizou-se um arranjo ortogonal L16 com quatro réplicas. Na fase exploratória, avaliou-se o volume de defeito em uma peça de produção normal da fundição. Na fase de validação, desenvolveram-se corpos de prova em forma de cubo, vazados sem alimentação. Depois de produzidos, os corpos de prova passavam por ensaio de ultrassom para identificação e localização do defeito e corte de serra, para posterior medição do volume do defeito. O sistema de medição foi validado por meio de um estudo de Análise do Sistema de Medição – MSA GR&R. Os resultados dos experimentos mostraram uma grande diferença entre a contração teórica prevista na literatura de 5,5% para o aço estudado para 2% determinado experimentalmente. As variáveis que mais influenciaram no tamanho do defeito foram: o percentual de carbono queimado no processo de elaboração do aço, percentual de alumínio na liga, percentual de óxido de ferro na areia de moldagem e a interação entre... / The number of variables involved in the steel foundry process makes it susceptible to the occurrence of defects related to the solidification, like shrinkages and pores. This charge the foundries scrap and rework costs. Exploring the opportunities at this context, this Thesis, developed inside an industrial environment of a steel foundry, purposes a mathematic model capable to predict the combination of levels of variables that minimizes the occurrence of shrinkage and pores in carbon steel cast parts, through no-bake sand molding and gravity pouring processes. Statistical Design of Experiments with fractional and factorial at the screening step and Taguchi Method for the final validation and development of the mathematic model were used. A L16 orthogonal array with four replicates was used in the validation step. The defect volume inside a regular cast part was evaluated in the screening step. In the validation step, cube form specimens were designed, poured without feeders and ultrasound tested to bound the defect then saw cut to measure the defect’s volume. In order to the validation of the measurement system, a MSA – Measurement Systems Analysis and Gauge R & R - Gauge Repeatability and Reproducibility were used. The experiments results shown a considerable difference between the theoretical expected contraction, 5.5%, and the 2% experimentally determined. The variables that influenced the defect size the most were: the carbon drop in the steel making, the Aluminum content of the alloy, the Black Iron Oxide content of the molding sand and the interaction between the pouring temperature and the carbon drop. The mathematic model was validated by experiments, which showed a good performance in the prediction of defects.

Reconfigurable antennas are attractive for many military and commercial applications where it is required to have a single antenna that can be dynamically reconfigured to transmit or receive on multiple frequency bands and patterns. RF-MEMS is a promising technology that has the potential to revolutionize RF and microwave system implementation for next generation telecommunication applications. Despite the efforts of top industrial and academic labs, commercialization of RFMEMS switches has lagged expectations. These problems are connected with switch design (high actuation voltage, low restoring force, low power handling), packaging (contamination layers) and actuation control (high impact force, wear, fatique). This Thesis focuses on the design and control of a novel ohmic RF-MEMS switch specified for reconfigurable antennas applications. This new switch design focuses on the failure mechanisms restriction, the simplicity in fabrication, the power handling and consumption, as well as controllability. Finally, significant attention has been paid in the switch’s electromagnetic characteristics. Efficient switch control implies increased reliability. Towards this target three novel control modes are presented. 1) Optimization of a tailored pulse under Taguchi’s statistical method, which produces promising results but is also sensitive to fabrication tolerances. 2) Quantification of resistive damping control mode, which produces better results only during the pull-down phase of the switch while it is possible to be implemented successfully in very stiff devices. 3) The “Hybrid” control mode, which includes both aforementioned techniques, offering outstanding switching control, as well as immunity to fabrication tolerances, allowing an ensemble of switches rendering an antenna reconfigurable, to be used. Another issue that has been addressed throughout this work is the design and optimization of a reconfigurable, in pattern and frequency, three element Yagi-Uda antenna. The optimization of the antenna’s dimensions has been accomplished through the implementation of a novel technique based on Taguchi’s method, capable of systematically searching wider areas, named as “Grid-Taguchi” method.

Orientador: Marcos Valerio Ribeiro / Resumo: A superliga de ferro-níquel VAT 32® foi desenvolvida como um substituto da liga de níquel 751 (UNS N07751) usada na fabricação de válvulas automotivas para motores de combustão interna de alto desempenho. A formação de carbonetos de nióbio confere a esta liga elevada resistência ao desgaste, desejada na aplicação em válvulas automotivas, criando-se, porém, uma maior dificuldade na usinagem deste material. Este trabalho estudou o torneamento da liga VAT 32® com dois tipos de ferramentas cerâmicas, sendo pastilhas de materiais cerâmicos experimentais, a primeira de Al2O3 + MgO e a segunda de Al2O3 + Cr2O3. Por meio do método de Taguchi de planejamento experimental, foram obtidas as combinações e os efeitos principais dos parâmetros de entrada: velocidade de corte, avanço, profundidade de corte para avaliar cada uma das variáveis respostas, sendo elas potência de usinagem, desgaste das ferramentas, qualidade superficial das peças usinadas, e os sinais de emissão acústica e vibração do processo. Foi obtida também, através da análise da razão sinal ruído (S/N) de Taguchi, a composição dos parâmetros de corte em que o processo apresentou menor variabilidade das características de qualidade, o chamado processo robusto. Os tipos de desgaste e tipos de cavaco também foram observados, e foi evidenciado que a pastilha contendo óxido de cromo teve um resultado melhor que a pastilha com óxido de magnésio, sendo superior nos parâmetros de desgaste de flanco (Vb), rugosidade (Ra), apresenta... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre

The thesis outlines the use of Orthogonal Arrays for the training of Artificial Neural Networks. Such arrays are popularly used in system optimisation and are known as Taguchi Methods. The chief advantage of the method is that the network can learn quickly. Fast training methods may be used in certain Control Systems and it has been suggested that they could find application in ‘disaster control,’ where a potentially dangerous system (for example, suffering a mechanical failure) needs to be controlled quickly. Previous work on the methods has shown that they suffer problems when used with multi-layer networks. The thesis discusses the reasons for these problems and reports on several successful techniques for overcoming them. These techniques are based on the consideration of the neuron, rather then the individual weight, as a factor to be optimised. The applications of technique and further work are also discussed.

by Wong Kwok Keung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1994. / Includes bibliographical references (leaves 50-52). / Chapter 1. --- Introduction --- p.1 / Chapter 2. --- Taguchi's methodology in design of experiments --- p.3 / Chapter 2.1 --- System design / Chapter 2.2 --- Parameter design / Chapter 2.3 --- Tolerance design / Chapter 3. --- Inverse Gaussian distribution --- p.8 / Chapter 3.1 --- Genesis / Chapter 3.2 --- Probability density function / Chapter 3.3 --- Estimation of parameters / Chapter 3.4 --- Applications / Chapter 4. --- Iterative procedures and Derivation of the GLIM 4 macros --- p.21 / Chapter 4.1 --- Generalized linear models with varying dispersion / Chapter 4.2 --- Mean and dispersion models for inverse Gaussian distribution / Chapter 4.3 --- Devising the GLIM 4 macro / Chapter 4.4 --- Model fitting / Chapter 5. --- Simulation Study --- p.34 / Chapter 5.1 --- Generating random variates from the inverse Gaussian distribution / Chapter 5.2 --- Simulation model / Chapter 5.3 --- Results / Chapter 5.4 --- Discussion / Appendix --- p.46 / References --- p.50