scholarly journals Neural Network Prediction of P-wave Log for Reservoir Characterization in the Niger Delta Basin

2020 ◽  
Vol 17 (1) ◽  
pp. 28-32
Author(s):  
A. Ogbamikhumi ◽  
S.A. Salami ◽  
W.N. Uwadiae
Keyword(s):  
Neural Network ◽  
Correlation Coefficient ◽  
Reservoir Characterization ◽  
Gamma Ray ◽  
P Wave ◽  
Wave Parameter ◽  
Estimation Methods ◽  
Sonic Log ◽  
Neural Network Prediction ◽  
Niger Delta Basin

This study present a new technique that integrates several logs for P-wave prediction to minimize some errors and uncertainties associated with most estimation methods. The adopted method involves application of an artificial neural network technique that integrates density, resistivity and gamma ray logs for data training and the prediction of P-wave log. The results obtained gave correlation coefficient of 0.77, 0.24 and 0.42 between the acquired P-wave log and the acquired density, resistivity and gamma ray logs respectively, to demonstrate the relationship between P-wave log and the selected logs for the prediction process. The correlation coefficient of the estimated P-wave from Gardner and Faust methods with the acquired P-wave log are 0.64 and 0.59 respectively, while that of the neural network derived P-wave gave a better correlation coefficient of 0.81. Cross plot validation of P-wave derive Acoustic Impedance against density for both lithology and fluid discrimination revealed clusters for neural network derived P-wave parameter similar to the acquired P-wave derived parameters. Results of the presented neural network technique have been demonstrated to be more effective than results of the two conventional techniques. Keywords: Sonic log, Gardner’s method, Faust method, Neural network, Cross plot.

scholarly journals The BP Network Classification Leafminer-Infected Leaves Based on the Fractal Dimension

2011 ◽  
Vol 1 ◽  
pp. 163-167
Author(s):  
Da Ke Wu ◽  
Chun Yan Xie
Keyword(s):  
Neural Network ◽  
Fractal Dimension ◽  
Correlation Coefficient ◽  
Bp Neural Network ◽  
Input Data ◽  
Bp Network ◽  
Neural Network Prediction ◽  
Infected Tomato ◽  
Network Prediction ◽  
Hidden Nodes

Leafminer is one of pest of many vegetables, and the damage may cover so much of the leaf that the plant is unable to function, and yields are noticeably decreased. In order to get the information of the pest in the vegetable before the damage was not serious, this research used a BP neural network to classify the leafminer-infected tomato leaves, and the fractal dimension of the leaves was the input data of the BP neural network. Prediction results showed that when the number of FD was 21 and the hidden nodes of BP neural network were 21, the detection performance of the model was good and the correlation coefficient (r) was 0.836. Thus, it is concluded that the FD is an available technique for the detection of disease level of leafminer on tomato leaves.

scholarly journals SEISMIC TRAVELTIME TOMOGRAPHY IN THE DOM JOÃO FIELD, RECÔNCAVO BASIN, BRAZIL

2015 ◽  
Vol 33 (1) ◽  
pp. 89
Author(s):  
Naiane Pereira de Oliveira ◽  
Amin Bassrei
Keyword(s):  
Reservoir Characterization ◽  
New Technologies ◽  
Real Data ◽  
Seismic Inversion ◽  
P Wave ◽  
Gradient Algorithm ◽  
High Rate ◽  
Traveltime Tomography ◽  
Sonic Log ◽  
High Resolution Images

