|
Dissertations |
|
1
|
-
FÁBIO SILVA FARIA
-
Estimation of Heat Input in Welding Process by Capacitive Discharge Using Problem Technique inverse
-
Advisor : SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
COMMITTEE MEMBERS :
-
ANDRÉ GUIMARAES FERREIRA
-
BRUNO DE CAMPOS SALLES ANSELMO
-
EDMILSON OTONI CORREA
-
SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
Data: Feb 16, 2022
-
-
Show Abstract
-
This work presents an experimental and numerical procedure for estimating the thermal input in the capacitive discharge welding process of K-type thermocouples wire. The objective of the work is to promote improvements in a device at the important temperature, the thermocouple. The work of proposed improvement in the solution of this in metals and also of possible energy solution of risk solution. The thermal model used is based on the equation of transient three-dimensional heat diffusion with phase change, modeled as a function of enthalpy. The thermal properties of chromel and alumel, materials that make up the two wires of k-type thermocouples, were considered temperature-dependent. The model was solved using the program COMSOL Multiphysics. The equation of the power curve in relation to time was defined from theoretical concepts of capacitor discharge. The nonlinear inverse problem technique used to solve the heat conduction problem was the iterative Function Specification Method. This technique estimates the heat rate that minimizes an objective function, defined as the square of the experimental and numerical temperature difference for each time interval. A numerical code was used in MATLAB together with software COMSOL Multiphysics to applied the inverse problem technique. The experimental method was used to obtain temperatures close to the welding region, which were used to apply the inverse problem technique. To calculate the efficiency of the process, the integral of the power estimated by time was used, and the energy stored in the capacitors bank of the equipment, defined by concepts of electricity. The proposed methodology was able for estimating the thermal input and efficiency for the capacitive discharge welding process.The heat rate was estimated from experimental temperature data, the temperatures calculated computationally, from the estimated heat rate, showed low divergence in relation to the experimental data. In addition, the development of a capacitive discharge welding equipment is presented. The equipment consists of a capacitors bank that stores the energy supplied by a switched source. The amount of energy stored in capacitors bank is controlled by an Arduino microcontroller. The equipment interface has a liquid crystal display and a button, which allow the operator to select the type of thermocouple he wants to weld. This equipment was able to weld thermocouples wire thicker than 24 AWG.
|
|
2
|
-
RONNY FRANCIS RIBEIRO JUNIOR
-
Machine Learning-Based Fault Detection and Diagnosis in Electric Motors
-
Advisor : GUILHERME FERREIRA GOMES
-
COMMITTEE MEMBERS :
-
GUILHERME FERREIRA GOMES
-
ANTONIO CARLOS ANCELOTTI JUNIOR
-
MARCUS VINICIUS GIRÃO DE MORAIS
-
Data: Feb 16, 2022
-
-
Show Abstract
-
Fault diagnosis is critical to any maintenance industry, as early fault detection can prevent catastrophic failures as well as a waste of time and money. In view of these objectives, vibration analysis in the frequency domain is a mature technique. Although well established, traditional methods involve a high cost of time and people to identify failures, causing machine learning methods to grow in recent years. The Machine learning (ML) methods can be divided into two large learning groups: supervised and unsupervised, with the main difference between them being whether the dataset is labeled or not. This study presents a total of four different methods for fault detection and diagnosis. The frequency analysis of the vibration signal was the first approach employed. This analysis was chosen to validate the future results of the ML methods. The Gaussian Mixture model (GMM) was employed for the unsupervised technique. A GMM is a probabilistic model in which all data points are assumed to be generated by a finite number of Gaussian distributions with unknown parameters. For supervised learning, the Convolution neural network (CNN) was used. CNNs are feedforward networks that were inspired by biological pattern recognition processes. All methods were tested through a series of experiments with real electric motors. Results showed that all methods can detect and classify the motors in several induced operation conditions: healthy, unbalanced, mechanical looseness, misalignment, bent shaft, broken bar, and bearing fault condition. Although all approaches are able to identify the fault, each technique has benefits and limitations that make them better for certain types of applications, therefore, a comparison is also made between the methods.
|
|
3
|
-
RONNY FRANCIS RIBEIRO JUNIOR
-
Machine Learning-Based Fault Detection and Diagnosis in Electric Motors
-
Advisor : GUILHERME FERREIRA GOMES
-
COMMITTEE MEMBERS :
-
ANTONIO CARLOS ANCELOTTI JUNIOR
-
GUILHERME FERREIRA GOMES
-
MARCUS VINICIUS GIRÃO DE MORAIS
-
Data: Feb 16, 2022
-
-
Show Abstract
-
Fault diagnosis is critical to any maintenance industry, as early fault detection can prevent catastrophic failures as well as a waste of time and money. In view of these objectives, vibration analysis in the frequency domain is a mature technique. Although well established, traditional methods involve a high cost of time and people to identify failures, causing machine learning methods to grow in recent years. The Machine learning (ML) methods can be divided into two large learning groups: supervised and unsupervised, with the main difference between them being whether the dataset is labeled or not. This study presents a total of four different methods for fault detection and diagnosis. The frequency analysis of the vibration signal was the first approach employed. This analysis was chosen to validate the future results of the ML methods. The Gaussian Mixture model (GMM) was employed for the unsupervised technique. A GMM is a probabilistic model in which all data points are assumed to be generated by a finite number of Gaussian distributions with unknown parameters. For supervised learning, the Convolution neural network (CNN) was used. CNNs are feedforward networks that were inspired by biological pattern recognition processes. All methods were tested through a series of experiments with real electric motors. Results showed that all methods can detect and classify the motors in several induced operation conditions: healthy, unbalanced, mechanical looseness, misalignment, bent shaft, broken bar, and bearing fault condition. Although all approaches are able to identify the fault, each technique has benefits and limitations that make them better for certain types of applications, therefore, a comparison is also made between the methods.
