17/12
João Victor Neves Neiva, PUC-Rio
Data: 17/12/2025 às 10h e 0min
Local: on-line
Orientador: Igor Braga de Paula e Luc Pastur
Área de Concentração: Mecânica Aplicada
Resumo
Laminar separation bubbles (LSBs) have a significant role in the aerodynamic performance of airfoils operating at low Reynolds numbers. In many applications, they are subject to a time-varying level of environmental disturbances, such as in turbine cascades and unmanned aerial vehicles(UAVs). This time–varying regime is not yet fully described. In this scenario, reduced order models (ROMs) emerge as an useful tool to help describing and modeling the phenomenon. In this work, a four-dimensional and a three dimensional Galerkin model using Proper Orthogonal Decompostion (POD) and Sparse Identification of Dynamical System(SINDy) are tested for modeling the formation of a LSB. The model is built using experimental data from time-resolved Particle Image Velocimetry (PIV) measurements of a flow over a flat plate in a water tunnel with a convergent divergent geometry. To this end, the vorticity transport formulation of the incompressible Navier-Stokes equation is used to identify the steady solutions. Two flow states are identified and used as flow fields for the modal decomposition, and therefore, for the models. The first modal decomposition has four most energetic modes, according to Proper Orthogonal Decomposition (POD). Two static modes (u1 and u2) and two oscillatory modes (u3 and u4). The time dynamics of the third and fourth modes indicate an oscillating permanent regime reached after a transient. The resulting four-dimensional POD-Galerkin model correctly predicts the full transient dynamics, from the beginning of the boundary layer separation until the permanent regime with vortex shedding. The second modal decomposition has three most energetic modes and is in accordance with literature. The first mode is a static mode (u1), known as shift mode, and two oscillatory modes (u2, u3). The model retrieves the famous Landau equation and conjectures a Hopf bifurcation as the route for the dynamical system. Although the model only captures the transient part just before the boundary layer transition, it correctly predicts the dynamics.
Link da defesa:
https://pucrio.zoom.us/j/93537405214pwd=yB4A3y3UhQ3VXnPti1NUsMvpmV6TfQ.1
17/10
Vinicius Oliveira Fontes, PUC-Rio
Data: 17/10/2025 às 09h e 0min
Local: por acesso remoto
Orientador: Anderson Pereira
Área de Concentração: Mecânica Aplicada
Resumo
Mechanical cloaking is the idea of hiding the influence of a hole in a structure by designing a surrounding region known as a cloak. Many solutions proposed to this problem in the literature introduce some sort of multiscale topology optimization approach with mechanical metamaterials, that is, materials that derive their properties from both their constituents and architecture, which present a wide range of mechanical properties. To analyze these complex multiscale structures, it is common to presume scale separation using homogenization theory, despite its limitations due to the prohibitive computational cost of full-scale analyses. In this work, we developed solutions to mechanical cloaking inspired by the recent compatible boundary condition (CBC) paradigm, in which the cloaking device is optimized apart from the surrounding structure. First, we adapted the discrete material optimization algorithm to select homogeneous candidates in the solution of an elastostatic cloaking problem using the CBC paradigm, with results showing a drastic reduction in computational cost associated with the significantly smaller problem size, both in the analysis and optimization, since only the cloaking device is considered during the main design process. A second approach proposed in our work introduces an original bottom-up gradient-based multiple-metamaterial optimization method, developed from a modified version of discrete-continuous parameterization. Our deterministic multi-metamaterial selection efficiently optimized the cloaking device in a two-stage process (selection of metamaterial family followed by the fine-tuning of the density distribution), contrasting the complex data-driven solutions found in the literature. Finally, we adapted our multi-material framework to solve a shape matching problem, where we designed a beam to deform in an arbitrary sine-wave pattern under compression. Results showed that enriching the search space with multiple suitable candidate metamaterials yielded better shape matching performance than just optimizing the density of a single candidate.
Link da defesa:
https://puc-rio.zoom.us/j/93815648499?pwd=UoOSMVJE574bNbHQbgiHhmuMixKCIr.1
14/10
Amanda Alvim de Bonis, PUC-Rio
Data: 14/10/2025 às 10h e 0min
Local: on-line
Orientador: Jessé Paixão, Arthur Braga e Dario Parra
Área de Concentração: Petróleo e Energia
Resumo
A deposição de resíduos de parafinas em dutos offshore é um problema operacional antigo que ocorre em várias bacias no Brasil e no mundo, devido as baixas temperaturas e altas pressões presentes no leito marinho. Esse fenômeno pode ocasionar uma redução da eficiência do escoamento de fluídos e o aumento nos custos de manutenção dos equipamentos de produção submarina. Está pesquisa apresenta a viabilidade técnica de uma ferramenta mais segura, eficiente e de menor custo, que utiliza a radiação de um laser de alta potência acoplado a um Veículo Operado Remotamente (ROV) para derreter a parafina, removendo assim as camadas aderidas às paredes internas dos dutos e restabelecendo o fluxo adequado. O feixe de laser é direcionado para o exterior do duto, aquecendo-o o suficiente para atingir a temperatura necessária para o derreter a parafina que está em seu interior, conduzindo o calor sem danificar o equipamento submerso. Foram realizadas simulações numéricas e testes de laboratório para otimizar os parâmetros e alcançar o melhor cenário, garantindo a eficiência térmica do processo. Neste trabalho, foi utilizado uma seção curta de tubo para simular um duto offshore e parafina industrial nos experimentos de laboratório. Essa dissertação de mestrado descreve os resultados obtidos em laboratório e os compara com as simulações, analisando tempo de radiação, distância e potência. Os resultados obtidos atestaram a eficiência térmica do processo, indicando que o laser foi capaz de transferir calor através do duto e elevar a temperatura da parafina acima de seu ponto de fusão, validando o conceito proposto.
Link da defesa:
https://puc-rio.zoom.us/j/91514030610?pwd=4phlHOPRdcbVZuMWvED6CKj1frIyrN.1
13/10
Denise Amato Bacic, PUC-Rio
Data: 13/10/2025 às 11h e 0min
Local: on-line
Orientador: Florian Pradelle e Fernando Zegarra
Área de Concentração: Termociências
Resumo
A descarbonização é fundamental para alcançar a neutralidade de carbono no transporte marítimo. O etanol tem se destacado devido a seu reduzido impacto ambiental. Entretanto, seu baixo número de cetano, elevado calor latente de vaporização, lubricidade limitada e maior corrosividade limitam sua capacidade de autoignição e a operação duradoura em condições convencionais de motores de ignição por compressão (CI). Este trabalho explora formulações de combustíveis visando aprimorar a combustão de etanol aditivado. Foram preparadas 9 misturas, compostas por etanol hidratado (50-80% em volume), combinadas com 5-25% de biodiesel e até 5% de aditivos comerciais. Ensaios
foram conduzidos em uma Máquina de Compressão Rápida para avaliar a autoignição das misturas, em condições de partida quente e fria, variando a razão de compressão de 25:1 até o limite de falha de ignição. Os resultados demonstram que as misturas contendo até 5% de biodiesel ou aditivos melhoraram significativamente a autoignição, assegurando combustão completa e estável para razões de compressão superiores a 23. A análise do instante de injeção confirmou que o aumento da razão de compressão reduziu o atraso de ignição das misturas. Nessas condições, as formulações aditivadas mantiveram atrasos inferiores a 2 ms, e, em partida quente, a mistura 2,5% de biodiesel e 5,0% de aditivo apresentou atraso de 2,6 ms e desempenho comparável ao diesel marítimo, confirmando o potencial da aditivação e do controle do instante de injeção na calibração de motores CI a etanol.
