Modernas empresas buscam a melhoria dos seus sistemas de produção a fim de aumentar a produtividade e a qualidade de produtos. A tecnologia de usinagem com alta velocidade de corte (HSC – High Speed Cutting) permite aumentar as taxas de remoção de material, as tolerâncias dimensionais das partes, assim como a melhoria do acabamento superficial. Esta tecnologia tem sido muito utilizada na área de moldes e matrizes, pois minimiza o problema de tempos de usinagem longos na produção e facilita a obtenção de elevada qualidade, normalmente requerida em produtos injetados/estampados. Atualmente diversos estudos em HSC buscam a compreensão de todos os fatores envolvidos nesta nova tecnologia, a qual ainda apresenta diversas lacunas para o seu domínio completo. Este trabalho apresenta um estudo teórico-experimental da condução de calor nos materiais utilizados na fabricação de moldes e matrizes, submetidos à usinagem com altas velocidades de corte. Os materiais utilizados foram os aços AISI D2 e AISI H13, muito utilizados nas indústrias de moldes e matrizes. Os ensaios foram realizados com fresas de ponta esférica (Ball-Nose) com e sem angulação de hélice e com ângulo de saída neutro, com revestimento de ('TI'AL')N e com uma ferramenta com uma placa de PcBN soldada. A geometria dos corpos-de-prova foi projetada para aproximar-se dos casos reais encontrados na indústria, casos em que superfícies complexas e paredes finas são comumente usinadas em matrizes e moldes. Dois sistemas de resfriamento foram utilizados o ar comprimido e o ar gelado utilizando o princípio de vórtice. Os resultados demonstraram uma pequena variação da temperatura em função dos materiais empregados e das ferramentas utilizadas em todos os testes. Os dois sistemas de resfriamento alternativos à usinagem sem sistema de resfriamento, o ar comprimido e o ar frio, também se mostraram influentes na variação da temperatura, apresentando-se como sistemas capazes de remover o calor no processo durante a usinagem. Os desvios de forma encontrados nos corpos-de-prova devem ser considerados na usinagem de perfis muito precisos. Foram encontrados valores da ordem de 0,05 mm, oriundos da combinação de distorções térmicas e tensões residuais, os quais podem ser signifcativos, por exemplo, na área de matrizaria / Modern industries improve their production systems aiming at increasing productivity and product quality. The high speed cutting (HSC) technology allows increasing material removal rates, workpiece dimensional tolerance and surface finishing as well. Such technology has been extensively used in mold and dies areas, since it minimizes long lead time problems and facilitates high quality machining, normally required in injected/formed products. Nowadays, several studies in HSC search for understanding all aspects involved in that new technology, which still presents many blanks for its complete knowledge. The present work shows a theoretical and experimental study on heat conduction flux in materials applied to mold and die manufacturing, submitted to HSC. It was used AISI D2 and AISI H13, which are very common in mold and die industry. Two types of cylindrical ball nose end milling tool were used in the experiments: coated with ('TI'AL')N and tipped with PcBN. The workpiece geometry was conceived to simulate real cases found in industry, in which complex surfaces and slim walls are commonly machined in molds and dies. Two cooling systems were used in the tests, compressed air and frozen air using vortex principle. Results indicated a small temperature variation as a function of machined materials and tools throughout the tests. Both alternative cooling systems, besides dry machining, compressed air and cool air, also demonstrated to affect temperature variation. They presented themselves as capable of remove heat from the process during the cutting operation. The form deviation found in the workpiece after machining must be considered when machining very precise profiles. Variation values of around 0.05 mm were found, coming from a combination of thermal distortion and residual stress, which could be significant, for example, in die manufacturing area

