Application of Common Contour Binning (CCB) and Back Propagation Neural Network (BPNN) for Oil Water Contact Prediction in Carbonate Reservoir (The Case Study at G404 Field)

Wahyu Tri Sutrisno, Septian Prahastudhi, Ayi Syaeful Bahri, Yulia Putri Wulandari

Indonesian Petroleum Association

... networks have an input, an output and generally a hidden layer. The number of inputs and outputs are determined by considering the characteristics...

Neural network assisted responses simulation and data correction of array laterolog in invaded formations

Yueyang Han, Lei Wang, Donghan Hao, Xiyong Yuan, Nan Wang, Zhen Yang, Jianwen Zhou, Liwei Li

International Meeting for Applied Geoscience and Energy (IMAGE)

... and output response vector, x={x1,…,xm} and y={y1,…,yn}. hj signifies the neuron value vector of j-th hidden layer. W and b are the weight matrix...

Feasibility of Moment Tensor Inversion From a Single Borehole Data Using Artificial Neural Networks; #42212 (2018)

Oleg Ovcharenko, Jubran Akram, Daniel Peter

Search and Discovery.com

... evaluation. We also investigate the effect of number of hidden layers and choice of optimization algorithm on the accuracy of inverted results. Our results...

A cross-shape deep Boltzmann machine for petrophysical seismic inversion

Son Dang Phan and Mrinal K. Sen

AAPG Bulletin

... at the vertices interconnected via a hidden neuron layer at the center. Four different input data types, including a seismic amplitude and three...

Fault zone deformation and displacement partitioning in mechanically layered carbonates: The Hidden Valley fault, central Texas

David A. Ferrill, Alan P. Morris, Ronald N. McGinnis, Kevin J. Smart, William C. Ward

AAPG Bulletin

... than the footwall (see for example, Figures 6B, D; 7). Layer dips in the hanging-wall damage zone are generally 10 or less and synthetic to the Hidden...

Forecasting Coalbed Methane Resources by Artificial Neural Network; #80335 (2013)

Yuping Yang, Jianhua Zhong, and Gangshan Lin

Search and Discovery.com

... layer, the hidden layer and the output layer. The neurons of the input layer are the known index numbers of the samples, and the neurons of the output...

Supervised Learning Applied to Rock Type Classification in Sandstone Based on Wireline Formation Pressure Data; #42539 (2020)

Jose Victor Contreras

Search and Discovery.com

..., hidden, and output layers, which are chained together and where each layer has several neurons. The loss function, which defines the feedback signal used...

A three-component field study of the effect of the low-velocity layer on converted-wave seismic data

Dan Cieslewicz, Don C. Lawton

CSPG Special Publications

...A three-component field study of the effect of the low-velocity layer on converted-wave seismic data Dan Cieslewicz, Don C. Lawton 1998 217 220...

ABSTRACT: NEURAL NETWORK-BASED COAL ACTIVITY SAFETY STANDARDIZATION ANALYSIS

Fei WU

The Society for Organic Petrology (TSOP)

... Input layer Hidden layer U1 Output layer O1 U2 O2 U3 Y O3 U4 Figure. 3. The 3-Layer BP Network Topology for Safety Standard Normalize safety...

Hidden Structure Revealed by a Simple 3D Velocity Model - McAllen Ranch Field, Hidalgo County, Texas

Richard C. Bain

GCAGS Transactions

...Hidden Structure Revealed by a Simple 3D Velocity Model - McAllen Ranch Field, Hidalgo County, Texas Richard C. Bain Hidden Structure Revealed...

ANN Method, A New Approach to Find Potential Bypass Zones in Mature Semberah Field, East Kalimantan

Gunna Satria H. Kusumah, Robhy C. Permana, Ridha S. Riadi, Yoseph R. Apranda

Indonesian Petroleum Association

... by the red blocks in Figure 9. ANN Model Optimization and Validation In order to acquire the optimal ANN model, several configurations of hidden layer...

Artificial Neural Networks for Corrosion Rate Prediction in Gas Pipelines

Sumarni, Deden Supriyatman, Kuntjoro Adjie Sidarto, Rochim Suratman, Rinaldy Dasilfa

Indonesian Petroleum Association

...), four hidden layers and one output layer (contains one node). The output layer is corrosion rate (see Figure 1). Data Processing Data sets were...

Predicting Brittleness for Wolfcamp Shales Using Statistical Rock Physics and Machine Learning; #42566 (2021)

Jaewook Lee, David Lumley

Search and Discovery.com

... to these variables for a more predictive model. In this study, we use the multi-layer feedforward neural network based on the Levenburg-Marquardt algorithm...