Link da defesa:
https://puc-rio.zoom.us/j/97364221195?pwd=LyLOTPZTpIlvOMYFn21f3oDWUyDNa6.1
13/10
Pedro Henrique Cardoso Paulo, PUC-Rio
Data: 13/10/2025 às 14h e 0min
Local: on-line
Orientador: Márcio Carvalho e Helon Ayala
Área de Concentração: Petróleo e Energia
Resumo
Machine learning has seen rapid adoption across diverse sectors—from healthcare and finance to energy—due to its ability to uncover patterns and make predictions from complex datasets. In the oil and gas industry, this work explores three key applications of machine learning: soft sensoring for virtual flow metering, early-stage economic assessment of exploratory assets, and hybrid modeling for multiphase flow simulation. For soft sensoring, system identification techniques combined with current-time data significantly improved predictive accuracy while maintaining model utility for forecasting. In economic assessment, black-box classifiers trained on imbalanced datasets demonstrated feasibility for rapid asset appraisal, with investment data and oversampling strategies yielding notable performance gains, albeit with trade-offs in recall and interpretability. For multiphase flow modeling, hybrid models integrating commercial mechanistic models with data-driven estimators consistently outperformed both standalone physical and black-box models, achieving up to 71% error reduction in pressure gradient prediction. These findings highlight the potential of machine learning to augment traditional engineering workflows, improve decision-making, and address longstanding challenges in oil and gas operations.
Link da defesa:
https://puc-rio.zoom.us/j/95039469339?pwd=ndQ143lEgpXwGeJMa9rsNbEHmkhPbF.1
10/10
Pedro Leon Fontes Cardoso Bazan, PUC-Rio
Data: 10/10/2025 às 9h e 0min
Local: on-line
Orientador: Marco A. Meggiolaro, Vivian Medeiros e Wouter Caarls
Área de Concentração: Mecânica Aplicada
Resumo
Blind locomotion refers to the challenge of navigating varied terrains without prior knowledge or exteroceptive data. Although quadruped robots often use external sensors, these can be unreliable in low-light or resource-constrained settings and cannot anticipate disturbances such as slippage. In such scenarios, quadrupeds must rely exclusively on proprioceptive feedback, using internal measurements — joint positions, velocities, and contact forces — to adapt their locomotion strategies. While slip detection and terrain-estimation methods exist, leveraging proprioceptive information offers advantages across many applications. This work explores Meta-Reinforcement Learning (Meta-RL) to enhance policy robustness and rapid adaptation for quadruped robots during blind locomotion on challenging terrain, with the goal of achieving zero-shot generalization — i.e., enabling the agent to perform effectively in unseen environments without additional training. It builds on the RL² algorithm, integrating recurrent neural networks into Proximal Policy Optimization (PPO) to implicitly encode task-specific information from experience. Two novel RL²-based architectures are proposed and evaluated in simulation with the ANYmal C quadruped robot across diverse terrain conditions, focusing on flat surfaces with stochastic slip and highly unstructured terrains. Results show that recurrent policies significantly outperform standard PPO, improving both adaptability and robustness under unpredictable ground dynamics and thereby advancing the state of blind quadrupedal locomotion in challenging simulated environments, with implications for real-world deployment.
Link da defesa:
https://puc-rio.zoom.us/j/91899345128?pwd=kG4WTUnpW5fVkDuQ3hvcSQb4hxEjHu.1
09/10
Filipe Fiuza de Paula Alves, PUC-Rio
Data: 09/10/2025 às 14h e 0min
Local: 154L
Orientador: Renato Vieira
Área de Concentração: Petróleo e Energia
Resumo
O presente trabalho baseia-se no estudo e comparação entre fatores de segurança da norma API 579-1/ASME FFS-1: Fitness-For-Service e fatores de segurança obtidos experimentalmente para um vaso de pressão de aço carbono que apresenta perda localizada de espessura. Para este experimento, foi construído um vaso de pressão feito de um tubo de aço carbono com tampos soldados em suas extremidades, com dimensões reduzidas em escala, de maneira a reproduzir as mesmas tensões atuantes de um duto comercial comumente utilizado em serviços de transporte de fluidos, quando aplicada uma pressão interna. Na avaliação da integridade estrutural do vaso de pressão, foi utilizada a técnica DIC (Correlação Digital de Imagens) – um método óptico-numérico de registro e mapeamento de uma configuração de referência para uma configuração deformada – sendo comparado neste caso, o vaso de pressão antes e depois da aplicação de pressão interna, para medir o campo de deslocamento na região que contém a perda localizada de espessura e, por diferenciação, obter o campo de deformações na superfície. Com as medições de deformação obtidas a partir de experimentos e trabalhando no regime elástico, foi possível calcular as tensões atuantes no espécime, o fator de concentração de tensões e o fator de segurança. Por comparação, os fatores de segurança da norma de adequação ao uso são avaliados e são apresentados comentários sobre quão conservadores eles podem ser. Além disso, como a DIC é uma técnica não destrutiva, sem contato, que utiliza equipamentos leves e pouco volumosos, é discutida a possibilidade de aplicá-la para avaliar vasos de pressão e dutos em serviço sem que haja grandes transtornos.
07/10
Gustavo Henrique Gomes dos Santos, PUC-Rio
Data: 07/10/2025 às 14h e 0min
Local: 250L
Orientador: Ivan Menezes
Área de Concentração: Petróleo e Energia
Resumo
Effectively designing and managing today’s complex oil and gas production systems, from the reservoir to the wellhead and surface facilities, requires more than isolated simulations of each component. Traditional modeling approaches, which treat the reservoir and surface network separately, often rely on assumed boundary conditions that fail to capture the real interactions between subsurface and surface flow. This separation can lead to inaccurate predictions and suboptimal operational strategies. A more unified simulation approach, where the entire system is modeled as a continuous flow network, enables a deeper understanding of system behavior, supports more accurate performance forecasts, and allows for better-informed decisions in field development and production optimization. This study introduces an implicit reservoir–wellhead coupling methodology within the GSIM numerical simulator, developed through a collaboration between Petrobras and PUC-Rio. The proposed approach addresses key pressure losses along the production system, which are major factors in maximizing hydrocarbon recovery. As a foundational step toward full integration of reservoir and surface simulations, the method leverages GSIM’s plugin architecture and focuses on the residual and Jacobian matrix terms in Newton iterations. It uses vertical lift performance (VLP) tables in conjunction with nodal analysis to achieve effective coupling between surface and reservoir models. Bilinear and multidimensional linear interpolation of the hydraulics tables were implemented, tested, and validated against results from the commercial IMEX simulator, showing excellent agreement. Additional test cases were designed to compare the proposed coupling methodology with IMEX. Single-layer wells exhibited strong correspondence, while multilayer configurations revealed minor transient differences that resolved over time, demonstrating both the robustness of the method and GSIM’s newly developed capability to replicate coupled systems. This implementation allows for precise specification of wellhead pressure without compromising the flexibility of GSIM’s modular framework. The result is a reliable platform for integrated reservoir-to-surface modeling, offering enhanced simulation accuracy and improved support for field development decisions by explicitly accounting for pipes and tubing pressure losses.
07/10
Henrique Nunes Uchôa, PUC-Rio
Data: 07/10/2025 às 10h e 0min
Local: 106L
Orientador: Paulo R. de Souza Mendes, Alexandra Alicke e Ivan Siqueira
Área de Concentração: Termociências
Resumo
The behavior of simple interfaces is well described by an isotropic surface stress tensor with a single thermodynamic property, the so-called interfacial tension coefficient, that plays the role of a surface pressure and depends only on the excess concentration of surface-active components and temperature; as such, simple interfaces can be readily characterized with a standard Langmuir trough coupled with a Wilhelmy plate. The behavior of complex interfaces, conversely, must be described by a more robust surface stress tensor that contemplates the joint contribution of both thermodynamic and rheological properties. As a result, complex interfaces cannot be characterized with the same standard Langmuir trough because the resulting components of the surface stress tensor depend simultaneously on both shear and dilatational interfacial rheological properties. Here, we detail the design and manufacture of a new radial trough device for the rheological characterization of complex interfaces. The interface is compressed/expanded isotropically through the vertical movement of a cone-shaped funnel and a Wilhelmy tube probe is used to measure the interface force as the interface area changes while maintaining circular symmetry.