<p>U ovoj doktorskoj disertaciji optimizacioni algoritam zasnovan na roju čestica primenjen je na problem optimalnog upravljanja pogonom sa sinhronim motorom sa stalnim magnetima pri velikim brzinama obrtanja. Kako bi se smaljili ukupni kontrolabilni gubici električne energije i povećala efikasnost generisan je optimalni referentni vektor statorske struje za širok opseg brzina i za različita opterećenja. Prilikom optimizacije uvažena su naponska i strujna ograničenja invertora, kao i promena induktivnosti statora. Sačuvani generisani vektor struje statora se koristi prilikom upravljanja pogonom u realnom vremenu, a predloženi algoritam se poredi sa standarnim upravljanjem.</p> / <p>In this PhD thesis, a particle swarm optimization algorithm was applied to the problems of optimal control of high speed permanent magnet motor drives. In order to minimize the total controllable electrical losses and to increase the efficiency, the optimum current vector references are calculated offline for the wide speed range and for different load conditions. The voltage and current limits of the drive system and the variation of stator inductances are all included in the optimization method. The stored optimal current vector references are used during the real time control and the proposed algorithm is compared with the conventional control algorithm.</p>

Three methods were developed to better understand and characterize the near-field dynamic processes of rotary bell atomization. The methods were developed with the goal of possible integration into industry to identify equipment changes through changes in the primary atomization of the bell. The first technique utilized high-speed imaging to capture qualitative ligament breakup and, in combination with a developed image processing technique and PIV software, was able to gain statistical size and velocity information about both ligaments and droplets in the image data. A second technique, using an Nd:YAG laser with an optical filter, was used to capture size statistics at even higher rotational speeds than the first technique, and was utilized to find differences between serrated and unserrated bell ligament and droplet data. The final technique was incorporating proper orthogonal decomposition (POD) into image data of a side-profile view of a damaged and undamaged bell during operation. This was done to capture differences between the data sets to come up with a characterization for identifying if a bell is damaged or not for future industrial integration.

Les objectifs essentiels des opérateurs ferroviaires sont d'augmenter la sécurité, de réduire les coûts d'exploitation et de maintenance, d'accroître l'attrait et le profit en offrant de nouveaux services aux passagers. Ces objectifs seront atteintsgrâce à une énorme augmentation des échanges de flux de données entre les infrastructures existantes et les technologies présentes utilisées au bord du train. L'efficacité spectrale, l'optimisation des ressources radioélectriques ainsi que l'interopérabilité mais aussi la fiabilité des communications sont des éléments majeurs pour les applications ferroviaires. Ces contraintes ainsi que l'utilisation sporadique des bandes fréquentielles à disposition ont donné le jour à la radio intelligente. Cette dernière se présente comme une technologie émergente qui améliore les performances des systèmes radio existants en intégrant l'intelligence artificielle avec la radio logicielle. Un système radio intelligent est défini par sa capacité à être conscient de son environnement radioélectrique. En effet, afin d'optimiser au maximum les opportunités spectrales qui lui sont offerts, le dispositif radio intelligent doit être capable de transmettresur des bandes laissées libres tout en réalisant un sondage spectral afin de ne pas interférer avec les utilisateurs ayant la priorité sur la bande mais aussi pour détecter d'autres fréquences vacantes. Dans le cadre de cette thèse, nous proposons de nous concentrer sur la problématique de détection de spectre dans un environnement à très forte mobilité. Certaines contraintes sont à prendre en compte, telles que la vitesse. À cela s'ajoute les contraintes de réglementation concernant les critères de détection, telles que la norme IEEE 802.22 WRAN qui stipule que la détection d'un utilisateur prioritaire doit être réalisée à -21dB dans un laps de temps de 2 secondes. L'objectif est donc de concevoir un terminal radio intelligent dans les conditions physiques et réglementaires de transmission dans un environnent ferroviaire. / The main objectives of railway operators are to increase safety, reduce operating and maintenance costs, increase attractiveness and profit by offering new services to customers. These objectives will be achieved through a huge increase of data fluxes between existing infrastructure and the technologies currently used on the train. Spectral efficiency, optimization of radio resources, interoperability and reliability of communications are major elements for railway applications. These constraints and the sporadic use of available frequency bands have gave rise to cognitive radio. Cognitive radio is an emerging technology that improves the performance of existing radio systems by integrating artificial intelligence with software radio. A cognitive radio system is defined by its ability to be aware of its radio environment. Indeed, to optimize as much as possible the available spectral opportunities, the cognitive radio device must be able to transmit on free bands while performing a spectrum sensing to not interfere with users having priority on the band and to detect other vacant frequencies. As part of this thesis, we propose to focus on the problem of spectrum detection in a highly mobile environment. Some constraints should be considered, such as speed. Added to this, there are regulatory constraints on detection criteria, such as the IEEE 802.22 WRAN standard, which stipulates that detection of a priority user must be performed at -21 dB within a period of 2 seconds. The objective is therefore to design an intelligent radio terminal in the physical and regulatory conditions of transmission in a railway environment.