Coalbed Methane Production Parameter Prediction and Uncertainty Analysis of Coalbed Methane Reservoir with Artificial Neural Networks

Harry Ramadhan Sumardi, Dedy Irawan

Indonesian Petroleum Association

... HIDDEN LAYER. TABLE 3B. NEURAL NETWORK DESIGN RESULT FOR DOUBLE HIDDEN LAYER. (a) 1 Hi dden La yer Neura l  Network Archi tecture (Log‐Log) [9‐3‐1] [9‐4...

Viable Solutions to Overcome Weaknesses of Deep Learning Applications in Production Forecasting: A Comprehensive Review

Y. Kocoglu, S. Gorell

Unconventional Resources Technology Conference (URTEC)

... hidden representations (ℎ 𝑡𝑡 ) of the previous layer (𝐴𝐴 𝑙𝑙 ) are taken as inputs of the next layer (𝐴𝐴 𝑙𝑙+1 ) at each time step as shown...

Estimation of mechanical properties of sandstones from petrographic characteristics using artificial neural networks (ANNs)

Yasin Abdi, Bijan Yusefi-Yegane, Amin Jamshidi

Geological Society of Malaysia (GSM)

... layers of nodes: an input layer, a hidden layer, and an output layer. Different ANN models have been constructed using petrographic properties as inputs...

Reservoir prediction using graph-regularized deep learning

Kaiheng Sang, Nanying Lan, Fanchang Zhang

International Meeting for Applied Geoscience and Energy (IMAGE)

... is the number of samples, WE represents the weight matrix between input layer and hidden layer, also known as encoding parameter. Then, decoding operation...

Abstract: Image Processing of Seismic Lines: a Tool for Stratigraphic Interpretation, by E. L. Rossetti; #90933 (1998).

Search and Discovery.com

Some Applications and Problems of the Seismic Refraction Technique in Civil Engineering Projects in Malaysia

B. K. Lim, S. J. Jones

Geological Society of Malaysia (GSM)

... surveys. Green (1962) showed that even if the velocity of each layer increases with depth, the intermediate layer would be hidden if the thickness...

Comparison of Machine Learning and Statistical Predictive Models for Production Time Series Forecasting in Tight Oil Reservoirs

Hamid Rahmanifard, Ian Gates, Abdolmohsen Shabib-Asl

Unconventional Resources Technology Conference (URTEC)

... (ML) methods especially Artificial Neural Networks (ANNs). Cao et al. (2016) proposed an ANN (a hidden layer containing 15 neurons) model using 2...

Fault displacement gradients on normal faults and associated deformation

Alan P. Morris, Ronald N. McGinnis, and David A. Ferrill

AAPG Bulletin

...-scale normal fault. Antonellini M. Aydin A. , 1994 , Effect of faulting on fluid flow in porous sandstones: Petrophysical properties : AAPG Bulletin...

Deep learning cross-basin identification of TOC-rich zones in shale formations

Adewale Amosu, Yuefeng Sun

International Meeting for Applied Geoscience and Energy (IMAGE)

... of an input layer, a hidden layer (memory block) and an output layer. The number of neurons in the input and output layers are respectively equal...

Prediction of Fracture Porosity from Well Log Data by Artificial Neural Network, Case Study: Carbonate Reservoir, DEPOKŽ Field

Okok Wijaya, Pebrian Tunggal P, Ahmat Dafit Hasim, Depta Mahardika, Sungkono, Bagus Jaya S

Indonesian Petroleum Association

... ability of a biological brain. Neural network modeling consists of three main parameters, input layer, hidden layer and output layer. The hidden layer...

Nano-enhanced drilling fluids: Effects of MgO and ZnO nanoparticles on rheological properties and ANN modeling for predictive analysis

Moamen Gasser, Taha Yehia, Hossam Ebaid, Nathan Meehan, Omar Mahmoud

International Meeting for Applied Geoscience and Energy (IMAGE)

... for hidden layer and output layer between tansig, logsig, purelin, and relu, and 3) number in the neurons from 1 to 30 and then from 1 to 60...

Comparison of three Bayesian methods for lithofluid facies prediction using elastic properties

Jingfeng Zhang, Kevin Wolf, Anar Yusifov, Matt Walker, Pedro Paramo, Jeffrey Winterbourne, Reetam Biswas, Atish Roy, Qiang Liu, Xingchao Liu

International Meeting for Applied Geoscience and Energy (IMAGE)

... placed immediately below brine sand. Furthermore, if reliable geologic prior knowledge such as layer thickness and transition patterns is available...