06/10
Matheus Pinto Xavier, PUC-Rio
Data: 06/10/2025 às 14h e 0min
Local: 106L
Orientador: Monica Naccache e Priscilla Varges
Área de Concentração: Termociências
Resumo
Para garantir o sucesso da operação de cimentação em poços de petróleo, é essencial que a pasta de cimento efetivamente desloque o fluido de perfuração do espaço anular entre o poço e o revestimento. Este processo é altamente complexo devido ao comportamento não newtoniano dos fluidos, adicionalmente complicado pela presença de regiões erodidas. A eficiência de deslocamento depende de vários parâmetros, incluindo as propriedades reológicas e densidades dos fluidos, vazão e parâmetros geométricos. Este estudo apresenta uma investigação experimental abrangente do processo de deslocamento em poços de petróleo com zonas erodidas, acompanhada por uma análise de sua eficiência de
deslocamento. O conjunto de fluidos compreende três fluidos newtonianos ou não newtonianos representativos daqueles encontrados em processos de cimentação reais. Diferentes conjuntos de fluidos, vazões, inclinação e excentricidades foram investigados. Caracterizações reológicas dos fluidos foram conduzidas usando um reômetro rotacional, e as densidades foram medidas com um densímetro digital. A seção de teste de acrílico consiste em uma geometria anular com uma expansãocontração adrupta, apresentando dimensões fixas e excentricidade variando de 0 a 100%. Uma bomba de parafuso único foi utilizada para impor uma taxa de injeção constante por meio de garrafas de transferência. Um medidor de vazão Coriolis foi instalado na saída da seção de teste para medir a taxa de vazão, o volume bombeado e a densidade dos fluidos que saem da seção de teste. Até agora, os resultados sugerem que a faixa de vazão e excentricidades exploradas não afeta significativamente a eficiência do deslocamento. Além disso, é mostrado que a razão entre viscosidade e densidade desempenha um papel importante no deslocamento os fluidos. Mais investigações são necessárias para compreender as interações entre os parâmetros em estudo.
25/09
Diego Santos Duarte, PUC-Rio
Data: 25/09/2025 às 09h e 0min
Local: 238L
Orientador: Ivan Menezes
Área de Concentração: Mecânica Aplicada
Resumo
Plates and shells are thin-walled structures with thicknesses that are small relative to their other dimensions. While plates primarily resist loads applied perpendicular to their mid-plane through bending, shells carry loads mainly via membrane forces along their curved surfaces. Their structural behavior is often studied using numerical methods, notably the Finite Element Method (FEM). Several types of finite elements have been developed, generally with triangular or quadrilateral geometries, for modeling these structures. More recently, polygonal elements have been proposed to provide greater geometric flexibility and mitigate numerical instabilities commonly encountered in topology optimization techniques, such as checkerboard patterns and one-point connection problems. This work presents a topology optimization framework for plate and shell structures using arbitrary polygonal finite elements. The primary motivation is to extend open-source educational software, such as PolyMesher and PolyTop, to include Reissner-Mindlin plate and shell formulations. The main numerical challenge lies in mitigating shear and/or membrane locking, which can lead to an overestimated stiff response in thin structures. For plates, a locking-free formulation is adopted that applies Timoshenko beam assumptions along element edges; for shells, a degenerated curved element with assumed shear and membrane strain fields is employed. The polygonal elements are validated using benchmark problems, and compliance-based topology optimization is then performed for both structural types. Additionally, this work presents local stress-constrained volume minimization results for plates via the augmented Lagrangian method and proposes the application of topology optimization for shell structures using polygonal meshes. The formulations are robust for both thick and thin structural regimes and also support structured quadrilateral meshes. A guideline for code modification is also provided to extend PolyTop for plate analysis, encouraging further research and educational applications of the proposed software.
16/09
Bruno Costa Pontes, PUC-Rio
Data: 16/09/2025 às 10h e 0min
Local: por acesso remoto
Orientador: Márcio Carvalho e Jonatas Grosman
Área de Concentração: Petróleo e Energia
Resumo
This master’s thesis investigates the application of machine learning techniques to pressure and temperature data collected from permanent downhole sensors in multi-zone oil wells of the Brazilian pre-salt carbonate reservoirs. These reservoirs feature complex architectures with intelligent completions that enable independent control and monitoring of multiple production zones, posing challenges for traditional physical modeling due to heterogeneous reservoir properties and interdependent flow dynamics. The research addresses the gap in existing studies, which mostly focus on single-zone wells, by applying supervised and unsupervised learning methods to a comprehensive dataset obtained from selective production tests of two wells with similar multi-zone completions. Clustering algorithms were used to identify patterns related to valve configurations, which supported the development of classification models to accurately predict valve status from sensor data. Regression models, enhanced by segmenting data according to zone combinations, effectively estimated total oil production rates. Model assessment across different wells made it possible to discuss robustness and generalizability, highlighting the importance of incorporating pressure- and time-dependent features, proper data normalization, and temporal cross-validation techniques. The results confirm that machine learning can successfully extract valuable information from complex sensor data, automate well monitoring, and improve decision-making processes in challenging multi-zone reservoir environments. This work contributes to advancing data-driven approaches for modern reservoir management, demonstrating their potential to complement and enhance traditional physical modeling in oil and gas production.
Link da defesa:
https://puc-rio.zoom.us/j/91940590477?pwd=bibGJdJVjtD4CTBJwHcaVsufROaCl0.1
31/07
Pedro Henrique Souza Calderano, PUC-Rio
Data: 31/07/2025 às 16h e 0min
Local: 106L, híbrida
Orientador: Márcio Carvalho e Helon Ayala
Área de Concentração: Termociências
Resumo
Two-phase flow in porous media is complex, as pore geometry and fluid properties strongly influence flow patterns. Macroscopic behavior is driven by pore-scale phenomena and typically described using parameters like absolute and relative permeability or capillary pressure. Determining these from pore-scale features is difficult, requiring extensive experiments or numerical simulations. Direct pore-scale simulation is challenging and computationally expensive. This work proposes machine learning-based tools to assess two-phase macro-scale porous media flow properties as a function pore space geometry and fluid properties. Recently, deep neural networks have been explored to predict properties of single-phase flow through porous media. However, only a few works have extended these models to infer two-phase flow behavior. In a first study, a data set is created with different pore space geometry based on Voronoi diagram patterns. The pore-scale two-phase flow in all geometries is solved using the finite element method. We propose as neural network architectures a Convolutional Neural Network and a Deep Operator Network to predict the evolution of produced oil volume that results from water injection. In a second study, models based on fully connected networks using a Fourier kernel and a Laplace-Beltrami eigenfunction kernel are proposed to predict the complex pore-scale flow patterns. The two-phase flow solution is also computed using the finite element method. Results indicate that the Laplace kernel-based architecture outperforms the Fourier-based network, which proved to be less reliable. We also explore training with a physics-informed loss in addition to a purely data-driven loss. However, the inclusion of the physics-based loss did not enhance model performance. The studies conducted in this thesis propose systems that are cheaper in terms of time and computational resources than the traditional methods due to the low cost of machine learning inference.