|
|
4
|
-
BRUNO DA SILVA BUNYA
-
Thermodynamic Analysis of BECCS Systems in the Sugar-Energy Sector
-
Advisor : CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
COMMITTEE MEMBERS :
-
CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
CÉSAR ADOLFO RODRÍGUEZ SOTOMONTE
-
JOSE ANTONIO PERRELLA BALESTIERI
-
JUAN JOSE GARCIA PABON
-
THIAGO GOTELIP CORREA VELOSO
-
Data: Feb 24, 2022
-
-
Show Abstract
-
In the 2015 Paris Agreement, Brazil pledged to reduce its GHG (greenhouse gas) emissions by 37% by 2050 and by 43% by 2030 compared to 2005, reinforcing its targets for reduction to zero emissions by 2050. Bioenergy systems with carbon capture and storage (BECCS) will be essential for countries to be able to meet the gas emission reduction targets established in the Paris Agreement. Brazil is considered the world's largest sugarcane producer with a 39% share. For BECCS systems, the carbon dioxide (CO2) capture method by chemical absorption is the most promising approach among the commercially available technologies in the short and medium term. This work aims to carry out a thermodynamic analysis of different configurations of a BECCS system with a chemical absorption post-combustion capture method (Monoethanolamine-MEA) from the burning of sugarcane bagasse and straw. Commercial software Aspen Plus®, Aspen Hysys® and Matlab were used to simulate different configurations of the BECCS system, covering biomass combustion, CCS (Carbon Capture System) and seven different configurations of the Rankine cycle with cogeneration. Three scenarios of technological maturity of the alcohol and sugar production process were considered in the analysis. Stochastic method and parametric analysis were applied having as main performance indicators evaluated the generation of net electric power, CO2 capture and specific emission per unit of energy generated annually. The analyzes evidenced a tradeoff between net electric power generation and carbon capture for all evaluated scenarios and configurations. Scenarios where there is greater availability of heat as a result of the alcohol and sugar production process being more efficient have been shown to favor carbon capture. Extracting steam from the turbines to supply the thermal and electrical demand of the CCS together with the process implied a high net electrical power penalty generated by the plant
|
|
5
|
-
LUIZ FILIPE ALMEIDA ROQUE
-
MODELING AND ANALYSIS OF THE LIFE CYCLE OF PRODUCTION AND MARKETING OF GREEN DIESEL (HVO) AND ITS APPLICATION IN DUAL-FUEL MODE WITH ETHANOL IN COMPRESSION IGNITION ENGINES
-
Advisor : CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
COMMITTEE MEMBERS :
-
MONA LISA MOURA DE OLIVEIRA
-
CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
JUAN JOSE GARCIA PABON
-
SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
Data: Feb 25, 2022
-
-
Show Abstract
-
The use of clean and renewable sources is a practice that is gaining more and more space and visibility in the energy generation sector. Resources are spent on a daily basis in search of new technologies and new fuels that meet the requirements of reducing pollutant emissions, emission of carbon dioxide and a sustainable development policy. Among the solutions found, the use of biomass as energy sources and improved technologies for combustion, such as the Dual-Fuel technique, are options that promote decarbonization and a reduction in polluting gas emissions in general. In this context, the present work aimed at the experimental study of HVO as a renewable fuel to be applied in a single-cylinder internal combustion engine, neat or in Dual-Fuel combustion with hydrated ethanol, in a dynamometric bench, where performance parameters and emission of polluting gases were evaluated. With the experimental data added to inventory data collected from the literature on the production cycle of HVO from palm oil and soybean oil, a Life Cycle Assessment of the biofuel was carried out. The adopted methodology consisted of developing the test bench with all instrumentation, data analysis methods and calculation of measurement uncertainties, test planning aimed at the application of motor generators, and analysis of the parameters of interest. Tests were carried out using neat HVO compared to neat conventional diesel and tests using dual combustion of Diesel-Ethanol and HVO. For Life Cycle Assessment, the SimaPro 9 software and the ReCiPe 2016 Midpoint (E) calculation method were used. The results obtained with pure HVO showed reductions in specific levels of emissions of NOx, HC, CO and particulate matter (PM) of 30%, 75%, 81% and 55.3%, respectively. The reduction is mainly due to the physicochemical properties of the biofuel. The Dual-Fuel application provides a significant reduction in CO2, NOX and MP emissions. The life cycle analysis showed that the HVO produced from palm oil has an environmental performance superior to that of soy, with a reduction in the Global Warming Potential of up to 75% and impact categories such as Terrestrial Acidification, Ozone Formation and Consumption of Non-Resources renewables present lower levels when compared to diesel. The environmental impact of the 'Dual-Fuel' operation shows a decrease in the levels of Global Warming Potential, Depletion of Fossil Resources and Ozone Formation.
|
|
6
|
-
Alexandre Ribeiro Souza Carvalho
-
Quasi-three-dimensional inverse methodology for the design of hydraulic flow machines
-
Advisor : NELSON MANZANARES FILHO
-
COMMITTEE MEMBERS :
-
ANDRE LUIZ AMARANTE MESQUITA
-
NELSON MANZANARES FILHO
-
RAMIRO GUSTAVO RAMIREZ CAMACHO
-
Data: Mar 3, 2022
-
-
Show Abstract
-
Inverse methodologies continue to be important computational tools for the design of flow machines, despite the sophisticated methods currently available for the analysis of viscous three-dimensional flows. A Quasi-Three-Dimensional methodology based on the interdependent solution of two inverse problems is described in this work: the through-flow problem and the blade-to-blade problem. The developments continue the research by Santos et al. (2012) and Barbosa (2018). The solution to the through-flow problem – referring to the work of Santos et al. (2012) - provides the shape of the mean flow surface and the results of the streamlines, to meet a given hydraulic momentum distribution. The inverse blade-to-blade problem – referring to the developments of Barbosa (2018) – provides the geometry of the blades, given a finite number of blades, on surfaces of revolution whose generatrices are the streamlines resulting from the through-flow problem. Until then, the effects of thickness and blocking were disregarded. In this work, a methodology for including thickness distributions along the blades is described. The thickness effect is analyzed through the direct solution of the blade-to-blade problem (now considering profiles) using the panel method based on linear vortex distributions. The corresponding blocking factors are calculated, and the results feed the input of a new solution of the through-flow problem. Thus, the through-flow problem and blade-to-blade become coupled. The geometry is recalculated, now considering the effect of thickness and blocking. Furthermore, the effects of a) hydraulic momentum distribution; b) finite number of blades; c) thickness; d) blocking; e) combination of b), c) and d); and f) profile stacking line on the geometry of the blades and their hydrodynamic loads are analyzed.
|
|
7
|
-
GUILHERME AUGUSTO VILAS BOAS VASCONCELOS
-
Optimization in Duplex Stainless Steel End Milling UNS S32205
-
Advisor : JOAO ROBERTO FERREIRA
-
COMMITTEE MEMBERS :
-
MANOEL CLEBER DE SAMPAIO ALVES
-
JOAO ROBERTO FERREIRA
-
MIRIAN DE LOURDES NORONHA MOTTA MELO
-
Data: Aug 3, 2022
-
-
Show Abstract
-
O estudo sobre o fresamento de topo do aço inoxidável duplex vem ganhando destaque, devido aos desafios encontrados em sua baixa usinabilidade. Isso ocorre, pois, esses materiais apresentam baixa condutividade térmica, alta tenacidade e alta taxa de encruamento. Entretanto, para garantir qualidade final em produtos manufaturados, é importante que o processo de fresamento seja bem planejado, visando menores desgastes nas fresas durante o processo e ao mesmo tempo apresentando bons indicadores de produtividade. Assim, este trabalho tem como objetivo realizar uma otimização robusta multiobjetivo no processo de fresamento de topo do aço inoxidável duplex UNS S32205. Foram realizados experimentos seguindo um planejamento composto central combinando as variáveis de controle: velocidade de corte, avanço por dente, profundidade de corte e largura fresada e as variáveis de ruído: desgaste de flanco (vb), vazão de fluido (Q) e o balanço da ferramenta (lt0). As variáveis de respostas avaliadas foram a rugosidade da peça usinada e a taxa de remoção de material. Foram aplicadas as metodologias de superfície de resposta, de projeto de parâmetro robusto, do erro quadrático médio e da interseção normal à fronteira. Em seguida, foram analisados e discutidos os efeitos das variáveis de controle e de ruído, bem como as interações destas sobre as características de interesse. O desgaste da ferramenta foi a variável que mais influenciou a rugosidade Ra. A taxa de remoção de material foi influenciada por todas as variáveis controláveis do processo. Os valores obtidos para a rugosidade Ra variaram entre 0,24 a 1,10 µm e a taxa de remoção de material variou entre 40,39 a 187,52 mm³/s. A otimização da média e da variância de cada característica de interesse foi realizada, bem como a otimização do erro quadrático médio. Assim, 21 soluções Pareto-ótimas foram obtidas, contribuindo para a melhoria da qualidade superficial e da produtividade no processo de fresamento. Nos ensaios de confirmação empregou-se um arranjo ortogonal de Taguchi (L9) onde obteve-se os setups ótimos capazes de mitigar a influência das variáveis de ruído, o que corroborou a boa adequação da metodologia proposta.
