Made available in DSpace on 2015-03-04T18:51:05Z (GMT). No. of bitstreams: 1 Dissert_MSc_JaquelineAngelo.pdf: 926474 bytes, checksum: da4b07a3aac6c41fe497e0351128bde1 (MD5) Previous issue date: 2008-07-24 / Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior / This dissertation presents the BicriterionAnt, MACS and MONACO Ant Colony algorithms, available in literature, to solve the Multi-Objective Traveling Salesman Problem (MOTSP). The characteristics of the problem and of each algorithm used are presented. Those algorithms were tested in six bi-objective instances of MOTSP. Changes in the original algorithms were implemented to try to produce better results than the original ones. To validate the results and to measure the quality of the solutions, metrics of performance were used which help to identify the best non-dominated solution sets. / Esta dissertação apresenta os algoritmos BicriterionAnt, MACS e MONACO, disponíveis na literatura, baseados em colônia de formigas, para resolução do Problema do Caixeiro Viajante Multiobjetivo (PCVMO). São apresentadas as características do problema e de cada algoritmo utilizado. Estes algoritmos foram testados em seis instâncias bi-objetivo do PCVMO. Foram implementadas algumas alterações na estrutura original dos algoritmos na tentativa de produzir resultados melhores do que os algoritmos originais. Para a avaliação dos resultados e medição da qualidade das soluções, foram utilizadas métricas de desempenho que auxiliam na identificação dos melhores conjuntos de soluções não-dominadas.

Made available in DSpace on 2016-01-26T12:51:19Z (GMT). No. of bitstreams: 1 jorgeadasdib_tese.pdf: 343099 bytes, checksum: ab3f9830cf7060129a9dcec26d66b83c (MD5) Previous issue date: 2008-06-06 / No previous study has addressed the question of prognostic determinants for patients with Chagas? cardiomyopathy at the terminal stage listed for heart transplantation. Casuistic and Method: All patients listed for heart transplantation at our institution from August, 2000 to March, 2005 were considered for the study. Patients removed from the waiting list for clinical status improvement were excluded from the investigation. Patients were followed until death, cardiac transplantation or the end of the study period. Cardiac transplant recipients were censored at the time of transplantation. No patient was lost to follow up. A Cox regression hazards model was used to establish independent predictors of all-cause mortality. Variables previously demonstrated to predict mortality in either Chagas or non-Chagas? disease heart failure were entered the univariate analysis. Separate analyses were performed for Chagas and non-Chagas? disease patients. Results: Median follow up was 32 (15,121) days in Chagas disease and 79 (14,151) days in non-Chagas? disease patients. In Chagas disease patients, the hemodynamic instability (p=0.01; hazard ratio=0,077, 95% confidence interval, 0.01 to 0.58) as well as the transpulmonary gradient (p=0.02; hazard ratio=1.15, 95% confidence interval, 1.02 to 1.30) were retained as independent predictors of all-cause mortality. Serum sodium levels (p=0.002; hazard ratio=0.81; 95% confidence interval, 0.71 to 0.93) was independent predictor of all-cause mortality for non-Chagas? disease patients. Conclusion: The hemodynamic instability and transpulmonary gradient were independent predictors of all-cause mortality for Chagas? disease patients listed for heart transplantation. A larger, prospective cohort study is needed to validate our findings. / At? agora nenhum estudo preocupou-se em estabelecer determinantes progn?sticos para pacientes com insufici?ncia card?aca cr?nica terminal secund?ria ? cardiomiopatia da doen?a de Chagas na fila de espera para transplante card?aco. Casu?stica e M?todo: Todos os pacientes alocados em fila de espera de transplante card?aco em nossa institui??o, de agosto de 2000 a mar?o de 2005, foram inicialmente considerados para o estudo. Os pacientes que foram removidos da lista de espera em virtude de melhora no estado cl?nico foram retirados do estudo. Os pacientes foram acompanhados at? a morte, transplante card?aco ou a data final estipulada para o estudo. Os pacientes receptores de transplante de cora??o foram retirados do estudo na data em que o ato operat?rio ocorreu. N?o se perdeu contato com os pacientes durante o acompanhamento cl?nico enquanto na fila de espera de transplante card?aco. O modelo de an?lise de riscos proporcionais de Cox foi utilizado para se estabelecer vari?veis de predi??o independentes de mortalidade geral. As vari?veis que eram capazes de predizer mortalidade geral em pacientes com insufici?ncia card?aca cr?nica secund?ria ? cardiomiopatia da doen?a de Chagas ou a outras cardiomiopatias foram utilizadas no modelo univariado. An?lises univariadas foram feitas nos pacientes chag?sicos e n?o chag?sicos separadamente. Resultados: A mediana do tempo de acompanhamento cl?nico foi 32 (15, 121) dias nos pacientes chag?sicos e 79 (14, 151) dias nos pacientes n?o chag?sicos. Nos pacientes chag?sicos, a instabilidade hemodin?mica (p=0,01; raz?o de risco=0,077, intervalo de confian?a de 95% entre 0,01 e 0,58) e o gradiente transpulmonar (p=0,02; raz?o de risco =1,15, intervalo de confian?a de 95% entre 1,02 e 1,30) foram as vari?veis de predi??o independentes de mortalidade geral. Os n?veis s?ricos de s?dio (p=0,002; raz?o de risco =0,81; intervalo de confian?a de 95% entre 0,71 e 0,93) foi a vari?vel de predi??o independente para os pacientes n?o chag?sicos na fila de espera para transplante card?aco. Conclus?es: A instabilidade hemodin?mica e o gradiente transpulmonar foram preditores independentes de mortalidade geral em pacientes com insufici?ncia card?aca cr?nica secund?ria ? cardiomiopatia da doen?a de Chagas na lista de espera para transplante card?aco. Um estudo prospectivo de coorte longitudinal ? necess?rio para validar os resultados obtidos nesta investiga??o.

