Dissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Tecnologico / Made available in DSpace on 2016-01-08T20:58:14Z (GMT). No. of bitstreams: 0 Previous issue date: 1996 / Este trabalho visa fornecer um sistema de gerenciamento da qualidade para as fábricas montadoras nacionais. O objetivo principal do trabalho é obter uma melhor utilização da Assistência Técnica dentro das fábricas montadoras. Para isso, viu-se a necessidade de se incorporar qualidade ao produto desde o projeto, garantindo que essa qualidade se mantenha até o envio do mesmo ao consumidor, para, neste momento, o departamento de Assistência Técnica garantir que o usuário tenha a melhor utilização possível do produto, durante a sua vida útil. Então, optou-se pela utilização do QFD (Desdobramento da Função Qualidade) e do Método Taguchi, ferramentas modernas e eficientes na incorporação de qualidade ao produto desde o projeto, adaptando as mesmas para utilização nas fábricas montadoras nacionais.

Tese (doutorado) - Universidade Federal de Santa Catarina, Centro Tecnologico / Made available in DSpace on 2016-01-08T22:24:22Z (GMT). No. of bitstreams: 1 107974.pdf: 8035098 bytes, checksum: 0bdc89f9d47d6bcdce12574bf83136bc (MD5) Previous issue date: 1997 / Revisão e análise dos modelos de planejamento econômico dos gráficos de controle por atributos e modelos correlatos existentes. Agrupamento dos modelos em conjuntos de enfoques semelhantes, com base, especialmente, na formulação da função custo/perda, seus pressupostos básicos e número de parâmetros de custo/risco envolvidos. Indicação do modelo que melhor se adapta às necessidades, características e capacidades de um sistema produtivo. Instruções adicionais, para a tomada de decisão, com base em análise de sensibilidade dos modelos. Estudo comparativo entre o modelo de Taguchi e outros que lhe são compatíveis.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-02-17T19:02:33Z No. of bitstreams: 1 JannaynaDominguesBarrosFilgueira_TESE.pdf: 2533006 bytes, checksum: da08862481c89d6eac3b9988828ee15f (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-02-17T21:41:02Z (GMT) No. of bitstreams: 1 JannaynaDominguesBarrosFilgueira_TESE.pdf: 2533006 bytes, checksum: da08862481c89d6eac3b9988828ee15f (MD5) / Made available in DSpace on 2017-02-17T21:41:02Z (GMT). No. of bitstreams: 1 JannaynaDominguesBarrosFilgueira_TESE.pdf: 2533006 bytes, checksum: da08862481c89d6eac3b9988828ee15f (MD5) Previous issue date: 2016-07-18 / As superf?cies seletivas em frequ?ncia (FSS) pertencem a uma subclasse dos metamateriais composto por patchs met?licos bidimensionais. Por apresentar um comportamento seletivo de frequ?ncia, essas estruturas t?m sido aplicadas nas ?reas de micro-ondas e sistemas de comunica??o, al?m de ser tema de pesquisa por parte de estudiosos. A literatura define basicamente uma superf?cie seletiva de frequ?ncia como um arranjo peri?dico de patches que exibe um determinado comportamento de filtragem de frequ?ncia podendo apresentar caracter?sticas de filtro rejeita-faixa ou passa-faixa que permite a reflex?o ou a transmiss?o de sinais em uma determinada faixa de frequ?ncia. As FSS s?o utilizadas em diversas aplica??es que v?o desde sistemas de micro-ondas e antenas at? aplica??es em radomes e comunica??es via sat?lite. Estruturas de FSS com comportamento Eletromagn?tico adequado t?m sido estudadas por pesquisadores da ?rea. Por?m, a an?lise das caracter?sticas espectrais dessas estruturas requer t?cnica com elevada complexidade computacional. Para reduzir o esfor?o computacional relacionado ao tempo de processamento, podem ser utilizadas t?cnicas de otimiza??o. Neste aspecto, esta pesquisa prop?e a utiliza??o de um m?todo de otimiza??o no projeto de FSS. Os m?todos de otimiza??o t?m como objetivo obter os resultados de um sistema por meio do ajuste dos par?metros de entrada, ou seja, ao ajustar estes par?metros os m?todos de otimiza??o procuram um melhor resultado de sa?da de modo que o desempenho do sistema, em termos de qualidade, custo e efici?ncia, possam ser otimizados. Esses m?todos podem ser aplicados em diversas ?reas. Atualmente, na ?rea do Eletromagnetismo, os