|
|
8
|
-
LUCAS ROGÉRIO EMILIANO
-
Optimized tip grinding of martensitic stainless steel piston rings with chromium nitride coating
-
Advisor : JOAO ROBERTO FERREIRA
-
COMMITTEE MEMBERS :
-
JOAO ROBERTO FERREIRA
-
MARCOS VALERIO RIBEIRO
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
Data: Oct 26, 2022
-
-
Show Abstract
-
In the scope within in the automotive industry the machining processes are widely applied. Among them, the grinding process is widely used for finishing parts that require low roughness and dimensional tolerance, with the premise of achieve product features and reducing losses as much as possible that is, assurance the product quality and reduce manufacturing costs. In this context, this work presents a strategical approach to reduce variability in the grinding process of the tips of nitrided martensitic stainless steel (MS066-N1) piston rings, with chromium nitride (CrN) coated applied by the physical vapor deposition process (PVD). In this work, an experimental design (DOE) was created, in the Response Surface methodology configuration with 3 levels and 3 factors. The input variables were the abrasive grain size of CBN grinding wheel, the grinding wheel rotation parameter (RPM) and the feed parameter (GPM). The methodology applied was experimental planning, which support in defining the optimal machine parameterization, aiming to reduce the process variability and reach dimensional features stability in the specification average of the characteristic gap between piston rings tips. The closest optimization parameters and available on the machine was with abrasive grain size of CBN grinding wheel (B46), the grinding wheel rotation speed parameter (5200 RPM) and the feed parameter (9,0 GPM). To validate the optimization parameters, a study of capability analysis was created which demonstrated through the levels of Cp, Cpk and Pp, Ppk, that the process reached the expected stability, where the percentage of parts outside the specified or defective was reduced from 5,79% to 0,00%.
|
|
9
|
-
GUSTAVO VIEIRA FREZ
-
Fluid dynamics simulation in ANSYS – CFD of dual-fuel combustion in compression ignition engine using renewable diesel (HVO) and biogas
-
Advisor : CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
COMMITTEE MEMBERS :
-
ANDRÉS ARMANDO MENDIBURU ZEVALLOS
-
CHRISTIAN JEREMI CORONADO RODRIGUEZ
-
FAGNER LUIS GOULART DIAS
-
JUAN JOSE GARCIA PABON
-
Data: Nov 28, 2022
-
-
Show Abstract
-
Population and urban growth in Brazil in recent decades have increased the national vehicular fleet, which, in general, uses fossil fuels whose burning results in pollutant emissions. Allied with economic factors, there is an urgent concern to replace these fuels with those from renewable sources. Much of the research in this area uses experimental techniques, which are not always feasible. Thus, combining computational analysis with these techniques is of great relevance. In this context, this work analyzed computationally the combustion cycle of a Diesel engine operating with fossil diesel, HVO and HVO with biogas. For this purpose, its 3D combustion chamber geometry was modeled in a CAD software and the CFD simulations were carried on ANSYS/Forte software. A kinetic model with 35 species and 74 chemical reactions, together with the RANS RNG k-ε model, the KH-RT model, a uniform injection profile and experimental values for initial pressure and temperature were considered in the modeling step. In addition, a 0-D analysis in ANSYS/Chemkim-Pro software was performed to estimate some parameters not measured experimentally, and also to obtain two adapted kinetics models to describe the of pure HVO and HVO-biogas blend, which was done together with a mono-objective optimization process by the Lichtenberg algorithm. Both 0-D and 3D approaches were validated with experimental data. From the simulations, in-cylinder pressure and its rise rate, rate of heat release, and its accumulation, turbulence, and emissions characteristics were obtained as a function of crank angle. 3D fields of velocity, temperature, the spray of diesel, and molar fractions of CO, CO2, NO, and NO2 were also obtained. The in-cylinder pressure obtained computationally showed a good agreement with the experimental ones (which were obtained from the R&D project ROTA 2030 by GETEC-UNIFEI at 2021 and 2022), but with a longer ignition delay. It was observed that the in-cylinder pressure is strongly dependent on the duration of injection. It was also verified that the regions closer to the TDC presented the most variations in the analyzed parameters and with the maximum values. Besides, the combustion of HVO and HVO with Biogas reduced the peaks of pressure and heat release, as well as NOx and soot emissions and also presented more significant homogeneity in the velocity and temperature fields.
|
|
10
|
-
PÂMELA BADESSA MAIA
-
Geometric Optimization of Rotor Wells Applied in Oscillating Water Column Plants
-
Advisor : RAMIRO GUSTAVO RAMIREZ CAMACHO
-
COMMITTEE MEMBERS :
-
NELSON MANZANARES FILHO
-
RAMIRO GUSTAVO RAMIREZ CAMACHO
-
WALDIR DE OLIVEIRA
-
Data: Dec 2, 2022
-
-
Show Abstract
-
In a first approach, the validation of the numerical simulation (CFD) of a Wells turbine rotor with 8 blades formed by NACA 0015 airfoil so Constant chord length is carried out. The results obtained for the turbine performance characteristics (torque coefficient, total pressure drop coefficient across the turbine and efficiency) are compared to experimental and nu-merical values available in the literature. In order to carry out the numerical simulation so steady state flow in the Wells turbine rotor, a periodic hydraulic channel and a single computational domain were used. Anunstruc-tured mesh was employed, with prismatic layers close to the rotor blade surface. The k-ε Rea-lizable turbulence model was used withim proved wall treatment. For the pressure-velocity coupling, the SIMPLEC method and second-order discretization and interpolation schemes-were used. In a second approach, anoptimization methodology for maximizing the torque coeffi-cient and the rotor efficiency of the Wells turbine is presented. The optimization processcon-sisted of generating a DOE (Design of Experiments) experiment plan, wheret he profiles chord length and blade thickness were defined as geometric design variables, both paramete-rized and referring to the average height and tip of the blade. After carrying out numerical simulation soft He flow atall points of the DOE, a statistical analysis was applied to the results obtained, verifying that the thickness of the blade does not statistically have a significantin fluence on the aerodynamic performance ofthe Wells turbine rotor. The results obtained provided two optimal geometries for the Wells turbine rotor, the-first geometry (Optimized A) presented a maximum gain of 31.11% for the torque coefficient and 13.59% for maximum efficiency. The second geometry (Optimized B) showed a gain of 20.18% for maximum efficiency and a reduction of 5.44% for maximum torque coefficient.
