1-20 of 273
Keywords: neural network
Close
Sort by
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214888-MS
... & gas chatgpt geologist neural network united states government natural language cypher query information query node generated cypher query database openai region artificial intelligence snippet gpt-4 graphqa petrowiki snippet machine learning experiment petroleum engineer society...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214889-MS
... geologist sedimentary rock united states government neural network geology realization clastic rock reservoir simulation upstream oil & gas deep learning reservoir geomechanics machine learning prediction artificial intelligence dataset z-direction application reservoir...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214883-MS
... and qualitative visualization. In addition, this method is compared with other methods, such as GAN, Variational Autoencoder, and Super-Resolution Convolutional Neural Networks (SRCNN). The results indicate that the built model shows excellent potential in enhancing the resolution of heterogeneous carbonate rocks...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214922-MS
... to deal with image-based inputs. The convolutional neural network (CNN) can be adapted to image-to-value problems. Santoso et al. (2019) used CNN to identify the fractures from outcrop images. He et al. (2020) utilized CNN to construct the equivalent continuum models based on discrete fracture models...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215019-MS
... necessitate prior low-frequency models, and are not only susceptible to noise but also prone to introducing biases due to their dependence on initial assumptions. In recent years, deep learning, using deep neural networks (DNNs), has emerged as a competitive approach for seismic inversion, capable...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214985-MS
... realizations. In this work, we develop the sequential patch generative adversarial neural network (GAN), a computationally efficient method to perform machine learning- and patch-based, sequential subsurface modeling. The new machine learning method uses shift-invariant neural network structures to allow...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215028-MS
... learning regression models such as Artificial Neural Networks (ANN) and Random Forest (RF). While these models have their merits and can be useful, they also exhibit notable limitations. Some (ARIMA) fail to consider the influence of multiple parameters or can only perform single-step or short-term...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214818-MS
... traditional solvers that are more expensive to run. This can be addressed by adopting a hybrid workflow (e.g., using the proxy model to narrow down the design space and then using the physics-based model to validate the final decision). neural network seg unconventional resource technology conference...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215099-MS
..., we develop a new network architecture for automatic seismic facies class with attention mechanism for improved classification results. geology geologist upstream oil & gas sedimentary rock united states government neural network reservoir characterization information natural...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214769-MS
... of related time series and the uncertainty can be assessed. The objective of this work is to enable global modeling and probabilistic forecasting of a large number of related production time series using Deep Autoregressive Recurrent Neural Networks (DeepAR). The DeepAR model consists of three parts. First...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214881-MS
...-edge deep learning models were utilized. These models included Autoregressive Integrated Moving Average (ARIMA), Block Recurrent Neural Network (BlockRNN), Temporal Fusion Transformer (TFT), and the Neural Basis Expansion Analysis for Interpretable Time Series Forecasting (N-BEATS) using meta learning...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214831-MS
..., most of them focused on experimental methods. In this work, we proposed a new approach for recovering geometrical information of the stylolite zone (including its size and location) based on neural network architectures including convolutional neural network (CNN), recurrent neural network (RNN...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215103-MS
... simulation results at a grid level with less than 10 percent error. This approach offers significant potential in accelerating reservoir simulation processes and optimizing oil and gas recovery strategies. Neural Network Architecture upstream oil & gas fluid dynamics unconventional play asia...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215072-MS
... results. To address these limitations, we propose a Deep Convolutional Neural Networks (DCNN) based workflow that leverages context-specific algorithms for accurate subsurface-driven infill planning optimization. The framework was validated on real-world data from 2 fields of 630 wells in the Permian...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215117-MS
... ; Reinhardt et al., 2022 ). Particularly, encoder-decoder models based on convolution neural networks (CNNs) have shown great promise in overcoming the limitations of traditional methods, which have set a compelling precedent for their use in DRP studies ( Asgari et al., 2021 ; Niu et al., 2020...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-215085-MS
... operation physics-based machine learning model drilling operation algorithm rock type xgb coefficient deviation estimation neuron neural network svm real-time prediction petroleum engineer engineering artificial neural network dataset funnel viscosity application plastic viscosity...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 16–18, 2023
Paper Number: SPE-214782-MS
... and adopted to train a one-dimensional convolutional neural network (1D CNN) and the optimal convolutional neural network (CNN) is obtained by minimizing mean square error. In the CNN, the wellbore pressure is used as input of the network after nondimensionalization, and the interpreted parameters...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 3–5, 2022
Paper Number: SPE-209959-MS
...-scale GCS deployment. In this work, we introduce a deep-learning-based algorithm using a hybrid neural network for detecting CO 2 leakage based on bottom-hole pressure measurements. The proposed workflow includes the generation of train-validation samples, the coupling process of training-validating...
Proceedings Papers

Paper presented at the SPE Annual Technical Conference and Exhibition, October 3–5, 2022
Paper Number: SPE-210061-MS
... opportunities for dimension reduction. Purely data-driven methods exist, such as principal component analysis and autoencoder neural network, but the discovered latent variables are generally not conducive to physics-based interpretation. Hence, in this study, we focus on down-selecting variables from a pool...

Product(s) added to cart

Close Modal