Produção Científica



Apresentação
23/05/2022

Seismic attributes and machine learning techniques for identification and characterization of carbonate seismic facies of the Barra Velha Formation, in the Wildcat Prospect, Santos Basin.
Seismic data provides important information for prospect identification and reservoir characterization. For an effective seismic characterization, first we need to compute seismic attributes and combine them in a skillful way, to obtain as much information as possible to identify the different seismic patterns. Machine learning techniques applied to seismic interpretation have been successful in this regard and very useful in assisting with limitations involving data classification.

This work aims to apply a methodology for identification and characterization of carbonates facies from Barra Velha Formation, on the Wildcat Prospect in Santos Basin, using seismic attributes and a non-supervised facies classification that uses a machine learning method called Self-Growing Neural Network (SGNN). For the workflow, we performed the following steps: (i) Carbonate seismic patterns identification through seismic amplitude, where was possible to identify the build-up, characterized by chaotic seismic textures with a conical external geometry and internal fracturing; debris facies, exhibit prograde geometry with chaotic internal texture; carbonate platform facies showing a well defined flat parallel reflectors; and
the bottom lake facies, that does not have specific geometry and internally the reflectors are chaotic
(Neves et al., 2019). (ii) seismic attributes generation and analysis to characterize the carbonate seismic patterns, where the eigen coherence, dip steered enhancement, relief and relative acoustic impedance
assisted the seismic characterization. (iii) Principal component analysis (PCA), with amplitude filtered with
DSE, eigen coherence and relief as inputs, and (iv) non-supervised seismic classification with attribute clustering, using the seismic attributes and PCA results as input.

In the results, we could associate the seismic facies with the carbonate seismic patterns of Barra Velha Formation, previously identified. We obtained a clear differentiation between the carbonate platform and more fractured areas related to the build-ups. Similarly, the chaotic behavior of debris was well captured. However, the bottom lake patterns didn’t appear very evident. The PCA before the seismic classification brought better results. The attribute clustering method, has been an effective approach to differentiate the fractured zones mainly associated to build-ups and the facies layering of the carbonate platform in the Wildcat Prospect.

Apresentação
18/11/2019

An√°lise de escoamentos bif√°sicos em meios porosos para duas diferentes configura√ß√Ķes atrav√©s de simula√ß√£o num√©rica 2D utilizando OpenFOAM
Em um cen√°rio de baixos pre√ßos do petr√≥leo no mercado internacional e a necessidade de elevados investimentos para a ind√ļstria do petr√≥leo no desenvolvimento de novos campos descobertos, o investimento na otimiza√ß√£o da explora√ß√£o de campos maduros √© uma alternativa para aumentar a produtividade e os lucros. Inova√ß√Ķes na gest√£o e uso de novas tecnologias influenciam cada vez mais as decis√Ķes operacionais. Os constantes avan√ßos na computa√ß√£o de alto desempenho t√™m permitido o uso de softwares CFD (computational fluid dynamics) na obten√ß√£o de solu√ß√Ķes num√©ricas, em um tempo aceit√°vel, para as equa√ß√Ķes que modelam a din√Ęmica em reservat√≥rio. Neste trabalho, o software usado para a solu√ß√£o num√©rica das equa√ß√Ķes diferenciais foi o OpenFOAM [2], que √© uma plataforma de c√≥digo aberto que pode ser customizada. O solver usado em nossas simula√ß√Ķes implementa a solu√ß√£o num√©rica das equa√ß√Ķes modeladoras que tem como base a lei de Darcy e a equa√ß√£o de conserva√ß√£o de massa [1], em que as principais vari√°veis s√£o as press√Ķes e as satura√ß√Ķes dos fluidos. Neste trabalho, a produtividade de um campo de petr√≥leo madura foi avaliada atrav√©s da compara√ß√£o dos volumes acumulados de √≥leo no po√ßo produtor, para duas diferentes configura√ß√Ķes de po√ßos sob as mesmas condi√ß√Ķes iniciais e de fronteira em simula√ß√Ķes 2-D. A configura√ß√£o mais produtiva (com
o maior volume acumulado de óleo) foi ainda estudada através da aplicação de diferentes gradientes de pressão entre o poço de injeção de água e o poço de produção, em busca de uma relação entre a pressão aplicada e a quantidade de
óleo removido. Cinco diferentes gradientes de pressão foram testados e classificados pelo volume de óleo acumulado.
Apresentação
18/11/2019