|
|
11
|
-
ÁLVARO HENRIQUE CALAZANS GUEDES
-
ASSESSMENT OF MICROABRASIVE WEAR RESISTANCE OF WC-Ni-Mo2C, WC-Ni-Al and WC-Ni-Cr3C2 CEMENTED CARBIDES
-
Advisor : EDMILSON OTONI CORREA
-
COMMITTEE MEMBERS :
-
EDMILSON OTONI CORREA
-
GILBERT SILVA
-
NEIDE APARECIDA MARIANO
-
Data: Dec 12, 2022
-
-
Show Abstract
-
The Brazilian industrial sector has suffered great losses due to abrasive wear, since this type of wear is one of the main factors for machinery and equipment breakdown. In order to mitigate this problem, cemented carbides have been increasingly used in applications involving abrasive wear. Conventional carbide is a cobalt and WC material, which has certain disadvantages, one of the main ones being the low corrosion resistance due to the cobalt binder. Thus, research conducted by Penrice (1987), Almond and Roebuck (1988), Tracey (1992), point to nickel as more promising and lower cost. The nickel binder is a less expensive material, has superior performance regarding oxidation and corrosion in acidic and aqueous environments compared to cobalt, but loses in wear resistance by presenting lower hardness. Thus, the combination with auxiliary carbides arises to increase the hardness of the composite. An important factor regarding the manufacture of carbide, which is by powder metallurgy, is the fact, that the composite produced presents a certain level of porosity. This porosity is responsible for performance losses to abrasive wear. Thus a good final product should have a low level of porosity. The evaluation of abrasive wear consisteddistância in subjecting three specimens with distinct nickel binder and auxiliary carbides and one with cobalt binder to a microabrasive test, monitoring the wear evolution, the acting micromechanisms, and later comparing the results between both specimens. Scanning electron microscopy was used to evaluate the worn surface and study the actuating wear mechanisms, where it was possible to observe the action of the two-body and three-body mechanisms. In the wear test, compared to the samples with the conventional carbide 90WC-10Co, sample 90WC-9.5Ni-0.5Al performed better, sample 90WC-8Ni-2Mo2C performed similar wear behavior, and finally sample 90WC-8Ni-2Cr3C2 showed the lowest performance. Given the results it is possible to say that cemented carbides produced with nickel and added carbides to improve performance are efficient since they maintain quality in manufacturing in relation to porosity and thus can be an excellent option to replace conventional cemented carbide.
|
|
12
|
-
MÁRCIO VITAL DE ARRUDA
-
Comparative analysis of residual stress results and dynamic response variation of welded plates using finite element method
-
Advisor : EDMILSON OTONI CORREA
-
COMMITTEE MEMBERS :
-
EDMILSON OTONI CORREA
-
ELISAN DOS SANTOS MAGALHÃES
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
Data: Dec 12, 2022
-
-
Show Abstract
-
Welding processes may cause undesirable residual stresses and their detection is possible using different methods. Some residual stress detection methods or techniques are destructive or semi-destructive, therefore not always applicable. Some methods are non-destructive, such as X-ray diffraction, neutron diffraction, and ultrasound. However, they are complex and high cost. One possible non-destructive method to detect welding residual stresses, lower cost than the current techniques, is being studied by some researchers. This method is based in the phenomenon which a structure natural frequencies vary when welding residual stress is applied. It consists in measuring the structure natural frequencies after welding, and comparing them to natural frequency values considered ideal values. These ideal values are obtained from finite element method simulation. However, the welding process simulation is not trivial; it implies at least four finite element analysis which can be executed in different manners. Different researchers have been developing studies with different types of structures, materials and results. The aim of this study is to analyze which parameters can improve Finite Element Method (FEM) model performance for obtaining simulation results closer to experiments. In this study, five models were developed and validated with experimental studies. It was observed that models using 2D shell elements generate better results than odels using 3D solid elements. In addition, it was observed that the symmetry technique which can be used in plate simulations leads to significant lower computational times, but affect modal results and, in addition, do not generate natural frequency values for all vibrational modes; therefore, the symmetry technique should be avoided in this type of analysis. The birth and death technique, which simulates filler metal deposition, was also analyzed and it was concluded that this technique has small effect in the results. Finally, this work proposes an interpolation technique for the natural frequency values to evaluate the modal results variation due to welding residual stresses.
|
|
|
Thesis |
|
1
|
-
MARCELO LUÍS SIQUEIRA
-
Effect of Chemical Composition on Development of the Martensitic Structure in Cast Iron Nodular In The as Casted State
-
Advisor : MIRIAN DE LOURDES NORONHA MOTTA MELO
-
COMMITTEE MEMBERS :
-
ANTONIO AUGUSTO ARAUJO PINTO DA SILVA
-
JOAO ROBERTO MORO
-
JOSE CARLOS DE LACERDA
-
MIRIAN DE LOURDES NORONHA MOTTA MELO
-
NEIDE APARECIDA MARIANO
-
PAULO JUNHO DE OLIVEIRA
-
Data: Feb 23, 2022
-
-
Show Abstract
-
Nodular cast iron is one of the most used ferrous materials by engineering currently, whether as components for the automotive industry, for the basic sanitation industry or for the component industries in general. One of the main raw materials for the production of cast iron is steel scrap. Currently, this raw material is collected from different generating sources and is composed of different types of steel with different chemical compositions. Many chemical elements present in steel scrap, even in small percentage concentrations, are undesirable in the production of cast iron, since some of them have the effect of causing significant changes in their microstructure. The contribution of this study is to investigate the effect of the elements copper, nickel, molybdenum and chromium, commonly present in steel scrap, as a promoter of the martensitic matrix in a nodular cast iron in the as casted state. Contents lower than 1%w were used for each element investigated and the pieces were produced by centrifugation casting process. By analyzing the scanning electron microscopy images and semiquantitative analysis via Energy Dispersive Spectroscopy, it was observed that nickel and copper were uniformly distributed in the matrix, while chromium and molybdenum formed carbides in the intercellular contours. Among the four elements investigated, molybdenum proved to be a strong promoter of martensite formation and the fractions of this phase increased with the increase in the amount added of this element (35%, 40% and 50%). The results obtained indicate the probable path to obtain the totally martensitic matrix in as casted state and the possibility of eliminating costs related to production on an industrial scale, such as the austenitization heat treatment followed by quenching, which has an estimated operating cost of 26173.35 USD per month. It is noteworthy that the analysis of the feasibility of eliminating the heat treatment should consider the cost with the consumption of the ferro-molybdenum alloy and the particularities of the process of each foundry.