Link da defesa:
https://puc-rio.zoom.us/j/95144718618?pwd=oolOKYzsCb90fsaaYKvqDn0fIIf5B7.1
13/06
Alandmara Rosa Dionizio Leoncio, PUC-Rio
Data: 13/06/2025 às 15h e 0min
Local: por acesso remoto
Orientador: Márcio Carvalho e Ranena Ponce
Área de Concentração: Termociências
Resumo
Naturally fractured reservoirs (NFR) provide over 20% of the world's oil reserves. The coexistence of matrix and fracture with distinct permeability and porosity properties results in a rapid oil production decline and a high level of complexity. Emulsion flooding is an effective chemical Enhanced Oil Recovery (cEOR) method. The fundamentals of pore- and Darcy-scale flow of oil-in water emulsion and its connection to the observed micro and macroscopic flow mobility reduction remain unclear, despite recent advancements in emulsion injection as an EOR technique. The impact of emulsion drop size distribution (DSD) and its stability, pore throat distribution, injection flow rate, and pressure behavior on the aqueous phase's mobility reduction is a crucial consideration in the design of an effective emulsion injection process. This thesis investigated O/W emulsion flow dynamics in fractured porous media to assess the impact of pore and Darcy scale phenomena on the aqueous phase mobility reduction, as evaluated by mobility reduction factor (f) and residual resistance factor (RRF). A microfluidic device with a central fracture flanked by stratified zones of high and low permeability and a fracture model created by a longitudinally cut Bentheimer sandstone core assembled with a polyoxymethylene spacer, filled the voids with spheres of controlled size to establish a porous medium with permeability contrast, were used to study emulsion flow through NFR representative. For the purpose of simplifying the injection tests conducted through fractured porous medium, emulsions composed of mineral oil without salts were used in experiments involving oil displacement and injectivity tests. Permeability reduction of the fractured models as a function of emulsion DSD and capillary number indicated water mobility control exerted by emulsion droplets. Emulsion injection using DSD in the fracture region's average pore diameter range resulted in moderate damage and major pressure oscillation, demonstrating emulsion mobility control effect.
Link da defesa:
https://puc-rio.zoom.us/j/97066767077?pwd=CmY2aLBCfyhphogUfUPKlQC6sxokVF.1
28/05
Bruno Lima Davico de São João, PUC-Rio
Data: 28/05/2025 às 16h e 0min
Local: por acesso remoto
Orientador: Arthur Braga e Luiz Paulo Souza
Área de Concentração: Mecânica Aplicada
Resumo
A busca por maior eficiência, redução de custos e mitigação de riscos constitui pilares fundamentais na indústria de Petróleo e Gás (O&G). Nesse contexto, o tamponamento e abandono (P&A) permanente de poços petrolíferos é uma etapa operacional mandatória, de alta relevância técnica e ambiental, que impõe significativos desafios técnicos e econômicos ao setor. Conforme regulamentação da Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP), a efetivação do abandono permanente requer a existência de, no mínimo, dois Conjuntos de Barreiras Solidárias (CBS), sendo a interface entre o revestimento e o cimento um componente crucial de uma dessas barreiras. Um desafio particular desta fase reside na avaliação da integridade da cimentação do poço através da coluna de produção (perfilagem through-tubing). Diante desse cenário, esta dissertação detalha o desenvolvimento e a construção de uma bancada de testes laboratorial. Esta estrutura foi projetada para simular os principais tipos de defeitos encontrados na cimentação de poços reais, permitindo também a variação controlada da excentricidade da coluna de produção (tubing) em relação ao revestimento. Adicionalmente, é apresentado um protótipo de ferramenta de perfilagem acústica, concebido com características análogas às de ferramentas comerciais. Este protótipo foi empregado em ensaios na bancada para investigar a viabilidade técnica da avaliação da qualidade da cimentação por meio da coluna de produção.
Link da defesa:
https://puc-rio.zoom.us/j/99255254311?pwd=ow7EuMHjLFpgq77iosX5ymu3ux92Fa.1
23/05
Pedro Henrique Leite da Silva Pires Domingues, PUC-Rio
Data: 23/05/2025 às 14h e 0min
Local: por acesso remoto
Orientador: Igor de Paula, Helon Ayala e Alan Kubrusly
Área de Concentração: Mecânica Aplicada
Resumo
Monitoring systems are essential for structural integrity and industrial process characterization. However, their development faces challenges due to limited labeled data and the need for expert knowledge in feature extraction. This thesis proposes scalable and efficient training frameworks based on Transfer Learning (TL) and Unsupervised Learning (UL) --- aiming to reduce dependency on labeled data, simplify implementation, and increase adaptability across different contexts. Three case studies were conducted. The first introduces a UL framework for estimating tensile stress in aluminum plates using ultrasonic guided waves. Features were extracted using machine, deep, and transfer learning techniques, feeding a k-means model. The best approach achieved 96.00% labeling accuracy, with a 20% error reduction compared to the baseline, and is suitable for real-time execution on in edge computing modules. The other studies apply TL with speech-domain pre-trained models to extract features from accelerometer and acoustic signals for wind turbine blade damage detection and two-phase flow regime classification. The proposed framework achieved 99.60% accuracy in the second study (vs 99.20% for the unisensory baseline), inferring in half the time, and 99.69% in the third (vs 80.00%). The results validate the use of speech models for monitoring tasks involving vibrational and acoustic signals. This thesis provides practical insights into model selection, feature transferability, and computational scalability.
Link da defesa:
https://puc-rio.zoom.us/j/98360632156?pwd=BnCDTxvx6KobrYhvt13xixylngvL2R.1
21/05
Carlos Roberto Hollanda Lopes, PUC-Rio
Data: 21/05/2025 às 10h e 30min
Local: 106L, híbrida
Orientador: Rafael Menezes e Pedro Henrique Anjos
Área de Concentração: Petróleo e Energia
Resumo
Investigamos as instabilidades de dedos viscosos que surgem nas duas interfaces formadas pela injeção sequencial de três fluidos newtonianos em uma célula de Hele-Shaw retangular. Para este fim, utilizamos a análise de estabilidade linear e desenvolvemos um conjunto de equações diferenciais ordinárias de primeira ordem, formuladas para descrever a evolução das perturbações nas interfaces. A partir dessa abordagem, avaliamos o problema de forma perturbativa, com ênfase na análise do efeito de acoplamento entre as duas interfaces, quantificada pela espessura do fluido intermediário, no regime linear do escoamento. Em paralelo ao estudo teórico, realizamos uma série de testes experimentais para investigar o crescimento das instabilidades de Saffman-Taylor no problema de três fluidos e duas interfaces. A construção do dispositivo experimental é descrita detalhadamente, e inclui as adaptações realizadas no aparato experimental ao longo deste estudo. Os resultados indicam que a interação entre as interfaces é governada pela força de acoplamento entre elas, influenciando tanto o tempo de formação das instabilidades quanto a morfologia dos dedos viscosos. Este efeito foi identificado inicialmente através da análise linear. Em seguida, verificamos experimentalmente que a manipulação da espessura do fluido intermediário permite atrasar o desenvolvimento das instabilidades, e que esses efeitos perduram no regime não linear. Isso indica que o volume da fase intermediária pode ser controlado para minimizar as deformações das interfaces e, consequentemente, aumentar a eficiência do processo de deslocamento. Este fenômeno pode ser relevante para o setor industrial, servindo como estratégia para processos de recuperação avançada de petróleo (EOR). Destacamos, em particular, a aplicabilidade desses métodos em processos de injeção alternada (WAG) e em métodos químicos, nos quais fluidos são bombeados sequencialmente em meios porosos para otimizar o processo de extração.