[pt] Os Ensaios Não-Destrutivos são técnicas utilizadas na inspeção de materiais e equipamentos sem dano a integridade física, podendo ser executados nas etapas de pós-fabricação, montagem, operação e manutenção. São utilizados pela indústria como ferramenta de controle e garantia da qualidade de materiais, produtos e processos. Esta pesquisa teve como objetivo identificar trincas e estimar o seu comprimento em corpos de prova de aço DIN 42CrMo4, adotando diferentes técnicas de END. Os corpos de prova foram de um eixo virabrequim de um motor Diesel, utilizado como unidades geradora em usinas termoelétricas, que falhou por fadiga de baixo ciclo. Os corpos de prova foram projetados de acordo com as normas ASTM E1820, ASTM E1290 e ISO 15653, bem como adotando sugestões de inspetores de ultrassom quanto a uma geometria que otimizasse a eficiência das medições. Subsequente ao trincamento dos corpos de prova em fadiga, os mesmos foram inspecionados com o emprego das técnicas de Radiografia Digital (RD), Ultrassom Phased Array (UT-PA), métodos Pulso-Eco (PE) e Time os Flight Diffraction (TfOD), e Tomografia Computadorizada (TC). Em sequência, os corpos de prova foram fraturados com o uso de uma carga em flexão e nitrogênio liquido, o que permitiu a medição real dos comprimentos das trincas. As técnicas de UTPA, RD e TC indicaram tamanhos de trincas próximos entre si. Nos métodos de PE e ToFD, de UT-PA, se observou que a indicação das trincas sofre influência da distancia entre o cabeçote do equipamento e a posição da trinca. Em trincas menores, essa influência é mais acentuada, levando a uma medição errônea do comprimento da trinca. A técnica de RD apresentou um bom resultado da medida lateral da trinca, porem foi necessário um processamento digital da imagem para obtenção de melhores resultados, explicado devido a sensibilidade do ensaio à geometria do corpo de prova e da densidade do material utilizado. Já a TC apresentou imagens melhores e medidas de trinca similares se comparada com a RD, o que pode influenciar em outros tipos/dimensões de defeitos. Em compensação, sofre mais influencia do volume de material presente no entorno do defeito, o que impossibilitou a detecção de defeitos em algumas amostras. / [en] Non-Destructive Testing (NDT) are techniques applied to materials and equipment inspection without integrity degradation and can be performed in the post-manufacture, assembly, operation and maintenance stages. Those techniques are used by the industry to control a guarantee the quality of materials, products and processes. The aim of this research was to identify cracks and estimate their length on DIN 42CrMo4 steel specimens adopting different NDT techniques. The material to the test specimen was taken from a crankshaft of a Diesel engine, utilized as generating units in thermoelectric power plant, which has failed due to low cycle fatigue. The test specimen were designed according to ASTM E1820, ASTM E1290 and ISO 15653 standards, as well as suggestions made by NDT inspectors for a geometry that optimized the efficiency of the measurement. Subsequent to the fatigue induced crack on the specimens, they were inspected using, Digital Radiography (RD), Phased Array Ultrasound (UT-PA), Pulse-Echo (PE) and Time of Flight Diffraction (ToFD) methods, and Digital Tomography (TC). The specimens were then fractured with a bending load and liquid nitrogen, which allowed the measurement of the real crack length. UT-PA, PE and ToFD, and RD techniques showed a closer estimate crack size among then. It was observed that both UT-PA methods, PE and ToFD, presented an influence of the distance between the equipment head and the crack position, on the crack measurement. This was more pronounced in smaller cracks, leading to an erroneous measurement of the crack length. The RD technique presented a good a good result of lateral crack length, but a good digital image processing was necessary to obtain these results, explained by the sensitivity of the test to the geometry to the geometry of the specimen and to the material density. TC presented better imaging and similar lateral crack lengths when compared to RD, which could influence the detection of other types/dimensions of defects. In contrast, it is more influenced by the volume of material around the object, which made it impossible to detect defects in some samples.