circuitos de micro-ondas e antenas t?m aplicado diversas t?cnicas de otimiza??o como Algoritmos Gen?ticos (AG), Redes Neurais Artificiais (RNA) e Otimiza??o com Enxame de Part?culas (PSO). O m?todo de otimiza??o descrito nesta pesquisa foi desenvolvido com base no conceito do arranjo ortogonal, o M?todo Taguchi. Este m?todo fornece uma maneira eficiente para escolher os par?metros do projeto em um processo de Resumo otimiza??o e reduz efetivamente o n?mero de experimentos requerido na aplica??o. O M?todo Taguchi ? aplicado com sucesso nas ?reas de Engenharia Qu?mica e Engenharia Mec?nica e mais recentemente t?m-se aplica??es na ?rea de Eletromagnetismo, por?m, sua aplica??o em FSS ? in?dita. Neste estudo aplicamos o M?todo Taguchi em tr?s estruturas de FSS: FSS quadrada, FSS quadrada com grade e dipolo cruzado. O objetivo do m?todo ? obter os melhores valores para par?metros f?sicos da FSS, de modo que esta opere de forma satisfat?ria, de acordo com especifica??es determinadas no projeto. Utilizamos tamb?m nesta pesquisa o M?todo do Circuito Equivalente. A precis?o do m?todo Taguchi ? verificada experimentalmente e comparada com as simula??es realizadas no ANSYS Designer. Em seguida os prot?tipos s?o constru?dos e caracterizados experimentalmente. / Frequency Selective Surfaces (FSS) belong to a subclass of metamaterial consisting of two-dimensional metallic patches. By presenting a selective frequency behavior, these structures have been applied in the microwave and communication systems fields, in addition it is a research topic by scholars. The literature defines basically the Frequency Selective Surface as a periodic arrangement of patches that displays a certain frequency filtering and it may present characteristics of a band-stop or band-pass filter, which allows the reflection or the transmission of signals in a given frequency range. The FSS are used in diverse applications ranging from microwave and antennas systems to applications in radomes and satellite communications. Researchers have studied FSS structures with appropriate electromagnetic behavior. However, the analysis of spectral characteristics of these structures requires technique with high computational complexity. In order to reduce the computational effort related to the time of processing, it may be used optimization techniques. Concerning to it, this study proposes the use of an optimization method in the FSS design. Optimization methods are intended to get the best results of a system by tuning of the input parameters, that is, when these parameters are adjusted, the optimization methods seek a better output result so that the system performance, in terms of quality, cost and efficiency can be improved. These methods can be applied in various areas. Currently, in the Electromagnetism area, the microwave circuits and antennas have applied various optimization techniques like genetic algorithms (GA), Artificial Neural Networks (ANN) and Particle Swarm Optimization (PSO). The optimization method described in this research was developed based on the concept of the Orthogonal Arrangement, Taguchi Method. This method provides an efficient way to choose the project parameters in a process of optimization and it effectively reduces the number of experiments required in the application. Taguchi method is successfully applied in the chemical engineering and mechanical engineering fields and recently, there have been applications in the Abstract Electromagnetism area, however, its application in FSS is unprecedented. In this study, it was applied the Taguchi Method in three structures of FSS: Square FSS, Square with grid FSS and Cross Dipole. The purpose of the method is to obtain the best values for physical parameters of the FSS, so that is operating in a satisfactory manner, in accordance with certain specifications in the project. It was also used in this research the Equivalent Circuit Method. The accuracy of Taguchi Method is verified experimentally and it is compared to the simulations performed in the ANSYS Designer ?. Then, the prototypes are built and measured in order to validate the experimental results.