<p>In the last decades, high speed cutting has become an attractive technology in the wood industry. The possibility of reducing global costs in addiction with an increase in productivity, were the main reasons for the enlargement of the use of this technology. </p><p>As usual, these advantages are accompanied by disadvantages that should be carefully analyzed. If on the one hand cutting forces are reduced with increasing cutting speeds, on the other hand, the centrifugal forces affecting the tool are higher. Exposed to such high loads, there is a considerable risk of tool failure that embeds hazards for both machine and workers. </p><p>To prevent the risk of accidents and to guarantee safety in use, security standards have been implemented in industrial fields, imposing specific experimental tests, with defined procedure modes. Accordingly with these standards, the results obtained through the tests should fall inside limited ranges. The experimental tests suggested on the European Standards are intended to simulate the real working conditions of a rotating cutting tool, where extreme centrifugal forces are imposed by the high values of speed. Although their main importance, these destructive tests aren’t always practicable. It happens, for instance, with tools produced in small batch sizes, or as an ascertainment for the fail-critical speed during the development stage, or even due to physical incompatibilities between the tool and the laboratory testmachines. </p><p>The high value of weight associated with the cutting tool prototype developed and patented by Verktygs Larsson AB was an impediment to run the laboratorial tests specified by the standards, forcing the company to find a new way to assure the safety requirements of their product. </p><p>The main goal of this project was the development of an alternative method based on finite element theory to perform a safety analysis to the prototype of a wood cutter. This tool is used as a component in high speed planers. </p><p>In addiction to this primary objective, some considerations were made about other available models, with increased dimensions or even with different parameters. If there was the need, design changes could be assumed in order to guarantee that the tool reached the requisites of the safety standards. Considering an optimization effort, material changes would also be considered, to aim in the direction of reducing the tool weight and the consequent centrifugal forces.</p>

Experimental and numerical analysis of loads and responses for planing craft in waves is considered. Extensive experiments have been performed on a planing craft, in full-scale as well as in model scale. The test set-ups and significant results are reviewed. The required resolution in experiments on planing craft in waves, concerning sampling frequencies, filtering and pressure transducer areas, is investigated. The aspects of peak identification in transient signals, fitting of analytical cumulative distribution functions to sampled data, and statistical convergence are treated. A method for reconstruction of the momentary pressure distribution at hull-water impact, from measurements with a limited number of transducers, is presented. The method is evaluated to full-scale data, and is concluded to be applicable in detailed evaluation of the hydrodynamic load distribution in time-domain simulations. Another suggested area of application is in full-scale design evaluations, where it can improve the traceability, i.e. enable evaluation of the loads along with the responses with more confidence. The presented model experiment was designed to enable time-domain monitoring of the complete hydromechanic pressure distribution on planing craft in waves. The test set-up is evaluated by comparing vertical forces and pitching moments derived from acceleration measurements, with the corresponding forces derived with the pressure distribution reconstruction method. Clear correlation is found. An approach for direct calculations of loads, as well as motion and structure response, is presented. Hydrodynamic loads and motion responses are calculated with a non-linear time-domain strip method. Structure responses are calculated by applying momentary distributed pressure loads, formulated from hydrodynamic simulations, on a global finite element model with inertia relief. From the time series output, limiting conditions and extreme responses are determined by means of short term statistics. Promising results are demonstrated in applications, where extreme structure responses derived by the presented approach, are compared with responses to equivalent uniform rule based loads, and measured responses from the full-scale trials. It is concluded that the approach is a useful tool for further research, which could be developed into a rational design method.