Estudo de Variabilidade em Sistemas Turbulentos por Fluxo Interno Bidimensional Utilizando Simulação Numérica
Durante algum tempo as dificuldades encontradas em estudar equa√ß√Ķes que regem escoamentos turbulentos, implicaram na exist√™ncia de linhas de investiga√ß√Ķes, te√≥ricas e pr√°ticas, ambas mesmo que usando caminhos distintos tinham como
finalidade identificar padr√Ķes de turbul√™ncia. Em 1904, Prandtl, formulou a teoria da camada limite [2] aproximando estas duas correntes em mec√Ęnica
dos fluidos. Foi em virtude do aprimoramento da computa√ß√£o de alto desempenho, que surge uma nova linha de estudo chamada Din√Ęmica dos Fluidos Computacional, que utiliza m√©todos num√©ricos para encontrar solu√ß√Ķes das equa√ß√Ķes que governam os escoamentos. Embasado pela teoria e motivado por um dos principais problemas encontrados no transporte de fluidos l√≠quidos atrav√©s de dutos submetidos a varia√ß√Ķes de temperatura que √© a obstru√ß√£o parcial e at√© total em toda linha de transporte, como exemplo, a deposi√ß√£o de parafinas no transporte de petr√≥leo bruto em dutos que operaram em ambientes frios [3], o presente trabalho tem o intuito de apresentar alternativas que contribuam com a detec√ß√£o do surgimento de tais problemas. Os estudos aqui realizados em CFD (Computational Fluid Dynamics) foi utilizada a biblioteca OpenFOAM [1], que √© um software de c√≥digo aberto para solu√ß√£o num√©rica de equa√ß√Ķes diferenciais usando o m√©todo de volumes finitos, em particular, obten√ß√£o de solu√ß√Ķes num√©ricas das Equa√ß√Ķes de Navier-Stokes, onde foram simuladas geometrias em dutos 2-D com algumas varia√ß√Ķes de obst√°culos onde as velocidades foram calculadas em pontos fixados em eixos paralelos as paredes do duto. Os comprimentos relativos dos obst√°culos da parede superior para o da parede inferior s√£o alterados, a fim de variar a quebra de simetria. √Č
neste sentido que o presente trabalho se prop√Ķe a estudar os efeitos dessas obstru√ß√Ķes na natureza da turbul√™ncia por elas induzidas, visando identificar
atrav√©s dos c√°lculos realizados a dist√Ęncia m√°xima em que o efeito da quebra de simetria √© preservado. √Č simulado um duto de se√ß√£o transversal retangular, parcialmente obstru√≠do, com um escoamento turbulento de um fluido incompress√≠vel
Apresentação
18/11/2019

Escoamentos externos: influ√™ncia da turbul√™ncia na estabilidade e manobrabilidade de ve√≠culos submarinos aut√īnomos uma an√°lise num√©rica
As an√°lises em Din√Ęmica de Fluidos, especialmente a partir de resolu√ß√Ķes das equa√ß√Ķes de NavierStokes [5], se tornaram vi√°veis com o desenvolvimento tecnol√≥gico da computa√ß√£o de alto desempenho. N√£o sendo ainda poss√≠vel resolver tal equa√ß√£o de maneira anal√≠tica, m√©todos num√©ricos s√£o aplicados para obten√ß√£o de respostas para problemas regidos por equa√ß√Ķes desse n√≠vel de complexidade ¬Ī para este trabalho, as equa√ß√Ķes de Navier-Stokes s√£o resolvidas numericamente atrav√©s da plataforma open source OpenFOAM, que
utiliza o Método dos Volumes Finitos (MVF) [2] em suas estruturas internas.
O trabalho, a priori, consiste em observar as componentes hidrodin√Ęmicas [3] exercidas sobre tr√™s modelos simplificados de ve√≠culo submarino aut√īnomo (VSA) em duas dimens√Ķes (vide figura 1) quando estes est√£o submetidos √† corrente mar√≠tima. Essas medidas de drag, lift e torque s√£o utilizadas para avaliar a estabilidade e a manobrabilidade do VSA quando sua posi√ß√£o varia em pequenos √Ęngulos (corre√ß√Ķes de curso) em rela√ß√£o √† corrente.
Apresentação
18/11/2019

Effects of torque produced by wake on the maneuverability of a flatfish autonomous underwater vehicle
Autonomous underwater vehicles (AUV) are important resources to be used in the oil exploration industry in deep waters as well as a platform for scanning devices used in open sea regions of difficult human access. This work aims to analyze through computer simulations the influence of marine currents on the maneuverability of a flatfish shaped AUV. The 3D realistic scale simulations were performed on the Yemoja supercomputer located at SENAI-CIMATEC and describe the temporal evolution of the torques in the three rotational degrees of freedom - roll, yaw and pitch. The torques were calculated for two different inlet velocities and three angles (yaw) of attack showing a significant gain in the amplitude of these with increasing velocity and pitch being the component with the greatest amplitude of oscillation.
Apresentação
18/11/2019