ABSTRACT. Tomography was incorporated in Exploration Geophysics with the intention of providing high-resolution images of regions in Earth’s subsurface that are characterized as potential reservoirs. In this work, seismic traveltime tomography in the transmission mode was applied to real data from the Dom João Field, Recôncavo Basin, State of Bahia, Brazil. This basin represents a landmark of oil exploration in Brazil and has been intensively studied since the 1950’s. Today, the Recôncavo Basin is still the principal oil producer in the State of Bahia, but there is a demand for new technologies, especially for mature fields, to improve hydrocarbon recovery. Acoustic ray tracing for the computation of traveltimes was used for forward modeling, and the conjugate gradient algorithm with regularization through derivative matrices was used as the inverse procedure. The estimated tomograms were consistent with available data from a sonic log near the acquisition area in terms of the layer geometry, as well as the P-wave velocity range. The results showed that traveltime tomography is feasible for the characterization of reservoirs with a high rate of vertical change, similar to the Dom Jo˜ao Field.Keywords: traveltime tomography, seismic inversion, regularization, reservoir characterization, Recˆoncavo Basin.RESUMO. A tomografia foi incorporada na Geofísica de Exploração justamente para fornecer imagens de alta resolução de regiões do interior da Terra, consideradas como potenciais reservatórios. Neste trabalho aplicamos a tomografia sísmica de tempos de trânsito no modo de transmissão em dados reais do Campo de Dom João, Bacia do Recôncavo, Estado da Bahia, Brasil. Esta bacia representa um marco da exploração de petróleo no Brasil e vem sendo exaustivamente estudada desde a década de 1950. Embora haja uma demanda por novas tecnologias, em especial para campos maduros, com o propósito de se aumentar a recuperação de hidrocarbonetos, a Bacia do Recôncavo é ainda a principal produtora do Estado da Bahia. Para o procedimento da modelagem direta foi utilizado o traçado de raios acústicos e para o procedimento inverso foi utilizado o algoritmo do gradiente conjugado com regularização através de matrizes de derivadas. Os tomogramas estimados foram consistentes com os dados provenientes do perfil sõnico de um poço próximo ao levantamento tomográfico analisado, tanto em termos de geometria de camadas, como também na faixa de velocidades da onda P. Os resultados mostraram que a tomografia de tempos de trânsito é viável para a caracterização de reservatórios com elevada taxa de variação vertical, que é o caso do Campo de Dom João.Palavras-chave: tomografia de tempos de trânsito, inversão sísmica, regularização, caracterização de reservatórios, Bacia do Recôncavo.

Determination of formation shear attenuation from dipole sonic log data

Geophysics ◽  
2019 ◽  
Vol 84 (3) ◽  
pp. D73-D79 ◽  
Author(s):  
Qiaomu Qi ◽  
Arthur C. H. Cheng ◽  
Yunyue Elita Li
Keyword(s):  
Reservoir Characterization ◽  
Wave Attenuation ◽  
Synthetic Data ◽  
Low Frequency ◽  
P Wave ◽  
Flexural Wave ◽  
S Wave ◽  
Additional Energy ◽  
Log Data ◽  
Sonic Log

ABSTRACT Formation S-wave attenuation, when combined with compressional attenuation, serves as a potential hydrocarbon indicator for seismic reservoir characterization. Sonic flexural wave measurements provide a direct means for obtaining the in situ S-wave attenuation at log scale. The key characteristic of the flexural wave is that it propagates at the formation shear slowness and experiences shear attenuation at low frequency. However, in a fast formation, the dipole log consists of refracted P- and S-waves in addition to the flexural wave. The refracted P-wave arrives early and can be removed from the dipole waveforms through time windowing. However, the refracted S-wave, which is often embedded in the flexural wave packet, is difficult to separate from the dipole waveforms. The additional energy loss associated with the refracted S-wave results in the estimated dipole attenuation being higher than the shear attenuation at low frequency. To address this issue, we have developed a new method for accurately determining the formation shear attenuation from the dipole sonic log data. The method uses a multifrequency inversion of the frequency-dependent flexural wave attenuation based on energy partitioning. We first developed our method using synthetic data. Application to field data results in a shear attenuation log that is consistent with lithologic interpretation of other available logs.

scholarly journals Geothermal modelling and its application to hydrocarbon generation from agbada formation. a case study of USANi field, Niger delta basin, Nigeria