[ES] En la presente Tesis Doctoral se investigó la aplicación del proceso de ultrafiltración (UF) y el fenómeno de ensuciamiento de las membranas en la eliminación de sustancias disueltas y coloidales (DCS) de efluentes tratados de la industria papelera (PMTE) para su reutilización en los diferentes procesos de fabricación de papel y cartón reciclado. El objetivo general de esta investigación se dividió en tres partes principales: i) describe cómo encontrar las condiciones óptimas de operación de cuatro parámetros de proceso: presión transmembrana (TMP), velocidad de flujo cruzado (CFV), temperatura y corte de peso molecular (MWCO) para maximizar el flujo promedio de permeado (Jp) y rechazo de la demanda química de oxígeno (COD) y minimizar el descenso del flujo de permeado acumulado (SFD) utilizando el método de Taguchi (Design Robusto) y utility concept aplicado a un proceso de UF a flujo cruzado, para remover DCS de efluentes tratados de la industria papelera, ii) el descenso del flujo de permeado y los mecanismos de ensuciamiento de las membranas de UF ensuciadas con PMTE se examinaron mediante modelos matemáticos semi-empíricos. Los resultados para los diferentes ensayos de UF se expresaron en términos de variación del Jp en función del tiempo para verificar la precisión del ajuste (mayor valor de R2 y menor valor de desviación estándar) de los distintos modelos de Hermia adaptados a flujo tangencial y del modelo de formación de torta en filtración a presión constante ajustados a los datos experimentales, y iii) describe métodos de identificación, caracterización y posibles orígenes de las sustancias contaminantes (foulants) en las membranas de UF. Técnicas como el análisis físico-química, FESEM, SEM-EDS, ATR-FTIR y 3DEEM se llevaron a cabo para comprender qué fracción de los contaminantes son responsables por la formación de incrustaciones en las membranas. Los resultados obtenidos durante la etapa de optimización de parámetros del procesos demostraron que TMP y MWCO tienen la mayor contribución en el Jp y SFD. En el caso de la tasa de rechazo de COD, los resultados mostraron que MWCO tiene la mayor contribución seguida de CFV. Por consiguiente, las condiciones óptimas se encontraron para el segundo nivel de TMP (2.0 bar), el tercer nivel del CFV (1.041 m/s), el segundo nivel de la temperatura (15°C) y el tercer nivel de MWCO (100 kDa). Bajo estas condiciones óptimas de operación Jp, rechazo de COD y SFD alcanzaron respuestas de 81.15 L/m2.h, 43.90% y 6.01 (alrededor de 28.96 % para (FD), respectivamente, valores dentro del rango previsto del intervalo de confianza del 95%. Además, los modelos de Hermia adaptados a UF en flujo tangencial fueron capaces de predecir con gran precisión el descenso del Jp y los mecanismos de ensuciamiento en función del tiempo para todas las membranas seleccionadas (10, 30 y 100 kDa) y bajo diferentes condiciones ensayadas de UF. Por lo tanto, los modelos que presentan un mayor grado de ajuste son el bloqueo completo de poros (coeficiente de determinación R2 >0.97) y bloqueo intermedio (R2 >0.96), seguido por el modelo de formación de torta (R2 >0.94), lo que indica que estés son los principales mecanismos de ensuciamiento de las membranas. Análisis de 3DEEM revelaron que la mayoría de la materia orgánica fluorescentes en las membranas sucias eran proteínas coloidales (componentes similares a proteínas I + II) y proteínas macromoleculares (componentes similares a SMP). Además, polisacáridos (especie celulósica) y sustancias como ácidos grasos y resinosos fueron identificadas en las membranas contaminadas mediante análisis ATR-FTIR. Por fin, análisis SEM-EDS para las membranas ensuciadas con PMTE se detectó concentración de contaminantes inorgánicos (iones metálicos multivalentes) especialmente el Ca2+ que podría acelerar la formación torta en la superficie de la membrana. / [CA] En la present Tesi Doctoral es va investigar l'aplicació del procés d'ultrafiltració (UF) i el fenomen d'embrutiment de les membranes en l'eliminació de substàncies dissoltes i col·loïdals (DCS) d'efluents tractats de la indústria paperera (PMTE) per al seu reutilització en els diferents processos de fabricació de paper i cartó reciclatge. L'objectiu general d'aquesta investigació es va dividir en tres parts principals: i) descriu com trobar les condicions