We have developed a new iterative geostatistical seismic amplitude variation with angle (AVA) inversion algorithm that inverts prestack data, sorted by gathers, directly for high-resolution density, P-wave velocity, S-wave and facies models. This novel inverse procedure is based on two key main principles: the use of stochastic sequential simulation cosimulation as perturbation technique model parameter space global optimizer crossover genetic to converge simulated earth models toward an...

Supervised machine learning is widely researched nowadays. Several works have already been developed using genetic algorithms (GAs) for classification tasks evolving IF-THEN rules. Oftentimes, these methods are built integers and real values from one’s chromosome structure. In this paper, new important improvements proposed to Non-linear Computation Evolutionary Environment (NLCEE), a GA-based rule-set generator by Amaral Hruschka. The GA, called BIN-NLCEE, uses binary representation in its...

Due to the nature of seismic inversion problems, there are multiple possible solutions that can equally fit observed data while diverging from real subsurface model. Consequently, it is important assess how inverse-impedance models converging toward For this purpose, we evaluated a new methodology combine multidimensional scaling (MDS) technique with an iterative geostatistical elastic algorithm. The algorithm inverted partial angle stacks directly for acoustic and impedance (AI EI) models....

ABSTRACT Seismic inversion plays an important role in reservoir modelling and characterisation due to its potential for assessing the spatial distribution of sub‐surface petro‐elastic properties. amplitude‐versus‐angle methodologies allow retrieve P‐wave S‐wave velocities density individually allowing a better existing litho‐fluid facies. We present iterative geostatistical seismic algorithm that inverts pre‐stack data, sorted by angle gather, directly for: density; P‐wave; velocity models....

In this letter, we show how a seismic inversion method based on Bayesian framework can be applied poststack data to estimate the wavelet, noise level, and subsurface acoustic impedance. We propose different linearized forward model discuss in detail some stochastic quantities are defined geophysical interpretation. The Gaussian assumption for likelihood distributions enable obtain conditional distributions. is divided into two sequential steps: wavelet level estimation, which posterior...

One of the main challenge problems in geophysics is getting reliable seismic inverse models while uncertainty assessed. Seismic may be tackled a probabilistic framework resulting set equiprobable acoustic and elastic impedance models. Here we show new geostatistical AVO method from where density, Vp Vs are retrieved. With Earth also compute correspondent synthetic pre-stack data zero-reflectivity R(0) Gradient (G) We successfully applied this workflow to 3D dataset were known. The final best...

Seismic inversion is an important technique for reservoir modeling and characterization due to its potential in inferring the spatial distribution of subsurface elastic properties interest. Two most common seismic methodologies within oil gas industry are iterative geostatistical Bayesian linearized inversion. Although first able explore uncertainty space related with inverse solution a more comprehensive way, it also very computationally expensive compared approach. In this paper, we...

This letter presents ACO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">\BBR</sub> - <i xmlns:xlink="http://www.w3.org/1999/xlink">V</i> , a new computationally efficient ant-colony-optimization-based algorithm, tailored for continuous-domain problems. The algorithm is well suited application in seismic inversion problems, owing to its intrinsic features, such as heuristics generating the initial solution population and facility deal with...

Numerical reservoir models are nowadays a common tool in any oil and gas characterization project. Geostatistical seismic inversion is one of the available methodologies that allow to retrieve acoustic and/or elastic three-dimensional from reflection data. Then, standard approach transformation these impedance into petrophysical ones (e.g. porosity, facies) following pre-defined technique. The use geostatistical, or stochastic, inverse have advantage assessing uncertainty associated with...

Multimodal optimization attempts to find multiple global and local optima of a function. Finding set optimal solutions is particularly important for practical problems. However, this kind problem requires techniques that demand high computational cost large amount parameters be adjusted. These difficulties increase in high-dimensional space In work, we propose niching method based on recent developments the basins (optimal locations) identification reduce costs perform better spaces. Using...