Link da defesa:
https://puc-rio.zoom.us/j/99151575448?pwd=i5zxxpYCDHH7UiP4H5RRjD9a1Wfgpa.1
21/05
Alan de Oliveira Pinto Neves, PUC-Rio
Data: 21/05/2025 às 14h e 0min
Local: 154L
Orientador: Igor de Paula e Omar Pinedo
Área de Concentração: Mecânica Aplicada
Resumo
Este trabalho apresenta a caracterização de uma bomba piezoelétrica sem partes móveis para controle de fluxo. O direcionamento do escoamento é obtido por meio das geometrias presentes na entrada e na saída do dispositivo. O objetivo principal é caracterizar três configurações de válvulas: Tesla de dois canais, Tesla de quatro canais e uma válvula do tipo bocal/difusor, sob condições de acionamento pulsátil e contínuo.Os resultados indicaram que a frequência de 10 Hz proporcionou a maior vazão para as três geometrias, com valores da ordem de 0,01 mL/min. Em regime contínuo, observou-se que a válvula Tesla de dois canais apresentou a menor vazão máxima (cerca de 4,5 mL/min), enquanto a válvula Tesla de quatro canais atingiu aproximadamente 5,5 mL/min, e a válvula do tipo bocal/difusor alcançou cerca de 6,5 mL/min. No sentido reverso, para um mesmo ΔP, as vazões foram de aproximadamente 3,8 mL/min para a válvula Tesla de dois canais, 5,0 mL/min para a Tesla de quatro canais e 5,5 mL/min para a válvula bocal/difusor, evidenciando o efeito da diodicidade.Por fim, os campos de velocidade internos das válvulas foram analisados por meio da técnica de Velocimetria por Imagem de Partículas (PIV). Espera-se que a caracterização detalhada dessas bombas, aliada às medições obtidas por PIV, contribua para o avanço no desenvolvimento desses dispositivos.
20/05
Samuel Nascimento Cândido, PUC-Rio
Data: 20/05/2025 às 9h e 0min
Local: DEQM 1, 5º andar Leme
Orientador: Angela Nieckele
Área de Concentração: Petróleo e Energia
Resumo
Reducing the emissions of greenhouse gases in hard-to-abate segments, such as the petroleum industry, is a major challenge of the energy transition. A promising pathway in this direction is increasing the energy efficiency of the production processes. In the petroleum industry, the recurrent formation and growth of wax deposits over the inner walls of pipelines can substantially reduce the energy efficiency of the petroleum transportation process. Numerous studies have been conducted to accurately model and predict wax deposition, with the goal of enhancing pipeline system design and optimizing removal operations. However, an unequivocal deposition criterion is still an open problem. A Newtonian prediction model that defines the deposit thickness through the molecular diffusion across the wax deposit interface or an empirical solids saturation threshold is often used. Recent advances indicate that the non-Newtonian behavior of the oil should be considered as a deposition mechanism. Once the oil and gas industry uses long pipelines and needs fast predictions, simplified and accurate prediction models are much required. In this regard, the present work develops a quasi-1D wax deposition model coupled with a robust thermodynamic model. The fluid rheology can be Newtonian or non-Newtonian. Additionally, molecular diffusion is considered. The velocity and pressure fields are 1D, while the temperature and dissolved wax concentration fields are one-direction 2D fields, meaning they are determined by marching along the pipeline length. The mesh is dynamic, adapting to the movement of the deposit interface. The thermodynamic properties (solids saturation included) are obtained from a lookup table constructed by the thermodynamic model in a pre-processing stage. The predictive performance of different deposition criteria is compared with experimental data on temperature and deposit thickness of cases under various operating conditions. The non-Newtonian and Newtonian models presented good results for higher thermal driving forces. The Newtonian model was more dependent on the wax solubility curve. The molecular diffusion model presented good results for lower thermal driving forces or higher Reynolds numbers. These results are a further indication that the wax deposition phenomenon is not governed by a single deposition mechanism.
Link da defesa
https://puc-rio.zoom.us/j/97404762297?pwd=ekaxQWly0eaFUlkLBpciZXzoDuZa9S.1
20/05
José Eduardo Sanson Portella de Carvalho, PUC-Rio
Data: 20/05/2025 às 08h e 0min
Local: 106L, híbrida
Orientador: Florian Pradelle, Remí Revellin e Romuald Rullière
Área de Concentração: Termociências
Resumo
Integrating intermittent renewable sources with energy storage systems is an effective strategy to enhance energy network reliability and ensure power availability. This study examines three different energy storage technologies: Compressed Air Energy Storage (CAES), hydrogen, and lithium-ion batteries. These systems are coupled with photovoltaic panels and applied to a residential building in Lyon (France) and Rio de Janeiro (Brazil). The analysis spans a full year of operation and is evaluated using a 4E approach, assessing energy, exergy, exergoeconomics, and environmental impact. In terms of efficiency, CAES and hydrogen systems exhibit similar performance, except in the trigeneration configuration, where hydrogen demonstrates superior results. From an economic standpoint in both Rio de Janeiro and Lyon, when compared to business as usual, the hydrogen system incurs the highest costs (+260.9% and +244.1%), followed by CAES (+206.4% and +197.8%) and the lithium-ion batteries (+71.0% and +67.6%). Regarding environmental impact, CAES and lithium-ion batteries effectively reduce CO2 equivalent emissions in both cities, with greater benefits observed in Rio de Janeiro. In contrast, the hydrogen system only becomes viable when biogas replaces natural gas, as the system's performance improves significantly, substantially reducing the cost of avoided CO₂ emissions. With this configuration, all systems are able to avoid emission, notably in Rio de Janeiro, with 24.3 tons, 50.0 tons, and 26.1 tons for CAES, hydrogen and lithium-ion batteries. In Lyon, the values are lower with 2.45 tons, 15.3 tons and 1.49 tons. Overall, all three systems show promising potential, but their feasibility depends on cost reductions and efficiency improvements of its components.
Link da defesa
https://puc-rio.zoom.us/j/91602964874?pwd=bTfsvaJrZ6uoPapGZqCMKGaoykkjnd.1
19/05
Lucas Alves Marçano, PUC-Rio
Data: 19/05/2025 às 13h e 0min
Local: 106L
Orientador: Florian Pradelle e Epifanio Ticona
Área de Concentração: Petróleo e Energia
Resumo
O transporte marítimo, responsável por cerca de 3% das emissões globais de CO₂, enfrenta crescente pressão para descarbonização. Nesse cenário, a amônia verde emerge como uma alternativa promissora para substituir os combustíveis fósseis em motores marítimos, especialmente em configurações dual-fuel. Este trabalho realiza uma revisão bibliográfica, incluindo o levantamento do estado da arte sobre as rotas de produção da amônia verde, suas características físico-químicas, os impactos técnicos e econômicos, os aspectos ambientais e os desafios regulatórios associados. Além disso, o levantamento identificou o potencial da amônia verde como combustível marítimo, bem como os principais gargalos para sua adoção em larga escala. Adicionalmente, investiga-se o desempenho termodinâmico e ambiental da operação de um motor marítimo de ignição por compressão, modelado com base em um motor MWM de médio porte, com razão de compressão de 16,8 e operando a 1700 RPM. O estudo considera a utilização de injeção direta de alta pressão de amônia (650, 1000 e 1350 bar), combinada com injeção piloto de diesel a 1317 bar. As análises tridimensionais foram realizadas no software CONVERGE CFD, focando em parâmetros críticos como atraso de ignição, taxa de liberação de calor (HRR), eficiência térmica e emissões equivalentes de CO₂ em um horizonte de 100 anos (CO2-eq (100y)), considerando diferentes frações de energia da amônia (AEF), pressões de injeção e momentos de injeção do diesel piloto (SODI). Os resultados mostram que, para 50% de carga, foi possível operar com AEFs de até 62,5% sem comprometer a estabilidade da combustão, mantendo a eficiência térmica equivalente à do diesel puro (cerca de 48,8%) e reduzindo as emissões de CO2-eq em aproximadamente 11%. Em regimes de carga elevada (75% e 100%), foram observadas eficiências térmicas superiores às do diesel puro, com ganhos de 0,9 a 3,1 pontos percentuais, e reduções de até 83% nas emissões de CO2-eq. Por outro lado, em condições de baixa carga (25%), a substituição de diesel por amônia não se mostrou vantajosa, resultando em menores eficiências térmicas e emissões superiores às do diesel puro. Esses resultados evidenciam o elevado potencial da amônia verde como combustível marítimo para a descarbonização do setor, especialmente sob condições operacionais otimizadas.