2017 ◽  
Vol 5 (2) ◽  
pp. 75
Author(s):  
Godwin Aigbadon ◽  
Anthony Okoro ◽  
Elesius Akpunonu ◽  
Rosemary Nimnu ◽  
Azuka Ocheli
Keyword(s):  
Gamma Ray ◽  
Geothermal Gradient ◽  
Hydrocarbon Generation ◽  
Sonic Log ◽  
Resistivity Logs ◽  
Source Rock Evaluation ◽  
Geothermal Model ◽  
Niger Delta Basin ◽  
Gamma Ray Log

The geothermal model was done with the integration of surface. Subsurface temperature's data and formation depth values from suites of well log in the study field. The well comprises Gamma-ray log (GR log), Spontaneous Potential logs (SP log), Resistivity logs, Formationdensity, Neutron log and Sonic log. The suites of welllog within the studied sequences penetrates Agbada and the Benin Formation. The Benin Formation comprises mainly of continental sands, and the Agbada Formation consist of alternating sequence of sand and shales within the study wells. The estimated thickness and temperature values within the study field falls within the range from 1357- 3500m and 101 O C – 120.5 O C with estimated geothermal gradient range of (0.028 - 0.03 O C/100m) in the field. The geo-temperatures results range of 101.60 O C – 119.60 OCat modeled depth of 1357m- 3500m, indicating that the shale sequence at the basal path of the Agbada Formation is thermally matured with sufficient organic matter to generate hydrocarbon in the study field as earlier believe to be immature and cannot generate hydrocarbon. The geothermal model can be applicable to any sedimentary basin in the world. This work is also an important tool in source rock evaluation to compliment petroleum geochemistry and position the hydrocarbon generating window of the study field.

On the seismic wavelet estimative and reflectivity recovering based on linear inversion: Well-to-seismic tie on a real data set from Viking Graben, North Sea

Geophysics ◽  
2020 ◽  
Vol 85 (5) ◽  
pp. D157-D165
Author(s):  
Isadora A. S. de Macedo ◽  
José Jadsom S. de Figueiredo
Keyword(s):  
Reservoir Characterization ◽  
Gamma Ray ◽  
Real Data ◽  
Estimation Methods ◽  
Linear Inversion ◽  
Seismic Section ◽  
Data Set ◽  
Well Location ◽  
Well Data ◽  
Depth Relationship

Tying seismic data to well data is critical in reservoir characterization. In general, the main factors controlling a successful seismic well tie are an accurate time-depth relationship and a coherent wavelet estimate. Wavelet estimation methods are divided into two major groups: statistical and deterministic. Deterministic methods are based on using the seismic trace and the well data to estimate the wavelet. Statistical methods use only the seismic trace and generally require assumptions about the wavelet’s phase or a random process reflectivity series. We have compared the estimation of the wavelet for seismic well tie purposes through least-squares minimization and zero-order quadratic regularization with the results obtained from homomorphic deconvolution. Both methods make no assumption regarding the wavelet’s phase or the reflectivity. The best-estimated wavelet is used as the input to sparse-spike deconvolution to recover the reflectivity near the well location. The results show that the wavelets estimated from both deconvolutions are similar, which builds our confidence in their accuracy. The reflectivity of the seismic section is recovered according to known stratigraphic markers (from gamma-ray logs) present in the real data set from the Viking Graben field, Norway.

Research on Wind Power Prediction by Combining Mesoscale Numerical Model with Neural Network Model

2012 ◽  
Vol 512-515 ◽  
pp. 771-777
Author(s):  
Feng Ming Yu ◽  
Xi Cang Li ◽  
Jin Hua Song ◽  
Chun Xiang Gao ◽  
Chun Long Jiang
Keyword(s):  
Neural Network ◽  
Wind Power ◽  
Correlation Coefficient ◽  
Bp Neural Network ◽  
Wind Farm ◽  
Power Prediction ◽  
Wind Power Prediction ◽  
Neural Network Prediction ◽  
Network Prediction ◽  
Relative Root