[pt] As vigas de aço com abertura de alma propiciam a passagem de tubulações de serviços e instalações, fornecendo boa capacidade de resistência aos esforços cortantes. As vigas casteladas e as vigas celulares apresentam limitações de dimensões da abertura que estão ligadas à altura do perfil base, geralmente produzindo-se um perfil 50 por cento maior em altura. Neste trabalho, são estudadas as vigas casteladas expandidas em altura, através da soldagem de uma chapa complementar de alma, de modo a se obter vigas de maior inércia, mais leves e com maiores aberturas. O presente trabalho aborda os resultados obtidos através da modelagem por elementos finitos com análise não linear física e geométrica, com vistas a investigação do comportamento estrutural quanto a resistência da seção transversal e rigidez visando-se a redução de custos de fabricação pela eliminação da soldagem de chapas de reforço na região da abertura. / [en] Steel beams with web openings allow integration between structural and tubular services and facilities, providing good shear strength capacity. The castellated and cellular beams have limitations due to this web openings and results in a profile with 50 per cent higher depth. In this research, depth expanded castellated beams with the use of an additional welded plate will be studied, aiming the obtainment of higher inertia as well as wider openings on lighter beams. The present work addresses the results results obtained through the finite element method using physical and geometric nonlinear analysis, observing the structural behavior regarding stiffness and strength in order to reduce manufacturing costs since the reinforcement plates in the areas will no more be required.

[pt] As previsões de falhas por fadiga tornam-se críticas sob condições multiaxiais de carregamento, uma vez que a complexidade do estado de tensões requer uma análise com base em modelos mais avançados do que aqueles adotados para carregamentos uniaxiais. Neste contexto, alguns critérios foram desenvolvidos para tentar prever a falha de componentes submetidos a carregamentos cíclicos multiaxiais. Um exemplo de componente mecânico sujeito a fadiga multiaxial quando em serviço são os eixos virabrequim de unidades geradoras de plantas termoelétricas. Usualmente projetados segundo o critério de Findley, falhas recentes de eixos virabrequim ocorridas no parque termoelétrico nacional, tem provocado o questionamento da eficiência de tal critério. Desta forma, o presente estudo teve como objetivo analisar a aplicabilidade de seis modelos (Papadopoulos, Findley, Matake, McDiarmid, Carpinteri e Spagnoli e Liu e Mahadevan) de fadiga multiaxial de alto ciclo, baseados no plano crítico, para prever a falha de eixos virabrequim de unidades geradoras. Para aplicação dos modelos, adotaram-se diferentes parâmetros de carregamento (tensão média e defasagem de aplicação das tensões normais e cisalhantes) e propriedades de material (resistência à fadiga), partindo de valores encontrados na literatura. Observou-se uma divergência entre os resultados de cada modelo estudado, sendo uns mais conservadores do que os outros, em função do carregamento e propriedades do material. Entretanto, o modelo de Papadopoulos, com base na facilidade de aplicação e sensibilidade de previsão da falha, mostrou-se mais adequado. / [en] The fatigue failure prediction becomes critical under multiaxial loading conditions, since the complexity of the stress state requires analysis based on models more advanced than those adopted for uniaxial loads. In this context, some criteria were developed in an attempt to predict the failure of components subjected to multiaxial cyclic loads. An example of mechanical component subject to multiaxial fatigue is the crankshaft of thermoelectric plants. It is usually projected according to the Findley criterion. However, recent failures of crankshafts occurring in the national thermoelectric park, put on doubt the efficiency of such criterion. In this way, the present study aimed to analyze the applicability of six models (Papadopoulos, Findley, Matake, McDiarmid, Carpinteri e Spagnoli and Liu e Mahadevan) of high cycle multiaxial fatigue, based on the critical plane, to predict failure on crankshaft of generating units. For the models application, different loading parameters (medium stress and out-of-phase stresses) and material properties (fatigue limit) were adopted, starting from values found in the literature. A divergence between the results of each model was observed, some of them were more conservative than the others, depending on the loading and material s properties. However, the Papadopoulos model, based on easy application and failure prediction sensitivity, proved to be more suitable than the others.