òptimes d'operació de quatre paràmetres de procés: pressió transmembrana (TMP), velocitat de flux creuat (CFV), temperatura i tall de pes molecular (MWCO) per a maximitzar el flux mitjà de permeat (Jp) i rebuig de la demanda química d'oxigen (COD) i minimitzar el descens del flux de permeado acumulat (SFD) utilitzant el mètode de Taguchi (Design Robust) i utility concept aplicat a un procés de UF a flux creuat en escala pilot, per a remoure DCS d'efluents tractats de la indústria paperera (PMTE), ii) el descens del flux de permeat i els mecanismes de embrutiment (fouling) de les membranes de UF embrutades amb PMTE es van examinar mitjançant models matemàtics semi-empírics. Els resultats per als diferents assajos de UF es van expressar en termes de variació del flux de permeat (Jp) en funció del temps per a verificar la precisió de l'ajust (major valor de R2 i menor valor de desviació estàndard) dels diferents models de Hermia adaptats a flux tangencial i del model de formació de coca en filtració a pressió constant ajustats a les dades experimentals, i iii) descriu mètodes d'identificació, caracterització i possibles orígens de les substàncies contaminants (foulants) en les membranes de UF. Tècniques com l'anàlisi física-química, FESEM, SEM-EDS, ATR-FTIR i 3DEEM es van dur a terme per a comprendre quina fracció dels contaminants són responsables per la formació d'incrustacions sobre la superfície i adsorció dins dels porus de les membranes. Els resultats obtinguts durant l'etapa d'optimització de paràmetres del processos van demostrar que TMP i MWCO tenen la major contribució en el Jp i SFD. En el cas de la taxa de rebuig de COD, els resultats van mostrar que MWCO té la major contribució seguida de CFV. Per consegüent, les condicions òptimes es van trobar per al segon nivell de TMP (2.0 bar), el tercer nivell del CFV (1.041 m/s), el segon nivell de la temperatura (15°C) i el tercer nivell de MWCO (100 kDa). Sota aquestes condicions òptimes d'operació Jp, rebuig de COD i SFD van aconseguir respostes de 81.15 L/m².h, 43.90% i 6.01 (al voltant de 28.96% per a (FD)), respectivament, valors dins del rang previst de l'interval de confiança del 95%. A més, els models de Hermia adaptats a UF en flux tangencial van ser capaços de predir amb gran precisió el descens del Jp i els mecanismes de embrutiment en funció del temps per a totes les membranes seleccionades (10, 30 i 100 kDa) i baix diferents condicions assajades de UF. Per tant, els models que presenten un major grau d'ajust són el bloqueig complet de porus (coeficient de determinació R2 >0.97) i bloqueig intermedi (R2 >0.96), seguit pel model de formació de coca (R2 >0.94), la qual cosa indica que estigues són els principals mecanismes de embrutiment de les membranes. Anàlisi de 3DEEM van revelar que la majoria de la matèria orgànica fluorescents en les membranes brutes eren proteïnes col·loidals (components similars a proteïnes I + II) i proteïnes macromoleculars (components similars a SMP). A més, polisacàrids (espècie cel·lulòsica) i substàncies com a àcids grassos i resinosos van ser identificades en les membranes contaminades mitjançant anàlisis ATR-FTIR, tals substàncies exerceixen un paper important en el embrutiment de les membranes. Per fi, anàlisi SEM-EDS per a les membranes embrutades amb PMTE es va detectar concentració de contaminants inorgànics (ions metàl·lics multivalents) especialment el Ca2+ que podria accelerar la formació coca en la àrea de la membrana. / [EN] In this PhD Thesis, the application of ultrafiltration process (UF) and membrane fouling phenomenon used to remove dissolved and colloidal substances (DCS) from paper mill treated effluent (PMTE) for reuse in different recycled paper and cardboard manufacturing processes was investigated. The overall goal of this research has been divided into three main parts: i) describes how to find optimal operating conditions of four controlling parameters, such as transmembrane pressure (TMP), cross-flow velocity (CFV), temperature and molecular weight cut-off (MWCO) for maximizing the average permeate flux (Jp) and chemical oxygen demand (COD) rejection, and minimizing the cumulative flux decline (SFD) using Taguchi method and utility concept for a cross-flow UF in pilot scale, used to remove DCS from a paper mill treated effluent (PMTE), ii) flux