In the last decades, high speed cutting has become an attractive technology in the wood industry. The possibility of reducing global costs in addiction with an increase in productivity, were the main reasons for the enlargement of the use of this technology. As usual, these advantages are accompanied by disadvantages that should be carefully analyzed. If on the one hand cutting forces are reduced with increasing cutting speeds, on the other hand, the centrifugal forces affecting the tool are higher. Exposed to such high loads, there is a considerable risk of tool failure that embeds hazards for both machine and workers. To prevent the risk of accidents and to guarantee safety in use, security standards have been implemented in industrial fields, imposing specific experimental tests, with defined procedure modes. Accordingly with these standards, the results obtained through the tests should fall inside limited ranges. The experimental tests suggested on the European Standards are intended to simulate the real working conditions of a rotating cutting tool, where extreme centrifugal forces are imposed by the high values of speed. Although their main importance, these destructive tests aren’t always practicable. It happens, for instance, with tools produced in small batch sizes, or as an ascertainment for the fail-critical speed during the development stage, or even due to physical incompatibilities between the tool and the laboratory testmachines. The high value of weight associated with the cutting tool prototype developed and patented by Verktygs Larsson AB was an impediment to run the laboratorial tests specified by the standards, forcing the company to find a new way to assure the safety requirements of their product. The main goal of this project was the development of an alternative method based on finite element theory to perform a safety analysis to the prototype of a wood cutter. This tool is used as a component in high speed planers. In addiction to this primary objective, some considerations were made about other available models, with increased dimensions or even with different parameters. If there was the need, design changes could be assumed in order to guarantee that the tool reached the requisites of the safety standards. Considering an optimization effort, material changes would also be considered, to aim in the direction of reducing the tool weight and the consequent centrifugal forces.

This thesis explores the clock and data recovery (CDR) for the high-speed blind-sampling ADC-based receivers. This exploration results in two new CDR architectures that reduce the receiver complexity and save the ADC power and area compared to the previous work. The two proposed CDR architectures constitute the primary contributions of this thesis. The first proposed architecture, a 2x feed-forward CDR architecture, eliminates the interpolating feedback loop, used in the previously reported CDRs, in order to reduce the CDR circuit complexity. Instead of the feedback loop, the proposed architecture uses a feed-forward topology to recover the phase and data directly from the blind digital samples of the received signal. The 2x feed-forward CDR architecture was implemented and characterized in a 5 Gb/s receiver test-chip in 65 nm CMOS. The test-chip measurements confirm that the CDR successfully recovers the data with bit error rate (BER) < 10e-12 in the presence of jitter. The second proposed architecture, a fractional-sampling-rate (FSR) CDR architecture, reduces the receiver sampling rate from the typical integer rate of 2x the baud rate to a fractional rate between 2x and 1x in order to reduce the ADC power and area. This architecture employs the feed-forward topology of the first contribution of this thesis to recover the phase and data from the fractionally-spaced digital samples of the signal. To verify the proposed FSR CDR architecture, a 1.45x receiver test-chip was implemented and characterized in 65 nm CMOS. This test-chip recovers 6.875 Gb/s data from the ADC samples taken at 10 GS/s. The measurements confirm a successful data recovery in the presence of jitter with BER < 10e-12. With sampling at 1.45x, the FSR CDR architecture reduces the ADC power and area by 27.3% compared to the 2x feed-forward CDR architecture, while the overall receiver power and area are reduced by 12.5%.