Comparative Analysis of Two-Phase Flow in Porous Media in Two Different Configurations of Wells Through 3D Numerical Solution Using OpenFOAM
With the advance of high-performance computing (HPC) the use of CFD software has become common in research in Fluid Dynamics, either to obtain the numerical solutions of differential equations that model the various systems or the dynamics for viewing and analyzing data obtained in the simulations. In this work we look to benchmarking between two distinct two-phase flows from two geometrically different wells through the analysis of their respective accumulated volumes of non-wetting fluid in the outflow. Three simulations were performed, for each set of wells, both during the same interval of time and under the same pressure gradient. The OpenFOAM software and Finite Volume Method for solving differential equations were used.
Apresentação
18/11/2019

Evaluation of Static and Dynamic Elastic Properties in Carbonate Rock
The understanding of rocks mechanical properties is essential for some of the petroleum industry sectors, such as drilling, well stimulation and development. Rock mechanics data, as Young's modulus and Poisson's ratio, can be obtained by the static and dynamic conditions using triaxial compressive and ultrasonic tests, respectively. This work analyses the behaviour of static and dynamic elastic properties in a set of 20 carbonate core samples and compares with other literature results. Our approach is based in fit equations to predict static properties from the dynamic data, considering the occurrence of the frictional sliding or closing of cracks and microcracks, while performing triaxial tests. The results indicated a strong relationship among the effective pressures applied, porosity, density, and the efficiency of static and dynamic property relations. Additionally, porosity type could be indicated as one of the causes of the difference between static and dynamic moduli, since the inclusion of density and porosity in the relations demonstrated a significant improvement between Young's static and dynamic modulus correlations.
Apresentação
18/11/2019

Pressure Effects on the Prediction of the Dry Bulk Modulus Using the Nur Critical Porosity Model
Carbonate reservoirs are some of the most important plays in the oil and gas business around the world. The monitoring of fluid distribution within those rocks is not an easy task due to the complexity of their pore structure, which also influences the relationship between the petrophysical properties and seismic data. Fluid substitution theories like Biot-Gassmann often used for that monitoring, depend on the accurate info of the dry bulk modulus (Kdry), which is usually obtained from P- and S-wave velocities. In cases where those velocities are not available, it is possible to use other theories like the Nur (critical porosity) model to estimate Kdry from porosity and mineral content. This work aims to use a dataset of petrophysical data measured in core plugs to evaluate how the external pressure influence the prediction of Kdry based on the Nur Model. The results indicated the impact of pressure in the lab measured porosity affect the accuracy of such predictions and a modification in the Nur Model is proposed for accounting those effects.
Apresentação
18/11/2019

Correlation Between Hysteresis, Elastic Moduli and Petrophysical Properties in Sedimentary Rocks
ut on 05 sandstones and 03 carbonate core samples. The results indicated that hysteresis is severely dependent on depositional texture, and it is directly related to Young modulus and Bulk modulus. Tight rocks exhibited higher hysteresis than friable rocks, due to frictional sliding and grain contact adhesion, which causes permanent damage to pore structure. Porosity displays an inverse relation to hysteresis, as high pore density enables rock’s matrix to deform and recover its shape without frictional sliding and grain contact adhesion.
Apresentação
18/11/2019

Harvesting the computational power of heterogeneous clusters to accelerate seismic processing
Cluster environments are crucial to modern geophysics. Major processing companies make use of one or more computational environments, whether they be in-house clusters or third-party public clouds, to guarantee the efficient execution of their processing flows. But the diversification of such environments created a demand for software tools that are able to scale with efficiency in these ever-increasing ecosystems. Aside from efficiency requirements, these tools must also be able to handle and recover automatically from the faults that arise from these new and complex ecosystems. In this paper, we discuss how we leverage the Scalable Partially Idempotent Tasks System (SPITS) programming model and the PY-PITS runtime system to efficiently harvest the computing power of heterogeneous systems in order to solve geophysics problems. We also present an experiment in which we combine the computational resources from several clusters and workstations simultaneously to perform the regularization of seismic data and demonstrate the scalability and robustness of the system.
<<  <   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23   >  >>


An√ļncios

29/05/2023

 

IX SEMANA DE INVERNO DE GEOF√ćSICA 03-05 de Julho, UNICAMP, Campinas-SP

A Semana de Inverno de Geofísica é um evento regular do Instituto Nacional de Ciência e Tecnologia de Geofísica do Petróleo (INCT-GP/CNPq), organizado pelo Grupo de Geofísica Computacional (GGC) do IMECC/UNICAMP.O evento é direcionado a estudantes em fim de graduação e início de pós-graduação interessados na área de Geofísica Aplicada, principalmente nos temas ligados à exploração de petróleo. A Semana presta-se a divulgar a carreira de Geofísica, franqueando aos alunos de pós-graduação o acesso a temas atuais de pesquisa em Geofísica, por meio de especialistas renomados na área.

Maiores informa√ß√Ķes: https://www.lgc.ime.unicamp.br/sig/