Effective wind power prediction on wind farm can not only guarantee safe operation of wind farm, but also increase wind power storage and utilization efficiency. This research combines mesoscale numerical weather prediction model with BP neural network model for the use of wind power prediction. WRF model is used to recalculate the meteorological elements of trial wind farm from Jun. 2008 to Jun. 2009, and the accuracy check result shows that the correlation coefficient between predicted value and corresponding measured value of wind speed reaches 0.72. Predictions accuracy of wind direction, air temperature, humidity and air pressure are also precise, which meets the requirement of building BP neural network prediction model. The BP neural network prediction models of output power of 40 wind turbines are established on trial wind farm one by one, to analyze the influence of data normalization method and neuron number at the hidden layer on prediction accuracy. The prediction test every 10 minutes, with the actual effect of 24 hours, is done for 26 days, and prediction accuracy test is conducted by using independent samples. The result shows that relative root mean square error of the output power of the single wind turbine from 24.8% to 32.6%, and the correlation coefficient between predicted value and measured value is from 0.45 to 0.68; relative root mean square error of the whole wind farm is 21.5%, and the correlation coefficient between predicted value and measured value is 0.74.

scholarly journals A New Method to Estimate Total Organic Carbon (TOC) Content, an Example from Goldwyer Shale Formation, the Canning Basin

2017 ◽  
Vol 10 (1) ◽  
pp. 118-133 ◽  
Author(s):  
Munther Alshakhs ◽  
Reza Rezaee
Keyword(s):  
Organic Carbon ◽  
Total Organic Carbon ◽  
Multiple Regression ◽  
Gamma Ray ◽  
Source Rocks ◽  
Estimation Methods ◽  
Large Area ◽  
Canning Basin ◽  
Sonic Log ◽  
Regression Approach

Background:There is an increasing interest in the Goldwyer Formation of the Canning Basin as a potentially prospective shale play. This Ordovician shaly formation is one of the most prominent source rocks in the Canning Basin. One key property to evaluate the prospectivity of any shale oil or gas is its total organic carbon (TOC) richness.Objectives:This study investigates different TOC estimation techniques and validates the reliability of each, aiming to provide a best estimating approach for local and global applications.Method:The limited well distribution in the large area of the Canning Basin makes a basin-wide study not warranted at this stage. A focused look into the Barbwire Terrace was carried out instead. General TOC estimation methods, such as Schmoker and ∆logR were employed for TOC calculation. TOC relationships of single and multivariate regressions were also derived from wireline data and TOC rock sample measurements.Results:Both Schmoker and ∆logR methods tend to overestimate TOC when compared to the available Rock-Eval pyrolysis TOC measurements. The regression approach have shown to provide the best TOC estiamtes for wells in the Barbwire Terrace, where the best multiple regression approach for the terrace and global application was found to be the one derived from gamma-ray (GR), bulk density (RHOB), and sonic log transit time (DT).Conclusion:The generalized nature of the Schmoker method, as it provides a global relationship between density and TOC is probably the main reason why this approach does not provide a good fit in the case of the Goldwyer Formation. Furthermore, the uncertainty associated with the ∆logR method factors, such as the level of maturity (LOM), and resistivity and sonic baselines greatly influence the TOC estimation in this method, and hence, sometimes do not merit a reliable TOC estimation. The multiple regression approach have shown to be most accurate once lithology and compaction information (GR, RHOB, and DT) were incorporated in the regression process. TOC was reliably estimated for wells inside and outside the Barbwire Terrace, and also for wells of a global lacustrine shale. Such derivation have provided a more accurate technical assessment of the shale play and its prospectivity as a potential unconventional hydrocarbon resource.

Efficient Secure Neural Network Prediction Protocol Reducing Accuracy Degradation

2020 ◽  
Vol E103.A (12) ◽  
pp. 1367-1380
Author(s):  
Naohisa NISHIDA ◽  
Tatsumi OBA ◽  
Yuji UNAGAMI ◽  
Jason PAUL CRUZ ◽  
Naoto YANAI ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Network Prediction ◽  
Network Prediction
Sign in / Sign up

Export Citation Format

Share Document