Sootblowing is a process in which supersonic steam jets are used to periodically blast deposits off heat transfer tubes in kraft recovery boilers. However, sootblowing significantly consumes the valuable high pressure steam generated by the boiler, hence it should be optimized. A recovery boiler consists of three convective sections - superheater, generating bank and economizer. The tube arrangement in these sections, particularly the tube spacing is different from each other. Moreover, tubes in an economizer are finned. A sootblower jet will interact differently with these tube arrangements, potentially affecting its strength, and hence deposit removal capability. The objective of this work was to characterize jet/tube interaction in the three sections of a recovery boiler. Lab-scale experiments were conducted in which these interactions were visualized using the schlieren technique coupled with high-speed video, and were quantified by pitot pressure measurements. This work is the first to visualize the interactions. The offset between the jet and tube centrelines, the nozzle exit diameter relative to the tube diameter, and the distance between the nozzle and tube were varied to examine their effects on jet/tube interaction. Results showed that due to the very low spreading rate of a supersonic jet, a jet (primary jet) stops interacting with a superheater platen when the jet is only a small distance away from it. When the jet impinges on a tube, the jet deflects at an angle, giving rise to a weaker ‘secondary’ jet. Due to the large inter-platen spacing, a secondary jet has an insignificant impact in a superheater. In a generating bank, the primary jet weakens between the closely spaced tubes due to increased mixing. However, a secondary jet impinges on the adjacent tubes exerting a high impact pressure on those tubes. The primary jet also weakens between finned economizer tubes, but remains stronger for a greater distance than in a generating bank. As in the case inside a generating bank, a secondary jet also impinges on adjacent rows of tubes in an economizer. The results imply that in a superheater, a sootblower jet must be directed close to the platens to yield useful jet/deposit interactions, and to avoid wasting steam by blowing between the platens. In a generating bank, deposits beyond the first few tubes of a row experience a weaker sootblower jet, and thus may not be removed effectively. However, secondary jets may contribute to removing deposits from the first few adjacent tubes. They may also induce erosion-corrosion of those tubes. Secondary jets may also help remove deposits from adjacent rows in a finned tube economizer. In an economizer, the strength and hence, the deposit removal capability of a sootblower jet diminish only slightly beyond the supersonic portion of the jet. A mathematical model was also developed to determine the feasibility of using inclined sootblower nozzles in recovery boiler superheaters, and suggests that it may be possible to clean superheater platens more effectively with slightly inclined nozzles.

This thesis explores the clock and data recovery (CDR) for the high-speed blind-sampling ADC-based receivers. This exploration results in two new CDR architectures that reduce the receiver complexity and save the ADC power and area compared to the previous work. The two proposed CDR architectures constitute the primary contributions of this thesis. The first proposed architecture, a 2x feed-forward CDR architecture, eliminates the interpolating feedback loop, used in the previously reported CDRs, in order to reduce the CDR circuit complexity. Instead of the feedback loop, the proposed architecture uses a feed-forward topology to recover the phase and data directly from the blind digital samples of the received signal. The 2x feed-forward CDR architecture was implemented and characterized in a 5 Gb/s receiver test-chip in 65 nm CMOS. The test-chip measurements confirm that the CDR successfully recovers the data with bit error rate (BER) < 10e-12 in the presence of jitter. The second proposed architecture, a fractional-sampling-rate (FSR) CDR architecture, reduces the receiver sampling rate from the typical integer rate of 2x the baud rate to a fractional rate between 2x and 1x in order to reduce the ADC power and area. This architecture employs the feed-forward topology of the first contribution of this thesis to recover the phase and data from the fractionally-spaced digital samples of the signal. To verify the proposed FSR CDR architecture, a 1.45x receiver test-chip was implemented and characterized in 65 nm CMOS. This test-chip recovers 6.875 Gb/s data from the ADC samples taken at 10 GS/s. The measurements confirm a successful data recovery in the presence of jitter with BER < 10e-12. With sampling at 1.45x, the FSR CDR architecture reduces the ADC power and area by 27.3% compared to the 2x feed-forward CDR architecture, while the overall receiver power and area are reduced by 12.5%.