|
|
2
|
-
RODRIGO GUSTAVO DOURADO DA SILVA
-
Thermal efficiency estimation of TIG and laser welding processes using modified Function Specification Method with moving temperature sensor
-
Advisor : SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
COMMITTEE MEMBERS :
-
FRANCIS HENRIQUE RAMOS FRANÇA
-
GUILHERME BORGES RIBEIRO
-
ELISAN DOS SANTOS MAGALHÃES
-
JOSE CARLOS ESCOBAR PALACIO
-
ROGERIO FERNANDES BRITO
-
SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
Data: Mar 15, 2022
-
-
Show Abstract
-
The determination of the imposed heat rate in welding processes has always been an obstacle to the modeling of this problem. Thus, the use of inverse problem techniques is an alternative to estimate the unknown heat rate. This work presents a numerical-experimental methodology for estimating the thermal input in welding processes. A modification of the iterative Function Specification Method is used to consider the thermal sensitivity of the temperature moving sensor as a function of time and the position in relation to the weld bead. In this way, it is possible to solve highly nonlinear problems with a large gradient of temperature near the measurement sensors. Once the heat rate was estimated, the thermal efficiency of the process as a function of the welding power was determined. In this work, thermal efficiencies for TIG welding experiments on AISI 304 steel and laser welding on AISI 1020 steel were estimated. A detailed analysis of thermal efficiency as a function of welding parameters was also carried out using the Taguchi method for the TIG welding case. In the case of laser welding, a multiphysical model to estimate thermal efficiency was proposed, in which the movement of the molten metal in the weld pool is considered. A computer program using the feature Livelink for Matlab was developed to solve the proposed inverse problem and it is explained in this work.
|
|
3
|
-
MAICON QUEIROZ HILÁRIO
-
Comparative Evaluation of Thermal Power Systems in the Context of the Energy Transition
-
Advisor : ROGERIO JOSE DA SILVA
-
COMMITTEE MEMBERS :
-
CARLOS BARREIRA MARTINEZ
-
EDUARDO JOSÉ CIDADE CAVALCANTI
-
GENESIO JOSE MENON
-
JOSE ANTONIO PERRELLA BALESTIERI
-
ROGERIO JOSE DA SILVA
-
Data: Apr 19, 2022
-
-
Show Abstract
-
In several regions of the world and diferente countries there is a transition of the electricity matrix in progress, driven mainly by environmental issues. However, there are geopolitical aspects involved, suah as the interests of countries with a strong presence of coal in their energy matrix. In this context, natural gas appears as a fuel for this energy transition. This perspective follows the increase in investments in infrastructure of the natural gas market and the fact that, among fossil fuels, natural gas is the fuel with the lowest greenhouse gas emissions. In addition, natural gas thermal power plants are interesting for the complementarity and reliability of the system and energy security. In addition, natural gas power plants are interestinh fot the complementarity and reliability of the system and energy security. Therefore, the presente work aimed to carry out a comparative analysis of four termal power Generation systems, being three coal-fired power plants (subcritical, supercritical and ultrasupercritical) and a combined cycle plant with natural gas, from the exergetic point of view, emissions of CO2 and annualized eletricity cost. Based on this comparison, a discussion was held about under what conditions the use of natural gas can have its use expanded in the contexto of the energy transition. It has been demonstrated that a combined cycle plant powered by natural gas with 57.95 % exergetic efficiency has an emission factor 60 % lower than that of a subcritical coal plant and avoids about 3.5 Mt of annual CO2 emissions. An economic analysis was also carried out and it was determined that the volatility of fuel prices has a strong influence on the cost of electricity. According to market conditions, supply and demand, the combined cycle plants presented competitive costs.
|
|
4
|
-
DIEGO CORRÊA FERREIRA
-
Non Linear Inverse Problem Analysis in Heat Transfer for the Estudy of Coated and Uncoated Carbide Tool
-
Advisor : SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
COMMITTEE MEMBERS :
-
GUILHERME FERREIRA GOMES
-
JOAO ROBERTO FERREIRA
-
SANDRO METREVELLE MARCONDES DE LIMA E SILVA
-
RUBENILDO VIEIRA ANDRADE
-
ALISSON FLÁVIO BARBIERI
-
LEANDRO ALCOFORADO SPHAIER
-
Data: Apr 19, 2022
-
-
Show Abstract
-
This work presents the thermal influence of the coating in a carbide cutting tool. The influence of the convection is also included in the analysis. The thermal model consist of a carbide cutting tool, a shim and a tool holder. This model is represented by the threedimensional transient thermal diffusion equation which is solved by the FEM (Finite Element Method) through the software COMSOL Multiphysics® 5.6. The most important advantages of using this software is the easy way to model the thermal coating, whose thickness is in the order of (𝜇m). For the coating, a cutting tool coated by Titanium Nitride (TiN), Aluminium Oxide (Al2O3) and Titanium Carbo-Nitride (TiCN) was chosen. This model is subjected to the boundary conditions of convection and radiation in the regions exposed to the environment. Boundary condition of prescribed heat flux in the contact area between the cutting tool and the workpiece. The thermal properties adopted in this work are temperature dependent. This is possible due the non-linear Function Specification technique used to solve the inverse problem. In order to obtain experimental data, 13 turning experiments were performed in nodular cast iron sample. Eight experiments were done for the uncoated tool, 7 dry experiments and 1 with pressure air cooling. For the coated tool, 5 experiments were performed, 4 dry experiments and 1 with pressure air cooling. In all experiments termocouples temperatures, thermal camera temperatures and dynamometer forces data were aquired. An exploratory data analysis of the experimental data was done. The heat flux is estimated through a MATLAB® script together with COMSOL®, using the tool MATLAB livelinkTM. Once the heat flux is obtained, COMSOL® is used again to generate the temperature thermal gradient for the whole model. Numerical results such as: Maximum temperature in the contact area and temperature field for the coated and uncoated tool were obtained. A comparaison between the numerical and experimental results for the thermocouples were also performed. The results showed an increase in the heat flux and in the maximum temperature in the contact area for the coated tool in comparison with the uncoated tool.
|
|
5
|
-
FERNANDO BRUNO DOVICHI FILHO
-
Determination of the technical and economic potentials of electricity generation from biomass: case study of the State of Minas Gerais.
-
Advisor : ELECTO EDUARDO SILVA LORA
-
COMMITTEE MEMBERS :
-
DIEGO MAURICIO YEPES MAYA
-
ELECTO EDUARDO SILVA LORA
-
FLAVIO DIAS MAYER
-
JOSE CARLOS ESCOBAR PALACIO
-
MARCELO RISSO ERRERA
-
OSVALDO JOSE VENTURINI
-
QUELBIS ROMAN QUINTERO
-
Data: May 2, 2022
-
-
Show Abstract
-
The present investigation aims to determine the technical and economic potentials of electricity generation from different types of biomasses in the state of Minas Gerais, using different electricity generation technologies, supported by different thermochemical combustion and gasification routes. The five most available biomasses in the state of Minas Gerais were selected: corn, soybeans, coffee, eucalyptus and sugar cane, whose crops produce residues that can be used for electricity generation. Considering the technical, logistical and economic data of the selected biomasses, the research sought to define the most viable types of residual biomass for electricity generation, in different locations within the state, through a multi-criteria decision-making approach (MCDM). A bibliographic review of the maturity of different types of electricity generation technologies from biomass was carried out, based on available information, schemes and operational parameters of real installations. The level of technological readiness of the different alternatives was also evaluated through a survey with specialists in the sector, concluding that the technologies with the highest level of readiness are the conventional Rankine cycle, the organic Rankine cycle and the use of gasifiers integrated with internal combustion. These three technological alternatives were adopted as options for the generation of electricity from biomass, obtaining a scale of the types of biomasses available, through a multicriteria methodology. The determination of the criteria weight was carried out from an integration with the analytical hierarchy process (AHP). In conjunction with a GIS-MCDA approach, different micro-regions were evaluated considering their respective biomass resources, allowing the estimation of costs and performance of different electricity production technologies. A logical algorithm was developed for the selection of alternatives for energy production from biomass, giving priority to technical feasibility, and considering factors such as the availability of biomass and its characteristics such as calorific value, humidity and granulometry. Investment costs, operating costs, as well as logistical costs for generating electricity from the plant, according to the technology used, were surveyed. Typical energy efficiencies for these cycles are between 8 and 30%, with conversion efficiency increasing with generation scale. To be analyzed, the generation systems were divided according to their power ranges in industrial application through power curves, in which plants with power above 5 MWe were to use the conventional Rankine cycle, power between 0.5 and 4.1 MWe using organic Rankine cycles, and the lower capacity potentials adopting the gasification and syngas burning cycles in internal combustion engines. It was found that eucalyptus prevailed as the most suitable biomass for most of the cases analyzed, due to its high energy power, followed by sugarcane residues, which are produced in greater quantities in the state. The conventional Rankine cycle was identified as the most mature technology and also had the lowest generation costs. In the projects that showed economic viability, the generation costs were between US$ 0.10/kWh and US$ 0.24/kWh.