decline and fouling mechanisms of UF membranes fouled with PMTE were examined by theoretical modelling. The results from UF tests were expressed in terms of permeate flux (Jp) as a function of time to check modified Hermia's models adapted to crossflow filtration and cake formation in constant-pressure filtration, and iii) describes the Identification, characterization and possible origins of UF membrane foulants. Techniques such as chemical analysis, FESEM, SEM-EDS, ATR-FTIR and 3DEEM analysis were applied to understand which fraction of the foulants caused the fouling. This research found that the TMP and MWCO have the greatest contribution to the average permeate flux and SFD. In the case of the COD rejection rate, the results showed that MWCO has the highest contribution followed by CFV. The optimum conditions were found to be the second level of TMP (2.0 bar), the third level of the CFV (1.041 m/s), the second level of the temperature (15°C), and the third level of MWCO (100 kDa). Under these optimum conditions Jp, COD rejection and SFD resistance of 81.15 L/m2/h, 43.90% and 6.01 (around 28.96 % of (FD), respectively, were obtained and they were within of the predicted range at the 95% confidence interval. Furthermore, the results showed that the predictions of the modified Hermia's models adapted to cross-flow UF had good agreements with experimental data, under different conditions tested for PMTE. Therefore, it can be concluded that for all cases the best fit (higher accuracy) to the experimental data corresponds to the complete (coefficient of determination R2 >0.97) and intermediate (R2 >0.96) blocking, followed by the cake layer formation (R2 >0.94). Moreover, measurements of particle size distribution and zeta potential near the isoelectric point, showed a substantial reduction in colloidal compounds. The 3DEEM analysis revealed that the majority of the organic foulants with fluorescence characteristics on the fouled membranes were colloidal proteins (protein-like substances I+II) and macromolecular proteins (SMP-like substances). Further, polysaccharide (cellulosic specie), fatty and resin acid substances were identified on the fouled membrane by the ATR-FTIR analysis and they play an important role in membrane fouling. In addition, the membrane SEM-EDS analysis showed accumulate and adsorbed onto the membrane surfaces of inorganic foulants, such as multivalent metal ions and especially Ca2+ (acts as a binding agent) that could accelerate cake layer formation on the membrane. / Santos Sousa, MR. (2020). Optimization of Operation Parameters in Ultrafiltration by Experiment Design, Mathematical Modelling and Fouling Characterization of the Membranes Used to Remove Dissolved and Colloidal Substances from a Treated Paper Mill Effluent [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/155975 / TESIS

A method for performing robust optimal design that combines the efficiency of experimental designs and the accuracy of nonlinear programming (NLP) has been developed, called Search-and-Zoom. Two case studies from the RF and communications industry, a high-frequency micro-strip band-pass filter (BPF) and a rectangular, directional patch antenna, were used to show that sensitivity optimization could be effectively performed in this industry and to compare the computational efficiency of traditional NLP methods (using fmincon solver in MATLAB R2013a) and they hybrid method Search-and-Zoom. The sensitivity of the BPF's S11 response was reduced from 0.06666 at the (non-robust) nominal optimum to 0.01862 at the sensitivity optimum. Feasibility in the design was improved by reducing the likelihood of violating constraints from 20% to nearly 0%, assuming RSS (i.e., normally-distributed) input tolerances and from 40% to nearly 0%, assuming WC (i.e., uniformly-distributed) input tolerances. The sensitivity of the patch antenna's S11 function was also improved from 0.02068 at the nominal optimum to 0.0116 at the sensitivity optimum. Feasibility at the sensitivity optimum was estimated to be 100%, and thus did not need to be improved. In both cases, the computation effort to reach the sensitivity optima, as well as the sensitivity optima with RSS and WC feasibility robustness, was reduced by more than 80% (average) by using Search-and-Zoom, compared to the NLP solver.

