Le T.D.,LNCC |
Murad M.A.,LNCC MCT |
Pereira P.A.,LNCC |
Boutin C.,ENTPE |
And 2 more authors.
Society of Petroleum Engineers - SPE Reservoir Simulation Symposium 2015 | Year: 2015
In this paper we construct coupled gas flow models in shale-matrix and hydraulic fractures within the framework of the reiterated homogenization procedure in conjunction with the treatment of fractures as (n-1) interfaces (n = 2, 3) and adsorption isotherm of Langmuir-based monolayer model in the nanopores. At the nanoscale the Langmuir model is applied to reconstruct general monolayer adsorption isotherms of pure methane in the intra-particle porosity of the gas-wet organic matter. The nanoscopic model is upscaled to the microscale where kerogen particles and nanopores are viewed as overlaying continua forming the organic aggregates at thermodynamic equilibrium with the free gas in the water partially saturated interparticle pores. The reaction/diffusion equation for pure gas movement in the kerogen aggregates is coupled with both Fickian diffusion of dissolved gas in water and Darcy free gas flow in the interparticle pores also lying in the vicinity of the inorganic solid phase (clay, quartz, calcite) assumed impermeable. By postulating continuity of fugacity between free and dissolved gas in the interparticle pores and neglecting the bound water movement, we upscale the microscopic problem to the mesoscale, where organic and inorganic matter, and interparticle pores are homogenized. The upscaling gives rise to a new nonlinear pressure equation for gas hydrodynamics in the interparticle pores including a new storage parameter dependent on the total carbon content (TOC) and porosities. The new pressure equation in the shale matrix is coupled with single phase gas flow in the hydraulic fractures. The reduction of dimensionality method is applied to treat fractures as interfaces by averaging the flow equation across the fracture aperture. Combination of the methods give rise to a new matrix/fracture coupled problem. Numerical simulations illustrate the potential of the multiscale approach proposed for computing gas production curves and recovery factor in different gas flow regimes. Copyright © 2015 Society of Petroleum Engineers.
Novotny A.A.,LNCC MCT |
Szulc K.,Polish Academy of Sciences |
Zochowski A.,Polish Academy of Sciences
2012 17th International Conference on Methods and Models in Automation and Robotics, MMAR 2012 | Year: 2012
In the paper, a variant of the genetic algorithm for finding the location and size of small holes is considered. Both linear and non-linear elliptic boundary value problems are introduced in two domains with the conditions of transmission on the common boundary. The expansion of the shape functional for non-linear part and the expansion of Steklov-Poincare operator for linear part are provided in order to determine the form of topological derivative for the coupled model. The value of topological derivative is used for computing the probability density applied later in generating location of holes for genetic algorithm. © 2012 IEEE.
Coutinho D.F.,Pontifical Catholic University of Rio Grande do Sul |
De Souza C.E.,LNCC MCT
IFAC Proceedings Volumes (IFAC-PapersOnline) | Year: 2010
This paper proposes a linear matrix inequality based method to the estimation of a robust domain of attraction for a class of discrete-time nonlinear systems subject to uncertain constant parameters. Recursive algebraic representations of the system dynamics and of the Lyapunov stability conditions are applied to obtain convex conditions which guarantee the system robust local stability while providing an estimate of the domain of attraction. A large class of discrete-time nonlinear systems and of Lyapunov functions can be embedded in the proposed methodology including the whole class of regular rational functions of the system state variable and uncertain parameters. A numerical example illustrates the application of the proposed method. © 2010 IFAC.
Kritz M.V.,LNCC MCT |
Trindade dos Santos M.,LNCC MCT |
Urritia S.,Federal University of Minas Gerais |
Schwartz J.-M.,University of Manchester
Journal of Theoretical Biology | Year: 2010
Metabolic networks are among the most widely studied biological systems. The topology and interconnections of metabolic reactions have been well described for many species. This is, however, not sufficient to understand how their activity is regulated in living organisms. These descriptions depict a static set of possible chains of reactions, with no information about the dynamic activity of reaction fluxes. Cyclic structures are thought to play a central role in the homeostasis of biological systems and in their resilience to a changing environment. In this work, we present a methodology to help investigating dynamic fluxes associated to biochemical reactions in metabolic networks. We introduce an algorithm for partitioning fluxes between cyclic and acyclic sub-networks, adapted from an algorithm initially developed to study fluxes in trophic networks. Using this algorithm, we analyse three metabolic systems: the central metabolism of wild type and a deletion mutant of Escherichia coli, erythrocyte metabolism and the central metabolism of the bacterium Methylobacterium extorquens. This methodology unveils the role of cycles in driving and maintaining metabolic fluxes under perturbations in these examples, and may be used to further investigate and understand the organisational invariance of biological systems. © 2010 Elsevier Ltd.
Figueredo G.P.,Federal University of Rio de Janeiro |
Ebecken N.F.F.,Federal University of Rio de Janeiro |
Augusto D.A.,LNCC MCT |
Barbosa H.J.C.,LNCC MCT
Memetic Computing | Year: 2012
One issue in data classification problems is to find an optimal subset of instances to train a classifier. Training sets that represent well the characteristics of each class have better chances to build a successful predictor. There are cases where data are redundant or take large amounts of computing time in the learning process. To overcome this issue, instance selection techniques have been proposed. These techniques remove examples from the data set so that classifiers are built faster and, in some cases, with better accuracy. Some of these techniques are based on nearest neighbors, ordered removal, random sampling and evolutionary methods. The weaknesses of these methods generally involve lack of accuracy, overfitting, lack of robustness when the data set size increases and high complexity. This work proposes a simple and fast immune-inspired suppressive algorithm for instance selection, called SeleSup. According to self-regulation mechanisms, those cells unable to neutralize danger tend to disappear from the organism. Therefore, by analogy, data not relevant to the learning of a classifier are eliminated from the training process. The proposed method was compared with three important instance selection algorithms on a number of data sets. The experiments showed that our mechanism substantially reduces the data set size and is accurate and robust, specially on larger data sets. © 2012 Springer-Verlag.