|
|
6
|
-
MATHEUS BRENDON FRANCISCO
-
Metamodel-Based Optimization of Auxetic Structures
-
Advisor : GUILHERME FERREIRA GOMES
-
COMMITTEE MEMBERS :
-
CARLA TATIANA MOTA ANFLOR
-
BRUNO SILVA DE SOUSA
-
GUILHERME FERREIRA GOMES
-
MIRIAN DE LOURDES NORONHA MOTTA MELO
-
RICARDO MELLO DI BENEDETTO
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
Data: May 13, 2022
-
-
Show Abstract
-
A numerical-experimental study was performed to carry out a deep design optimization. It was considered four different auxetic models (reentrant auxetic tube, auxetic sandwich tube, reentrant beam, and double-V beam) and five different structural responses: mass, critical buckling load, natural frequency, Poisson’s ratio, and failure load under compression. The response surface methodology was applied to generate a metamodel based on non-linear equations that can represent reality. A new optimization algorithm called the Multi-Objective Lichtenberg Algorithm was used to find the best possible settings. In the reentrant beam, the failure load was increased by 26.75% in compression performance optimization. Furthermore, in the optimization of modal performance, the natural frequency was increased by 37.43%. For the double-V beam, the analysis done was able to increase the critical buckling load and failure load by 3.4% and 4.8%, respectively, in compression performance. With respect to modal performance, it was possible to increase the natural frequency by 26.88% and reduce the Poisson’s ratio by 24.59%. Regarding the reentrant auxetic tube, it was possible to improve Qu and λ in more than 40% while the mass was improved in 17.51% (in compression performance). In modal performance, the natural frequency and the Poisson’s ratio were reduced in 6.92% and 10%, respectively. Thus, for the auxetic sandwich tube, it was possible to improve (reduce) the Poisson’s ratio by more than 56% in the optimization of the modal performance and improve the mass, the Poisson’s ratio, the failure load, and the critical buckling load in the static performance. This is a numerical-experimental study unprecedented in the literature to date because it was possible to evaluate the static and modal performance of four auxetic model in a multi-objective optimization problem.
|
|
7
|
-
LUCAS LINCOLN FONSECA SOARES
-
VISCOUS/INVISCID INTERACTION METHOD FOR COMPUTATIONAL DOMAIN REDUCTION IN AERODYNAMICS PROBLEMS
-
Advisor : NELSON MANZANARES FILHO
-
COMMITTEE MEMBERS :
-
ALEX MENDONÇA BIMBATO
-
IVAN FELIPE SILVA DOS SANTOS
-
LUIZ ANTONIO ALCANTARA PEREIRA
-
MARCELO ASSATO
-
NELSON MANZANARES FILHO
-
RAMIRO GUSTAVO RAMIREZ CAMACHO
-
Data: Jun 30, 2022
-
-
Show Abstract
-
Since CFD simulations for external flow problems may cause high computational costs, the present work presents a viscous/inviscid interaction method that allows to simulate the viscous flow in a reduced computational domain, which encompasses the rotational flow region (also called viscous subdomain). To this end, an interactive methodology is proposed, which consists of: calculate the potential flow outside the rotational region, impose the velocity of potential flow as a boundary condition at the boundary of the viscous subdomain, calculate the transpiration velocity to correct the potential flow, recalculate the field of potential flow velocities outside the rotational region, and update boundary conditions in the viscous subdomain. To determine the transpiration velocity, the velocity decomposition approach was used, which has been explored to develop methods of viscous/inviscid interaction. In this approach, to calculate the transpiration, it is necessary to first determine a boundary where the vorticity is negligible (called the rotational boundary). Determining the rotational boundary can be a complex task. Thus, two proposals were made to address this issue: the use of an auxiliary surface to facilitate the determination of rotational boundary (and to serve as a control surface for the potential flow), and to use of alternative criteria to evaluate the negligible vorticity. Calculations were performed in the viscous subdomain for two-dimensional, incompressible, steady, laminar or turbulent flows problems, on the profile NACA 0012 , bluff bodies and multiple bodies. The use of the proposed methodology allowed to reduce the computational costs, and the results of the aerodynamic coefficients and of the viscous flow fields obtained in the reduced domain, in general, were satisfactory for the engineering levels.
|
|
8
|
-
CAMILA APARECIDA DINIZ
-
METAMODEL-BASED STATIC AND DYNAMIC OPTIMIZATION OF COMPOSITE STRUCTURES WITH PLY DROP-OFFS
-
Advisor : GUILHERME FERREIRA GOMES
-
COMMITTEE MEMBERS :
-
BRUNO SILVA DE SOUSA
-
GUILHERME FERREIRA GOMES
-
LUIZ CLAUDIO PARDINI
-
RICARDO MELLO DI BENEDETTO
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
YOHAN ALI DIAZ MENDEZ
-
Data: Jul 22, 2022
-
-
Show Abstract
-
Nowadays, there is no analytical equation able to analyze the issues related to the performance of structures with ply drop-offs. In order to address this issue, a metamodel using Design of Experiments and the SunFlower Algorithm for static and dynamic optimization of composite structures with ply drop-offs was developed in this study. Through numerical simulations and experimental tests, a characterization of the static and dynamic behavior of tapered hybrid and non-hybrid tubes was proposed. Then, a metamodel was developed considering the results obtained through numerical simulations, where the best ply drop-off location that provides the best static and dynamic conditions was identified, and, posteriorly, it was applied in the manufacture of the tubes. The numerical results revealed that the hybrid tube reinforced with carbon and glass of fibers supported a high loading in buckling conditions when compared with non-hybrid tubes. Before the manufacture of the tubular structures, an experimental comparative study using honeycomb sandwich structures with different face sheets and cores was proposed to analyze the fabric characteristics. The results showed that the hybrid fabric reinforced with glass and aramid of fibers was demonstrated to be not viable for tubular structure manufacture. Then, in the manufacture of the tubular structures, the carbon, glass, and carbon/aramid hybrid fabrics were applied. The experimental results obtained with the optimized structures revealed that the hybridization provided an increase in the level of damping. The modal analyses performed on the intact and damaged structures demonstrated a smooth reduction in the first natural frequency and in the damping factor for the damaged structures. Aiming a comparative analysis between tapered and non-tapered structures, tubular structures without ply drop-offs were manufactured and experimental tests were performed. The hybrid tapered structure manufactured with carbon, aramid, and glass of fibers proved to be a promising option in compression conditions, supporting a loading of 9.489 kN, while the non-tapered structure supported a loading of 13.283 kN. In addition, this hybrid structure revealed a lower manufacturing cost when compared with the other hybrid structures, and it was considered lighter with a mass of 53 grams. The non-tapered hybrid structure had a mass of 77 grams, 30% higher than the tapered structure’s mass. Therefore, metamodel-based static and dynamic optimization was demonstrated to be feasible and advantageous for determining the optimum ply drop-off location.