16/05
Paulo Teixeira Vale de Carvalho, PUC-Rio
Data: 16/05/2025 às 09h e 0min
Local: por acesso remoto
Orientador: Marco A. Meggiolaro e Vivian Medeiros
Área de Concentração: Mecânica Aplicada
Resumo
Quadruped robots have many applications in the field of mobile robotics. They operate in rugged environments that challenge the capabilities of wheeled or tracked robots. However, they demand higher computational power and accurate environmental perception to enable their autonomous operation. Slippage is particularly challenging for legged systems, occurring when the ground reaction force (GRF) violates the friction cone. Slippage affects state estimation, self-localization, control systems, and increases the risk of falling. One approach to address this issue is slip detection, which triggers a reaction strategy. However, among the methods available in the literature, there is little direct comparison between them, and there is potential to incorporate additional slip characteristics to assist in reaction and improve environmental understanding. This dissertation aims to compare proprioceptive slip detection methods. The comparison uses simulation data, analyzing results through graphs and quantifying true and false detections. A filter is proposed to correlate a robot's foot's tangential and orthogonal velocity to filter out undesired detections. The study also investigates slip direction, proposing an estimator for its angle. An exponential moving average filter with a single iteration is applied to smooth the measurements. The proposed approach is validated in simulations with the Go1 robot navigating on slip-prone scenarios such as icy surfaces, inclined planes, and flat terrain. The comparison of the slip detection methods highlights the limitations of each approach and the relationship between true and false positives. At the same time, the proposed filter, in simulation, successfully filters out undesired detections. The filtered slip angle shows smoothness, consistent convergence in both simulated and experimental data, and a correlation with slip intensity.
Link da defesa:
https://puc-rio.zoom.us/j/93322186355?pwd=eP01NA3obBW4jrBknc8P6g4rfrwi6V.1
14/05
Felipe da Costa Pereira, PUC-Rio
Data: 14/05/2025 às 13h e 0min
Local: por acesso remoto
Orientador: Márcio Carvalho e Helon Ayala
Área de Concentração: Petróleo e Energia
Resumo
Multiphasic flow metering and well logging are key information that helps oil and gas companies understand reservoir characteristics and make strategic decisions. Recently, machine learning models have become an alternative to traditional data acquisition procedures based on physical hardware or first principle models, since the latter are, in general, expensive, time-consuming, and often unreliable. In this work, we propose new approaches for machine learning models to enhance the performance of virtual flow metering models and synthetic well log generation. For that purpose, this dissertation develops three application cases based on actual well data from the Volve Field in the Nothern Sea. Firstly, the application of stacking ensembles and non-linear system identification techniques improved RMSE metrics by 4.5% to 29% compared to benchmark regressors. System identification proved to be an effective approach for assessing optimal lags while stacking ensembles enhanced well-rate prediction by integrating various base learners. Secondly, a two-step training approach, applying a residual predictor, improved the quality of the well rates predictions by 1% to 13% depending on the complexity of the base estimator. In the third case, seasonal decomposition was used whether as a feature extraction method or as an ensemble strategy for well log generation purposes. While the first promoted a 27% gain in RMSE for compressional slowness log, the second outperformed the baseline models for 6 of the 8 examined estimators, reducing the density log RMSE by 5.2%. In all four scenarios examined, both methods achieved better results than the baseline strategy. These findings provide new guidelines for research in the field of virtual flow metering and well logs generation as well as in other domains.
Link da defesa:
https://puc-rio.zoom.us/j/98410105940?pwd=SC5F2Vs6QY7ddxtcapXnOyJ28n8Reo.1
13/05
Maria Leticia Costa Lobato, PUC-Rio
Data: 13/05/2025 às 15h e 0min
Local: 106L
Orientador: Florian Pradelle e Fernando Sánchez
Área de Concentração: Termociências
Resumo
A crescente demanda por redução de emissões no setor marítimo tem impulsionado o estudo de alternativas aos combustíveis fósseis. Nesse contexto, misturas ternárias de diesel marítimo, biodiesel e etanol surgem como uma opção promissora. A viabilidade dessas misturas depende da análise de sua estabilidade — com o uso de um co-solvente formulado a partir de planejamento experimental para evitar a separação de fases — e do desempenho em motores. Propriedades físico-químicas como densidade, viscosidade cinemática, ponto de fulgor, número de cetano e poder calorífico foram avaliadas, pois influenciam diretamente a combustão. Misturas com 25% de biodiesel (B25) apresentaram bom desempenho até 20% de etanol (B25E20), exceto no ponto de fulgor. Os testes foram realizados em uma Máquina de Compressão Rápida, comparando misturas com 0%, 10% e 20% de etanol e quatro avanços de injeção (1, 2, 3 e 4 mm) antes do PMS. Foram analisados parâmetros como pressão e temperatura máximas, atraso de ignição, trabalho e calor transferido e eficiência, com suporte de análise estatística. O aumento de etanol reduziu a pressão, temperatura máxima e aumentou o atraso de ignição, efeitos parcialmente mitigados pelo adiantamento da injeção. A eficiência termodinâmica utilizando diesel marítimo variou de 32,19% a 48,68%, e a de combustão caiu de 20,19% para 14,25%. Para o B25E0, a eficiência termodinâmica foi de 39,51% a 43,57%.
12/05
João Henrique Paulino de Azevedo, PUC-Rio
Data: 12/05/2025 às 09h e 0min
Local: Sala verde do IEPUC, híbrida
Orientador: Sergio Braga, Florian Pradelle e Ulf Moslener
Área de Concentração: Petróleo e Energia
Resumo
This thesis develops and validates an integrated geospatial modelling framework to assess the production potential of green hydrogen and its derivatives - ammonia and methanol - generated from renewable energy sources. High‐resolution datasets for solar photovoltaic, onshore wind, and offshore wind resources are combined with a comprehensive suite of economic, environmental, social, and technical constraints to transform theoretical (gross) resource estimates into realistic (technical) potentials. The framework employs detailed production models for water electrolysis and subsequent chemical conversion processes, calibrated against region‐specific data, while incorporating exclusion criteria such as capacity factor thresholds, infrastructural setbacks, and environmental protection zones. Applied to the Brazilian context, the analysis reveals significant reductions in deployable capacity due to constraints, thereby exposing the divergence between idealized resource availability and practical implementation. Furthermore, a techno‐economic evaluation is performed using metrics such as the Levelized Cost of Electricity (LCOE) and Levelized Cost of Hydrogen (LCOH). The results emphasize the necessity of accounting for real‐world limitations in renewable energy assessments and provide critical insights for optimizing site selection and guiding policy decisions. This work thus contributes to advancing geospatial methodologies for renewable energy integration and supports the formulation of sustainable investment strategies that are essential for global decarbonization efforts.