Recent applications of seismic data in reservoir characterization, direct hydrocarbon indication, and production monitoring rely on the accuracy elastic logs (density sonic). Trying to develop (semi) quantitative seismic-based studies requires use well-to-seismic calibration; well information is used as a bridge between geology surface data. Nevertheless, we must take into account that are also indirect measurements rock properties subject various sources errors. Borehole wall rugosity due...

Summary We present a new iterative geostatistical seismic inversion algorithm that allows retrieving: density; P-wave velocity; S-wave velocity and facies models. This novel procedure is based on two key main ideas: stochastic sequential simulation co-simulation as the perturbation technique of model parameter space; genetic act global optimizer to converge towards an objective function, mismatch between recorded synthetic pre-stack data. At end each iteration, triplet elastic traces jointly...

This paper describes how geostatistical inversion based in a Bayesian framework can be modeled and applied on post- stack seismic data, yielding multiple stochastic realizations of acoustic impedance with improved vertical resolution conditioned to well data. The proposed method is capable jointly estimate not only the impedance, but also wavelet uncertainties results. Gaussian assumption for likelihood models enables ob- tain analytical expressions distribu- tions, which allows sampling...

An important feature present in neural network models is their ability to learn from data, even when the user does not have much information about particular dataset. However, most popular do perform well spatial interpolation problems due difficulty accurately modeling correlation between samples. On other hand, one of geostatistical methods for interpolation, Kriging, performs very but requires some expert knowledge fit model (semivariogram). In this work, we adapt Incremental Gaussian...

This study assesses the extent to which Brazilian Artificial Intelligence Plan (PBIA, Plano Brasileiro de Inteligência Artificial) of 2024 supports country’s pursuit AI sovereignty. The authors map financial allocations PBIA’s 54 proposed structural actions against components Belli’s (2023a; 2023b) key sovereignty enablers (KASE) framework: data, algorithms, computing capacity, connectivity, electricity, education, cybersecurity, and regulation. finds that support six KASE enablers, in...

The use of confidence estimation techniques on neural networks outputs plays an important role when these mathematical models are applied in many practical applications. However, few have the capability to deal with variable noise rate predictions over domain, making assumptions about reliability become not suitable their real accuracy. In this paper extension non-linear regression method estimate prediction intervals for feed forward is presented. main idea that residuals variance should be...

Abnormal pore pressures can result in drilling problems such as borehole instability, stuck pipe, circulation loss, kicks, and blowouts. Gradient pressure prediction is of great importance for risk evaluation planning new wells early stages development production oil reservoirs. In this paper, a stochastic simulation with point distributions method presented to integrate uncertain data cube characterization. The consists the use direct sequential distributions. Wells data, case, are...

Ant Colony Optimization (ACO) is an optimization metaheuristic based on the foraging behavior of ants. This was originally proposed to find good solutions discrete combinatorial problems. Many extensions ACO heuristic for continuous domain have been proposed, but even those that claim close similarity with classical (discrete domain) ACO, like ACOR, do not use information called visibility, commonly used in original algorithm. In this paper, we show importance visibility by proposing <sub...

Abstract PRAVAP 2 (Petrobras’ strategic project for reservoir characterization) aims mainly technology acquisition and development in characterization. The is based on a multidisciplinary approach, the synergism between operational research teams. New technologies are put together with others already consolidated. This paper presents some of developments acquired how they impact field oil production.

Time-lapse feasibility study is not only the first step to validate a 4D campaign; it also an essential tool for interpretation of acquired seismic volumes. In this paper we present practical example Brazilian offshore reservoir illustrate main issues and benefits analysis. Our goal emphasize which information may be accessed with analysis, as well uncertainties non-uniqueness in time-lapse interpretation.

The most effective way to integrate seismic data in the reservoir characterization process is through generation of impedance models derived from inversion. In this work we compared deterministic and geostatistical inversion results an oil field order improve generate a more accurate model, where behavior predictions could be done way. acoustic widely used technique, which yields single result, inverting available for parameters (P-impedance). Geostatistical generates multiple properties...