|
|
9
|
-
TANIA MARIE ARISPE ANGULO
-
LOW NOISE AXIAL FANS DESIGN METHODOLOGY THROUGH AEROACUSTIC ANALYSIS
-
Advisor : RAMIRO GUSTAVO RAMIREZ CAMACHO
-
COMMITTEE MEMBERS :
-
ISRAEL JORGE CARDENAS NUNEZ
-
LUIZ ANTONIO ALCANTARA PEREIRA
-
MARCELO ASSATO
-
NELSON MANZANARES FILHO
-
RAMIRO GUSTAVO RAMIREZ CAMACHO
-
SERGIO-RICARDO GALVÁN-GONZÁLEZ
-
WALDIR DE OLIVEIRA
-
Data: Jul 28, 2022
-
-
Show Abstract
-
The study and analysis of the sound generated by flows has become increasingly important in several areas of the industry, from household appliances to aeronautical propulsion systems. A particular case is the noise generated by axial fans applied to the cooling systems in the generator sets. These equipment represent considerable noise sources that must be treated so as not to exceed the permissible limits. In this sense, several studies have been carried out with the aim of analyzing and proposing noise control techniques caused by the fan flow. Therefore, aeroacoustics has become an important topic to be considered for the study and identification of noise sources and for the analysis of noise reduction techniques generated by rotor flow that can be incorporated from the design stage. This work presents a methodology for the aeroacoustic design of axial flow fan rotors, considering the lift wing theory, the radial equilibrium condition, the free and non-free vortex condition and the incorporation of the sweep effect based on a cubic function, as an aerodynamic noise control technique. The analysis methodology used is based on the integration of Computational Fluid Dynamics techniques and noise prediction models to determine the main performance characteristics of fans and to evaluate the influence of the incorporated geometric variations (sweep) in the acoustic and aerodynamic fields of the fan. Numerical analyzes were performed based on steady-state simulations to obtain fan characteristic curves and identify local noise sources. In a second approach, in transient regime, the sound pressure level was analyzed as a function of frequency for the analysis of the aeroacoustic behavior considering the ISO 13347-3: 2004 standard for the positioning of the receivers. The simulation methodology to obtain the aerodynamic magnitudes was validated through the experimental tests of a fan with constant thickness blades without torsion. The experiments were carried out in the test bench suitable for ASHRAE (Norma 5175) / AMCA (Norma 210-74) standards of the Ventilator Laboratory (LabVent) of IEM/UNIFEI. The main contribution of the methodology proposed in this work is the incorporation of aerodynamic noise control mechanisms from the design stage through an approach with low computational cost that consists of a sensitivity analysis to determine the parameters that allow reducing the local sources of aerodynamic noise without compromising the aerodynamic performance characteristics. The results showed that designs incorporating sweep based on free vortex and forced vortex condition reduce noise sources and improve the aerodynamic performance of the fan compared to the base rotor geometry. It is important to highlight that the designed axial fan rotors present a consistent aerodynamic behavior in terms of hydraulic efficiency and total pressure, since the maximum efficiency values are far from the stall region.
|
|
10
|
-
JOÃO LUIZ JUNHO PEREIRA
-
DEVELOPMENT OF A MULTI-OBJECTIVE OPTIMIZATION ALGORITHM BASED ON LICHTENBERG FIGURES
-
Advisor : GUILHERME FERREIRA GOMES
-
COMMITTEE MEMBERS :
-
SAMUEL DA SILVA
-
ANDERSON PAULO DE PAIVA
-
DOMINGOS ALVES RADE
-
GUILHERME FERREIRA GOMES
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
YOHAN ALI DIAZ MENDEZ
-
Data: Aug 22, 2022
-
-
Show Abstract
-
This doctoral dissertation presents the most important concepts of multi-objective optimization and a systematic review of the most cited articles in the last years of this subject in mechanical engineering. The State of the Art shows a trend towards the use of metaheuristics and the use of a posteriori decision-making techniques to solve engineering problems. This fact increases the demand for algorithms, which compete to deliver the most accurate answers at the lowest possible computational cost. In this context, a new hybrid multi-objective metaheuristic inspired by lightning and Linchtenberg Figures is proposed. The Multi-objective Lichtenberg Algorithm (MOLA) is tested using complex test functions and explicit contrainted engineering problems and compared with other metaheuristics. MOLA outperformed the most used algorithms in the literature: NSGA-II, MOPSO, MOEA/D, MOGWO, and MOGOA. After initial validation, it was applied to two complex and impossible to be analytically evaluated problems. The first was a design case: the multi-objective optimization of CFRP isogrid tubes using the finite element method. The optimizations were made considering two methodologies: i) using a metamodel, and ii) the finite element updating. The last proved to be the best methodology, finding solutions that reduced at least 45.69% of the mass, 18.4% of the instability coefficient, 61.76% of the Tsai-Wu failure index and increased by at least 52.57% the natural frequency. In the second application, MOLA was internally modified and associated with feature selection techniques to become the Multi-objective Sensor Selection and Placement Optimization based on the Lichtenberg Algorithm (MOSSPOLA), an unprecedented Sensor Placement Optimization (SPO) algorithm that maximizes the acquired modal response and minimizes the number of sensors for any structure. Although this is a structural health monitoring principle, it has never been done before. MOSSPOLA was applied to a real helicopter’s main rotor blade using the 7 best-known metrics in SPO. Pareto fronts and sensor configurations were unprecedentedly generated and compared. Better sensor distributions were associated with higher hypervolume and the algorithm found a sensor configuration for each sensor number and metric, including one with 100% accuracy in identifying delamination considering triaxial modal displacements, minimum number of sensors, and noise for all blade sections.