Link da defesa:
https://puc-rio.zoom.us/j/94079352995?pwd=hNpc2aoPecAa4iPIIaVO8naOEMn2PJ.1
09/05
Anna Luiza de Moraes y Blanco de Mattos, PUC-Rio
Data: 09/05/2025 às 10h e 0min
Local: 106L, híbrida
Orientador: Rafael Menezes e Eduardo Dias
Área de Concentração: Petróleo e Energia
Resumo
Investigamos o deslocamento radial de um fluido viscoso no interior de uma célula de Hele-Shaw horizontal e de geometria circular, promovido pela injeção de um fluido menos viscoso no centro da célula. Comparamos como diferentes protocolos de injeção afetam a instabilidade da interface entre os dois fluidos e a eficiência de deslocamento do sistema. Tais protocolos de injeção são modelados como funções lineares com platô, que apresentam vazão de injeção inicial nula e aumentam essa vazão de injeção linearmente com o tempo, até atingirem um platô. Este valor constante emula a máxima vazão de injeção possível em um sistema industrial de produção de óleo que utiliza a injeção de fluidos no reservatório como método de recuperação. As simulações numéricas que rastreiam a evolução dos dedos viscosos na interface são interrompidas quando esta atinge uma distância radial pré-definida, análoga à distância entre os poços injetor e produtor. Resultados mostram que vazões de injeção mais baixas tendem a gerar um número reduzido de dedos viscosos, de menores amplitudes, consequentemente levando a maiores eficiências de deslocamento. Isso permite que um volume maior do fluido menos viscoso seja injetado na célula, antes que a distância radial pré-definida seja atingida. Por outro lado, vazões de injeção mais baixas requerem maior tempo de injeção, o que se traduz em aumento dos custos operacionais em um cenário de campo. Assim sendo, apresentamos uma estratégia para otimizar o protocolo de injeção linear com platô, que maximiza o lucro, tendo como alvo o aumento da eficiência de deslocamento, mas também considerando a perda de tempo e o aumento dos custos operacionais relacionados a essa otimização.
Link da defesa:
https://puc-rio.zoom.us/j/94784211694?pwd=SI9R7vCCbkSSMc85BGNcr2ZiNbEqak.1
08/05
Ademir Freire de Medeiros, PUC-Rio
Data: 08/05/2025 às 09h e 0min
Local: 106L, híbrida
Orientador: Márcio Carvalho e Bruna Leopércio
Área de Concentração: Petróleo e Energia
Resumo
Enhanced oil recovery methods (EOR) are applied to mitigate non-uniform oil sweep. Gel treatment in the injection wells is one of these methods. It consists of blocking off the high permeability path with a polymeric gel to divert subsequently injected water into lesspermeable, oil-bearing pores. Fractures in reservoirs create preferential paths for the injected water and compromise the efficiency of the liquid displacement process. Sodium silicate gels could be injected to seal fractures and force the aqueous phase to flow through unswept regions. However, in this case, because the polymeric solution is injected with its cross-linker the gelation time is not predictable. As a result, it becomes difficult to control the depth at which the chemicals penetrate the fracture before solidifying, thus hindering the sealing of long fractures. Alternatively, microcapsules set a way to mitigate this problem, encapsulating the gelation activator making it possible to fill the fracture before activating the gelation. It allows controlling the timing and location of the content release. The microcapsules must be mechanically stable and impermeable to the encapsulated material during storage. Polydimethylsiloxane (PDMS) was used to encapsulate the hydrochloric acid (HCl), which is the activator of sodium silicate gelation, making it possible to better control the gelation reaction to reach deeper locations of fractured reservoirs. PDMS microcapsules were produced using a microfluidic device that combines co-flow and flow-focusing in coaxial glass capillaries. Fluorescent dyes were used in the inner and middle phases to enable microcapsule characterization. HCl release in hypotonic media (water) evaluation was made through pH measurement. Due to the osmotic difference, microcapsules swelled until burst. We discussed the effects of the mechanical properties of capsules on the HCl release process. Results show that the release time can be controlled by changing the eccentricity and stiffness of the capsules. A 3D device printed in white resin of a fractured porous medium was designed for our tests. Results show that the release time of HCl and the subsequent sodium silicate gelation can be controlled by adjusting the capsules’ shell material, thickness, stiffness, and eccentricity. PDMS microcapsules were injected alongside the sodium silicate solution. The device permeability was measured, before and after the gelation process, resulting in a considerable reduction in the device’s permeability.
Link da defesa
https://puc-rio.zoom.us/j/92846828910?pwd=5bRKJkfhGv3tpf5d6D0Jf9mmODaIg4.1
07/05
Vivian Mendes de Sousa, PUC-Rio
Data: 07/05/2025 às 10h e 0min
Local: Auditório Tecgraf, Pe Anchieta
Orientador: Márcio Carvalho e Brenda Costa
Área de Concentração: Petróleo e Energia
Resumo
Multiphase flow in highly heterogeneous systems, such as naturally fractured reservoirs, plays a crucial role in oil production. In the petroleum industry, hydrocarbons extracted from these reservoirs account for a significant portion of the global oil and gas production each year. The presence of fractures, vugs, and interconnected channels in such systems introduces complexity to fluid flow either by enhancing permeability through preferential flow paths or by acting as structural barriers that restrict flow. This behavior contrasts with that of homogeneous formations and requires a deeper understanding of fluid displacement mechanisms, especially at the pore scale. Thus, the goal of this research is to investigate two-phase flow behavior in fractured porous media, focusing on the determination and analysis of relative permeability curves as a function of pore-scale phenomena. To achieve this, an experimental approach was employed, using micromodels fabricated from PDMS. These devices replicate some geometric aspects of a porous matrix composed of a random arrangement of microchannels, into which different fracture geometries were incorporated. The experimental setup enabled real-time visualization of flow dynamics and the acquisition of the data and images needed for analysis. Steady-state water–oil injection experiments were performed on both fractured and non-fractured micromodels, aiming to direct comparison the resulting relative permeability curves. The results indicate that incorporating fractures into a porous matrix alters the relative mobility of the fluid phases, and anticipates water breakthrough. These effects reduce oil displacement within the matrix and eventually lower the oil recovery. Hysteresis effects were observed during the drainage and imbibition processes, highlighting the influence of fluid saturation history on phase distribution. Moreover, distinct flow regimes were identified within the fractures as a function of the fractional flow rate of water, which affected phase interactions and, consequently, influenced the relative permeability behavior. These findings emphasize the importance of pore-scale characterization in understanding multiphase flow in fractured porous media and reinforce the potential of microfluidic as a powerful tool for analyzing transport properties in complex porous systems.
03/04
Alexandre Carvalho Costa, PUC-Rio
Data: 03/04/2025 às 15h e 0min
Local: 414L
Orientador: Marcos Sebastião Gomes
Área de Concentração: Petróleo e Energia
Resumo
O Acordo de Paris (2015) estabeleceu os objetivos de longo prazo para manter o aumento da temperatura média da superfície global abaixo de 2°C em 2100, em relação aos níveis pré-industriais. O ideal é continuar com os esforços para limitar o aumento da temperatura a 1,5°C e atingir emissões líquidas zero (cenário Net Zero) em 2050 (IPCC AR6, 2021). Um dos maiores desafios para as empresas de petróleo e gás nos próximos anos é fornecer mais produtos energéticos com menor intensidade de carbono. Nesse contexto, marcos regulatórios para o mercado de carbono estão sendo implementados em vários países, que assinaram compromissos climáticos. Há um portfólio de tecnologias, em diferentes estágios de maturidade, que podem contribuir para a segurança energética e mitigação das mudanças climáticas. Dado esse contexto climático, a vulnerabilidade do setor de petróleo e gás a riscos físicos (integridade de instalações e equipamentos) e de transição (custos de carbono) enfatiza a urgência de enfrentar as mudanças climáticas. As reservas de petróleo e gás estão distribuídas globalmente e, em geral, estão localizadas em regiões sujeitas a condições climáticas extremas (ONU, 2023). Portanto, as decisões de investimento das empresas de petróleo e gás devem considerar, além das receitas da venda de hidrocarbonetos, o custo das emissões de carbono e as receitas de carbono oriundas de projetos de Captura, Utilização e Armazenamento de Dióxido de Carbono (CCUS). O objetivo deste trabalho é abordar o desafio de novos projetos de petróleo e gás para incluir questões climáticas em sua tomada de decisão e aprovação. O trabalho visa desenvolver uma metodologia para considerar as premissas de um mercado regulado de carbono e os investimentos, custos, créditos fiscais e/ou outras receitas de carbono associados ao CCUS. Os limites econômicos das reservas de petróleo e gás podem ser alterados por altos custos pelas emissões de carbono. O CCUS pode dar suporte à manutenção do desenvolvimento de projetos de petróleo e gás, durante a transição energética.