The research presented here supports the ongoing need for automatic heart volume calculation through the identification of the left and right ventricles in MRI images. The need for automated heart volume calculation stems from the amount of time it takes to manually processes MRI images and required esoteric skill set. There are several methods for region detection such as Deep Neural Networks, Support Vector Machines and Ant Colony Optimization. In this research Ant Colony Optimization (ACO) will be the method of choice due to its efficiency and flexibility. There are many types of ACO algorithms using a variety of heuristics that provide advantages in different environments and knowledge domains. All ACO algorithms share a foundational attribute, a heuristic that acts in conjunction with pheromones. These heuristics can work in various ways, such as dictating dispersion or the interpretation of pheromones. In this research a novel heuristic to disperse and act on pheromone is presented. Further, ants are applied to more general problem than the normal objective of finding edges, highly qualified region detection. The reliable application of heuristic oriented algorithms is difficult in a diverse environment. Although the problem space here is limited to MRI images of the heart, there are significant difference among them: the topology of the heart is different by patient, the angle of the scans changes and the location of the heart is not known. A thorough experiment is conducted to support algorithm efficacy using randomized sampling with human subjects. It will be shown during the analysis the algorithm has both prediction power and robustness.

Scheduling concerns the allocation of limited resources to tasks over time. In manufacturing systems, scheduling is nothing but assigning the jobs to the available processors over a period of time. Our research focuses on scheduling in systems of parallel processors which is challenging both from the theoretical and practical perspectives. The system of parallel processors is a common occurrence in different types of modern manufacturing systems such as job shop, batch shop and mass production. A variety of important and challenging problems with realistic settings in a system of parallel processors are considered. We consider two types of processors comprising discrete and batch processors. The processor which produces one job at a time is called a discrete processor. Batch processor is a processor that can produce several jobs simultaneously by keeping jobs in a batch form which is commonly seen in semiconductor manufacturing, heat treatment operations and also in chemical processing industries. Our aim is to develop efficient solution methodologies (heuristics/metaheuristics) for three different problems in the thesis. The first two problems consider the objective of minimizing total weighted tardiness in cases of discrete and batch processors where customer delivery time performance is critical. The third problem deals with the objective of minimizing the total weighted completion time in the case of batch processors to reduce work-in-process inventory. Specifically, the first problem deals with the scheduling of parallel identical discrete processors to minimize total weighted tardiness. We develop a metaheuristic based on Ant Colony Optimization(ACO) approach to solve the problem and compare it with the available best heuristics in the literature such as apparent tardiness cost and modified due date rules. An extensive experimentation is conducted to evaluate the performance of the ACO approach on different problem sizes with varied tardiness factors. Our experimentation shows that the proposed ant conony optimization algorithm yields promising results as compared to the best of the available heuristics. The second problem concerns with the scheduling of jobs to parallel identical batch processors for minimizing the total weighted tardiness. It is assumed that the jobs are incompatible in respect of job families indicating that jobs from different families cannot be processed together. We decompose the problem into two stages including batch formation and batch scheduling as in the literature. Ant colony optimization based heuristics are developed in which ACO is used to solve the batch scheduling problem. Our computational experimentation shows that the proposed five ACO based heuristics perform better than the available best traditional dispatching rule called ATC-BATC rule. The third scheduling problem is to minimize the total weighted completion time in a system of parallel identical batch processors. In the real world manufacturing system, jobs to be scheduled come in lots with different job volumes(i.e number of jobs) and priorities. The real settings of lots and high batch capacity are considered in this problem. This scheduling problem is formulated as a mixed integer non-linear program. We develop a solution framework based on the decomposition approach for this problem. Two heuristics are proposed based on the proposed decomposition approach and the performance of these heuristics is evaluated in the cases of two and three batch processors by comparing with the solution of LINGO solver.