|
|
11
|
-
ANGELA DE JESUS VASCONCELOS
-
Study of microhardness, electrical conductivity and dry wear of Al7Si0,4MgxFe Alloys obtained by Horizontal Solidification
-
Advisor : MIRIAN DE LOURDES NORONHA MOTTA MELO
-
COMMITTEE MEMBERS :
-
MIRIAN DE LOURDES NORONHA MOTTA MELO
-
ANTONIO AUGUSTO ARAUJO PINTO DA SILVA
-
GEOVANI RODRIGUES
-
MARIA ADRINA PAIXÃO DE SOUZA DA SILVA
-
ANTÔNIO LUCIANO SEABRA MOREIRA
-
OTAVIO FERNANDES LIMA DA ROCHA
-
Data: Aug 23, 2022
-
-
Show Abstract
-
Welding processes may cause undesirable residual stresses and their detection is possible using different methods. Some residual stress detection methods or techniques are destructive or semi-destructive, therefore not always applicable. Some methods are non-destructive, such as X-ray diffraction, neutron diffraction, and ultrasound. However, they are complex and high cost. One possible non-destructive method to detect welding residual stresses, lower cost than the current techniques, is being studied by some researchers. This method is based in the phenomenon which a structure natural frequencies vary when welding residual stress is applied. It consists in measuring the structure natural frequencies after welding, and comparing them to natural frequency values considered ideal values. These ideal values are obtained from finite element method simulation. However, the welding process simulation is not trivial; it implies at least four finite element analysis which can be executed in different manners. Different researchers have been developing studies with different types of structures, materials and results. The aim of this study is to analyze which parameters can improve Finite Element Method (FEM) model performance for obtaining simulation results closer to experiments. In this study, five models were developed and validated with experimental studies. It was observed that models using 2D shell elements generate better results than odels using 3D solid elements. In addition, it was observed that the symmetry technique which can be used in plate simulations leads to significant lower computational times, but affect modal results and, in addition, do not generate natural frequency values for all vibrational modes; therefore, the symmetry technique should be avoided in this type of analysis. The birth and death technique, which simulates filler metal deposition, was also analyzed and it was concluded that this technique has small effect in the results. Finally, this work proposes an interpolation technique for the natural frequency values to evaluate the modal results variation due to welding residual stresses.
|
|
12
|
-
MARCELA SILVA LAMOGLIA
-
Study of phase equilibrium in the Al-Fe-Nb system by the Mechanical Alloying process
-
Advisor : ANTONIO AUGUSTO ARAUJO PINTO DA SILVA
-
COMMITTEE MEMBERS :
-
ANTONIO AUGUSTO ARAUJO PINTO DA SILVA
-
DURVAL RODRIGUES JUNIOR
-
GEOVANI RODRIGUES
-
KATIA REGINA CARDOSO
-
MIRIAN DE LOURDES NORONHA MOTTA MELO
-
Data: Nov 21, 2022
-
-
Show Abstract
-
The knowledge regarding balance phases of Al-Fe-Nb system, and the understanding of its microstructures, are important for development of several families of alloys, such as niobiumbased super-alloys for use in high temperatures, light metal alloys used mainly in vehicles bolide, and aluminum added steels, which request low density and high-resistance to corrosion materials, used in aerospace and aeronautic applications. Although, arc fusion and induction processes commonly used for manufacturing of these alloys present some restrictions, such as high temperatures, lengthy times of process, high steam pressure that cause severe mass loss by evaporation, significant fusion temperatures of elements and the impossibility to obtain materials in nanometric scale. Thus, the Mechanical Alloying (MA) process becomes a viable alternative, which allows to synthesize different phase types, as well as supersaturated solid solutions, and intermetallic phases, from elementary powders, in a solid state reaction, into high-energy ball milling that offers shorter process time when compared to conventional methods for obtaining of phases in alloys with refractory materials. Consequently, heat treatments subsequent to MA for phase formation and/or stabilization, in significant lower temperatures than those used for phase stabilization after fusion are reached. Thereby, it is possible to makes viable the study of phase balance in regions yet or few explored in diagram, such as Nb rich region, in temperatures under 1000 °C. Based on this premise, in this study there were produced samples in Fe-Nb and Nb-Al samples, with compositions that present isothermal sections with balance of three distinctive phases. In initial experiments, in order to define Mechanical Alloying parameters, powders of Al (min. 99.9%), Fe (min. 99.8%) and Nb (min. 99.8%) were used, with three different process control agents (methanol, hexane and stearic acid), under four milling times (20, 40, 60 and 80 hours). The MA was conducted in a high-energy planetary mill, under inert atmosphere of argon gas, in a rotation of 350 rpm. The analysis focused on powder morphology, distribution of particle size, phase formation, reductions of crystallite sizes and lattice parameter. After set the process parameters, these were applied to all studied compositions. Heat treatment after MA were developed for 600 °C, 800 °C and 1000 °C temperatures for 48 hours. The morphology analysis was made with Scanning Electron Microscopy, in Secondary Electron mode (SEM/SE). The phases were obtained by Xray diffractometry, and phase volumetric fractions by Rietveld refining. Phase formation was reached in some systems just after MA, such as for Nb-60Al (%at.) alloy, which presented 33.0 %v. of NbAl3 phase for milled condition, and 40.3 %v. after heat treatment at 1000 °C. For ternary system, in Nb-55Al-5Fe (%at.) sample, it was possible to reach the expected phase balance for its isothermal section. Furthermore, it was possible to observe that for all studies, intermetallic phases were obtained, e.g. Laves from heat treatment at 600 °C for Fe-15Nb (%at.) composition, the phases Nb2Al and Nb3Al in heat treatment at 800 °C for Nb-23Al (%at.) composition, and phases Nb3Al, Nb2Al and NbAl3 after heat treatment at 600 °C for Nb-17Al-3Fe (%at.) composition.
|
|
13
|
-
RAFAELLA BARRETO CAMPOS
-
STUDY ON THE USE OF ADAPTIVE CONTROL BASED ON THE STRUCTURE OF THE UNIVERSAL INTEGRATIVE REGULATOR APPLIED TO AIRCRAFT FLIGHT CONTROL
-
Advisor : SEBASTIAO SIMOES DA CUNHA JUNIOR
-
COMMITTEE MEMBERS :
-
ALEXANDRE CARLOS BRANDAO RAMOS
-
EDSON HIDEKI KOROISHI
-
MARCELO BRAGA DOS SANTOS
-
MARCELO SANTIAGO DE SOUSA
-
SEBASTIAO SIMOES DA CUNHA JUNIOR
-
YOHAN ALI DIAZ MENDEZ
-
Data: Dec 16, 2022
-
-
Show Abstract
-
This thesis addresses studies carried out in order to integrate the adaptive control techniques and the Universal Integral Regulator (UIR). The studies started with the implementation of a non-constant gain, but based on the error. This controller was called MUIR (Modified Universal Integral Regulator). This will be the first step towards implementing adaptive gains. Simulations with flight control of a quadrotor were performed. In order to validate the proposed technique, comparisons were made with the UIR, MRAC (Model Reference Adaptive Control) techniques and combinations of these techniques were tested. Comparisons were based on the results plotted on graphs and on the calculation of two performance indices: the accumulated error (AE) and the control demand (CD). An analytical stability demonstration of the MUIR control technique was also presented. After that, ways of integrating adaptive control with UIR were analyzed, and the obtained ways were tested. The contributions of this work were the presentation and detailed study of the MUIR controller applied to a quadrotor, and the initial studies carried out in order to design a Universal Integral Regulator with adaptive gains (AUIR).
|
|