This project was funded by Automotive Partnership Canada (APC), an initiative created by the Government of Canada in an attempt to support significant, collaborative R&D activities in order to benefit the entire Canadian automotive industry. / High strength Al-Zn-Mg-Cu alloys have been increasingly employed in the transportation industry due to the increased demands for light structural components. However, their applications have been limited to relatively expensive wrought products. Application of the shape cast Al-Zn-Mg-Cu parts has never been the focus of attention due to their poor castability and mechanical properties. Improving the casting quality is expected to increase their utilization within the automotive industry. The poor castability and mechanical properties of some alloys in this family may be effectively improved through optimized chemistry control and melt treatment including grain refinement. The primary objective of this project is to optimize the chemistry and heat treatment of the Al-Zn-Mg-Cu alloy family that results in improved strength with acceptable level of ductility and casting quality relative to other shape cast Al alloys. The Taguchi experimental design method was used to narrow down the number of required casting experiments required to meet the research objective. Three levels across four elements yielded a total of 9 Al-Zn-Mg-Cu alloys, which were cast using a tilt pour permanent mold process. The effect of each major alloying element on the microstructure, and mechanical properties was investigated. Tensile measurements were made on the 9 alloys subjected to two steps solution treatments. Mechanical properties such as yield strength (YS), ultimate tensile strength (UTS), and elongation at fracture (El.%) were experimentally measured and statistically analyzed. An ANOVA analysis was employed to quantify the percentage contribution of the alloying elements on the material properties. Grain refinement was found to play a significant role in improving the hot tearing resistance and, thereby ameliorating quality. The alloying element that affected the YS and UTS to the greatest extent was Cu, followed by Zn. In contrast, the effect of Mg and Ti on YS and UTS was insignificant. Moreover, a decrease in Mg content had the greatest effect in enhancing the El.%. A regression analysis was used to obtain statistical relationships (models) correlating the material properties with the variations in the content of the major alloying elements. The R-square values of YS, UTS, and El.% were 99.7 %, 98 %, and 90 %, respectively, showing that the models replicated the experimental results. Verification measurements made on shape cast Al-6Zn-2Mg-2Cu alloy revealed that the material property model predictions were in agreement with the experimentally measured values. The results show that secondary and over ageing treatments of the shape cast Al-Zn-Mg-Cu alloys lead to superior combination of YS and El.%. The ongoing advances in shape casting of Al-Zn-Mg-Cu alloys with high will make them suitable choices for commercial load-bearing automotive components, when it comes to the selection of a material meeting the minimum requirements for strength, damage tolerance, cost and weight. / Thesis / Master of Applied Science (MASc)

Use of weight-saving materials to produce lightweight components with enhanced dimensional control is important to the automotive industry. This has increased the need to understand the material behavior with respect to the forming process at the microstructural level. A test matrix was developed based on the orthogonal array of Taguchi design of experiment (DOE) approach. Experiments were conducted for the V-bending process using 6022-T4 AA to study the variation of springback due to both process and material parameters such as bend radius, sheet thickness, grain size, plastic anisotropy, heat treatment, punching speeds, and time. The design of experiments was used to evaluate the predominate parameters for a specific lot of sheet metal. It was observed that bend radius had greatest effect on springback. Next, finite element simulation of springback using ANSYS implicit code was conducted to explore the limits regarding process control by boundary values versus material parameters. 2-D finite element modeling was considered in the springback simulations. A multilinear isotropic material model was used where the true stress-strain material description was input in discrete form. Experimental results compare well with the simulated predictions. It was found that the microstructure of the material used in this study was processed for sheet metal forming process.

This dissertation presents an empirical approach to identifying decisions for adapting cognitive radio parameters with no a priori knowledge of the environment. Cognitively inspired radios, attempt to combine observed metrics of system performance with artificial intelligence decision-making algorithms. Current architectures trend towards hybrid combinations of heuristics, such as genetic algorithms (GA) and experiential methods, such as case-based reasoning (CBR). A weakness in the GA is its reliance on limited mathematical models for estimating bit error rate, packet error rate, throughput, and signal-to-noise ratio. The CBR approach is similarly limited by its dependency on past experiences. Both methods have potential to suffer in environments not previously encountered. In contrast, the statistical methods identify performance estimation models based on exercising defined experimental designs. This represents an experiential decision-making process formed in the present rather than the past. There are three core contributions from this empirical framework: 1) it enables a new approach to decision making based on empirical estimation models of system performance, 2) it provides a systematic method for initializing cognitive engine configuration parameters, and 3) it facilitates deeper understanding of system behavior by quantifying parameter significance, and interaction effects. Ultimately, this understanding enables simplification of system models by identifying insignificant parameters. This dissertation defines an abstract framework that enables application of statistical approaches to cognitive radio systems regardless of its platform or application space. Specifically, it assesses factorial design of experiments and response surface methodology (RSM) to an over-the-air wireless radio link. Results are compared to a benchmark GA cognitive engine. The framework is then used for identifying software-defined radio initialization settings. Taguchi designs, a related statistical method, are implemented to identify initialization settings of a GA. / Ph. D.