[pt] Estudos recentes têm mostrado que para muitos materiais de Engenharia não existe um limite de fadiga. Numa análise convencional, se admite uma vida infinita em fadiga de alto ciclo quando o material atinge 10(6) - 10(7) ciclos sem a incidência de falhas. Entretanto, em função do desenvolvimento tecnológico atual, a vida-fadiga de inúmeros componentes mecânicos e estruturais pode ultrapassar a fronteira da fadiga de alto ciclo, fazendo com que a avaliação do comportamento em fadiga de altíssimo ciclo (10(7) – 10(12)), tenha se tornado extremamente importante para projetos, por ter estabelecido que uma tensão limite de fadiga não existe em muitos casos. Pesquisas recentes demonstram que a maioria dos materiais, incluindo ligas ferrosas, apresentam falhas em até 10(10) ciclos, com um decréscimo contínuo do limite de fadiga após 10(6) ciclos, o que torna a resistência à fadiga associada com um número de ciclos mais importante do que o próprio limite de fadiga. No regime de altíssimo ciclo de fadiga as trincas se iniciam a partir de defeitos internos do material, como inclusões, gerando mecanismos de iniciação de trincas caraterísticos das superfícies de fratura, tais como olho de peixe (fish-eye), ODA (Optically Dark Area) e FGA (Fine Granular Area). Neste estudo foram usinados corpos de prova do aço DIN 34CrNiMo6, que foram ensaiados entre 10(6) e 10(9) ciclos, sob fadiga ultrassônica do tipo tração-compressão, com frequência de 20 kHz e razão de carregamento -1. Os resultados mostraram que o material tem uma tendência maior a vida-fadiga sob valores baixos de tensão e na superfície de fratura de alguns corpos de prova formação de fish-eye. / [en] Recent studies have shown that for many engineering materials there is no fatigue limit. In a conventional analysis, infinite life in high cycle fatigue is allowed when the material reaches 10(6) - 10(7) cycles without the occurrence of failures. However, due to the current technological development, the fatigue life of several mechanical and structural components can exceed the boundary of high cycle fatigue, making the evaluation of ultra high cyle fatigue behavior (10(7)-10(12)), or fatigue of very high cycle, has become extremely important for projects, because it has established that a stress of fatigue limit does not exist in many cases. Current research has shown that most materials, including ferrous alloys, exhibit failures in up to 10(9) cycles, with a continuous decrease in the fatigue limit after 10(6) cycles, which makes strength fatigue associated with a number of cycles most important than own fatigue limit. In the very high fatigue cycles regime, cracks start from internal defects of the material, such as inclusions, generating a feature mechanisms of crack initiation on the fracture surfaces, such as fish-eye, ODA (Optically Dark Area) and Fine Granular Area (FGA). This study, specimens of DIN 34CrNiMo6 steel were machined and were tested between 10(6) and 10(9) cycles, under ultrasonic fatigue, with a frequency of 20 kHz and a loading ratio of -1. The results showed that the material has a tendency to fatigue life under low stress values and in some fracture sufaces of the specimens the fish-eye formation.

First part of the thesis is about literature research of optimization algorithms. Three of the algorithms were implemented and tested, concretely the ant colony algorithm, tabu search and simulated annealing. All three algorithms were implemented to solve the traveling salesman problem. In second part of the thesis the algorithms were tested and compared. In last part the influence of the ant colony parameters was evaluated.

Thesis deals with discrete optimization problems. It focusses on faster ways to find good solutions by means of heuristics and parallel processing. Based on ant colony optimization (ACO) algorithm coupled with k-optimization local search approach, it aims at massively parallel computing on graphics processors provided by Nvidia CUDA platform. Well-known travelling salesman problem (TSP) is used as a case study. Solution is based on dividing task into subproblems using tour-based partitioning, parallel processing of distinct parts and their consecutive recombination. Provided parallel code can perform computation more than seventeen times faster than the sequential version.