14/03
José Carlos Canazas Rodriguez, PUC-Rio
Data: 14/03/2025 às 10h e 0min
Local: por acesso remoto
Orientador: Mônica Naccache e Lorena Moraes
Área de Concentração: Termociências
Resumo
Xanthan gum (XG) is a biopolymer extensively utilized as a rheology modifier across several industries, including food, and cosmetics or oil and gas, where enhancements in physicochemical properties are essential. In this regard, the rheological response of XG under various thermophysical conditions can be significantly influenced by theinclusion of nanomaterials like graphene oxide (GO). The goal of this work is to investigate how the rheology and microstructure of XG aqueous solutions are affected by the addition of GO nanostructures. The Modified Hummers technique was used to synthesize GO nanosheets. The rheological response was analyzed through steady-state and oscillatory flow tests, in order to assess the impact of GO concentration in a fixed concentration of XG in water at various pressures and temperatures. In addition, samples containing sodium chloride at concentrations close to saturation were examined. XG suspensions displayed a shear-thinning and viscoelastic response. The addition of GO results in a more robust suspension with an increase in shear viscosity and viscoelastic moduli, as well as thermal stability at high temperatures, although less impact was observed at high pressure. In contrast to GO, the addition of sodium chloride showed a reduction in the rheological response at low temperatures and a diminished impact at high temperatures. A phenomenological model with two power-law regimes was used to fit flow curves. Cryomicrographs provided a visual representation of the microstructure of the suspensions and corroborated the rheological results. The observed improvements in thermal stability and viscoelastic characteristics present encouraging avenues for the creation of novel functional fluids.
Link da defesa:
https://puc-rio.zoom.us/j/99213788496?pwd=uNqa5k4KAEL3bDYnZE15StbJ5vaDnj.1
19/02
Caio Vinicius Santos Cartaxo, PUC-Rio
Data: 19/02/2025 às 14h e 0min
Local: 106L
Orientador: Márcio Carvalho e Amanda Pessoa
Área de Concentração: Termociências
Resumo
A way of reducing CO2 emissions is to decrease energy consumption in buildings. One promising approach to achieve this goal is the use of phase change materials (PCMs) as thermal energy storage systems for thermoregulation applications. However, the incorporation of PCMs in building components is challenging. Microencapsulation of PCMs has shown promising results as a passive form to improve the thermal performance of building materials, reducing the wasted energy with thermal comfort. In this context, this work is focused into two different aspects of microencapsulated phase change materials. The first is a numerical investigation of the internal temperature behavior of building walls with a thin layer of concrete with capsules. A one-dimensional transient methodology is proposed, with the presence of the microcapsules being simulated by a concentration function and the phase change process with the effective heat capacity method. The effects of the layer’s position, concentration, PCM melting temperature, and phase change enthalpy in the wall’s internal temperature profile were assessed. Moreover, we fabricated microencapsulated PCMs using poly (dimethylsiloxane) (PDMS) as a shell material and calcium chloride hexahydrate (CaCl2 · 6H2O) as PCM core. The production was performed using glass capillary microfluidic devices to ensure the production of monodisperse microcapsules with tunable geometrical properties such as size and shell thickness. The study shows a direct relationship between the position and concentration on the internal temperature as well as the reduction of internal temperature of walls. We were able to produce and stock PDMS/CaCl2 · 6H2O microcapsules with tunable properties. Finally, preliminary thermal efficiency tests were performed, demonstrating that this combination of shell/core was not ideal for these applications, however, due to the promising results presented by the numerical model, new combinations will be developed and tested in future works.
Link da defesa
https://puc-rio.zoom.us/j/91586983995?pwd=Lb21ROSVIa4yoH5cu4WtMqYi5bXK0R.1
18/02
Stélio Henrique Lopes Neto, PUC-Rio
Data: 18/02/2025 às 09h e 0min
Local: Híbrida, 106L
Orientador: Márcio Carvalho, Sergio Ribeiro e Florian Pradelle
Área de Concentração: Termociências
Resumo
Slot coating is largely used in the manufacturing of different functional films, including battery electrodes. For optimum performance, the thickness of the film deposited over the moving substrate must be uniform along the cross-web direction. The flow uniformity is achieved by properly designing the geometry of the internal cavities of the coating die. The optimized design is strongly influenced by the liquid rheology, which needs to be accurately described in the flow models. The rheological behavior of particle suspensions, as those encountered in slot coating of battery electrodes, is very complex. The viscosity varies orders of magnitude with the shear rate, and they show thixotropic behavior. Usually, flow models used in the design of slot die cavities do not take into account the time dependency of particle suspensions and the flow is described assuming that the local viscosity is only a function of the local deformation rate. This can lead to inaccurate description of the flow, since the liquid structure does not change instantaneously after a step change to a new stress level. This study addresses the three-dimensional internal flow of thixotropic fluids. Time-dependent behavior is modeled using a kinetic equation for the material fluidity, which is defined as the reciprocal of viscosity, and specifies uniquely the material microscopic state. The model accounts for the mechanisms of buildup and breakdown of the microstructure with no postulated functions nor additional auxiliary parameters. A 3D finite element numerical model, implemented in Python, based on the precompiled library named Dolfin, developed within the open-source project FeniCs, allows for a realistic representation of thixotropic fluid flow in complex geometries, such as those present in the internal cavities of slot coating dies. The results show the effect of thixotropy in the flow behavior and how they should be included in the design of slot dies used for coating particle suspensions.
Link da defesa:
https://puc-rio.zoom.us/j/96054249056?pwd=ToWSK32cDl8iJ6LoxhMBmSFA23lcDG.1
02/04
Monique Feitosa Dali, PUC-Rio
Data: 02/04/2019 às 14h e 0min
Local: Sala 1007 Prédio IAG
Orientador: Márcio da Silveira Carvalho e Frederico Carvalho Gomes
Área de Concentração: Petróleo e Energia
Resumo
The present work focuses on a numerical investigation of the flow through porous media with macropores, such as carbonates. The presence of vugs and fractures have a strong effect on the flow characteristics. The flow through the porous matrix is usually described by the Darcy equation and the flow through the macropores by Stokes equation. The coupling between these two distinct approaches brings great complexity to the modeling of such flows. In this work, we use Brinkman formulation that is able to describe the flow, both in the porous matrix and macropores with a single differential equation. We solved the set of differential equations using the finite element model and implemented the code in the FEniCS platform. We first solved the 1-D flow through parallel plates with one of the walls being a porous material. The goal was to compare the predictions obtained with the Brinkman formulation to that obtained by using the Beavers-Joseph boundary condition. Then, we solved a 2-D flow through a porous medium with macropores. The geometry of the pore structure was obtained from 2D slices of tomographic images of carbonates. The goal of this analysis was to evaluate an equivalent permeability as a function of macropores area and structures.