With the intention of improving process efficiency and product quality, Design of Experiments (DoE) is often applied in industry as a tool for engineers in planning experiments and product design. One of the methods in DoE is the Taguchi method that provides a practical procedure to design experiments with an appropriate number of tests for identifying optimal parametric settings of products. The purpose of this study is to investigate how Design of Experiments, specifically the Taguchi method, can be applied to analyze the performance of the lubricants used to stainless fasteners which are common products in various industries. The study has been implemented at the case company which manufactures stainless fasteners. Combining with qualitative and quantitative methods, this thesis work started from reviewing the previous research on DoE, the Taguchi method and various analysis tools as well as information on lubricants. Experiments were designed based on observations, interview, and literature collection. The actual tests were carried out on the case company's wax process where lubricants with different levels for the respective parameters were applied to screws and nuts. The fasteners were tested in a friction machine and UV lamp at the case company. The implemented experiments and data analysis have demonstrated which levels for each parameter of the lubricant resulted in the least friction. The parameters considered in the experiments are mixing ratio, number of dips and time in dips, which have varying degrees of influence on friction. The outcomes contribute to improved process efficiency and product quality within the steel industry and similar sectors. The visualization of the lubricant by UV lamp confirmed that the coating varied according to the levels of the parameters. The theoretical contribution can facilitate further research while the practical contribution provides an approach for experiments in quality improvement on lubricants. The thesis work has also identified the need for further research into the drying process after lubricant application and the use of UV lamp to improve control of lubricant smoothness.

Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2009. / ENGLISH ABSTRACT: Polysulfone hollow-fibre membranes were fabricated via the dry-wet solution spinning technique. The objective was to demonstrate the influence of the various fabrication parameters on the diameter and mechanical properties of the hollow-fibre membranes and to optimize the spinning process by controlling these parameters with a computer control system. The effects of the operation parameters were investigated using an experimental design based on a fractional factorial method (Taguchi’s design of experiments). The parameters that were considered are the spinneret size, dope solution temperature, bore fluid temperature, coagulation bath temperature, dope extrusion rate, bore flow rate and the take-up speed. A new pilot solution spinning plant was installed and upgraded, and a computer control system, based on LabView, was developed to control, monitor and log the experimental data. The diameter of the hollow-fibres were determined using a scanning electron microscope (SEM) while the mechanical properties were measured using a tensile tester. The effects of diameter size and wall thickness of the hollow-fibres on the performance of the membranes were studied. The results showed the significance of the fabrication parameters that dominate the diameter and strength of the hollow-fibres. / AFRIKAANSE OPSOMMING: Polisulfoon holvesel membrane is met ‘n droë-nat oplossingspin proses vervaardig. Die doel hiermee was om die invloed van verskeie vervaardigingsparameters op die deursnee en meganiese eienskappe van die holvesel membrane te demonstreer asook om die spin proses te optimeer deur gerekenariseerde beheer van die aanleg. ‘n Eksperimentele ontwerp, gebaseer op ‘n gedeeltelike faktoriaal metode (Taguchi se eksperimentele ontwerp) is gebruik om die invloed van die vervaardigingsparameters te ondersoek. Die parameters wat oorweeg is, is spindop grootte, materiaal temperatuur, boorvloeistof temperatuur, stolbad temperatuur, materiaal ekstrusie tempo and opwen spoed. ‘n Nuwe oplossingspin loodsaanleg was geïnstalleer en opgegradeer en ‘n rekenaar beheerstelsel, gebaseer op LabView, is ontwikkel om die aanleg te beheer, moniteer en eksperimentele data te stoor. Die deursnee van die holvesel is gemeet met ‘n skanderingelektron mikroskoop (SEM) terwyl die meganiese eienskappe bepaal is met ‘n trektoets apparaat. Die effek van die deursnee en wanddikte van die holvesels op die werkverrigting van die membrane is ook bestudeer. Die resultate toon watter vervaardigingsparameters is beduidend vir die deursnee en sterkte van die holvesels.