4.6 Article

Data-driven model predictive control for closed-loop refracturing design and optimization in naturally fractured shale gas reservoir under geological uncertainty

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Energy & Fuels

Data-driven model for hydraulic fracturing design optimization. Part II: Inverse problem

V. M. Duplyakov et al.

Summary: This study developed a stacked model using a combination of Ridge Regression and CatBoost algorithms to predict cumulative fluid production for multistage-fracture completion oil wells. The model was based on an extensive digital field data base and demonstrated physically explainable dependencies plots of the target on the design parameters. Various methods were developed for selection of offset wells and optimization of fracturing design parameters, ultimately leading to a recommendation system for production stimulation engineers.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2022)

Article Environmental Sciences

Model-Reduced Adjoint-Based Inversion Using Deep-Learning: Example of Geological Carbon Sequestration Modeling

Cong Xiao et al.

Summary: In this study, a two-level hybrid neural network architecture is proposed to efficiently derive a model-reduced adjoint-based optimization workflow for large-scale inverse modeling in geological carbon-storage applications. The first level utilizes a deep convolutional autoencoder for nonlinear dimensionality reduction, while the second level emulates a neural-network-powered linear transition unit based on the reduced subspace. The experimental results demonstrate the feasibility and accuracy of the proposed method in capturing the spatial-temporal evolution of the CO2 saturation fields using a relatively small number of training data.

WATER RESOURCES RESEARCH (2022)

Article Energy & Fuels

Machine-learning-based well production prediction under geological and hydraulic fracture parameters uncertainty for unconventional shale gas reservoirs

Cong Xiao et al.

Summary: This study proposes an efficient and robust post-history production forecasting framework using latent-space learning. By employing a convolutional autoencoder to regularize multi-well time-series data, this method achieves high robustness and computational efficiency in shale gas production prediction.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2022)

Article Energy & Fuels

Integrated optimization design for horizontal well spacing and fracture stage placement in shale gas reservoir

Lian Wang et al.

Summary: Horizontal well drilling and hydraulic fracturing technologies are crucial for improving gas recovery from shale reservoirs. This study proposes a novel optimization method called WSF-MFSVR, which integrates multi-fidelity support vector regression surrogate model and particle swarm optimization algorithm, to optimize horizontal well spacing and fracture stage placement. The results show that the WSF-MFSVR method is computationally efficient and produces superior convergence and diversity in obtaining optimal solutions.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2022)

Article Energy & Fuels

Optimization design of horizontal well fracture stage placement in shale gas reservoirs based on an efficient variable-fidelity surrogate model and intelligent algorithm

Guoxiang Zhao et al.

Summary: The study proposed a novel intelligent variable-fidelity radial basis function (VFRBF) surrogate-assisted model to reduce the computational burden of numerical simulation-based production optimization. The method showed efficiency and accuracy in shale gas horizontal well fracture optimization, and performed competitively compared to high-fidelity model-based methods.

ENERGY REPORTS (2022)

Article Environmental Sciences

Deep-Learning-Based Adjoint State Method: Methodology and Preliminary Application to Inverse Modeling

Cong Xiao et al.

Summary: An efficient adjoint model based on deep learning surrogate was presented to address high-dimensional inverse modeling, providing a nonintrusive way to approximate model derivatives and compute gradients for inverse modeling. By combining autoencoder and autoregressive structures in a deep convolutional neural network architecture, the proposed method efficiently generates an analytical adjoint with the help of auto-differentiation modules in popular deep learning packages. The proposed adjoint state method was tested on a synthetic two-dimensional model for parameter estimations, showing potential feasibility in terms of computational efficiency and programming flexibility.

WATER RESOURCES RESEARCH (2021)

Article Engineering, Civil

Treatment of model error in subsurface flow history matching using a data-space method

Su Jiang et al.

Summary: Subsurface flow models are imperfect and model error can lead to unreliable predictions. This study introduces a method to handle model error in a data-space inversion framework, showing that the coupled model error treatment provides highly accurate results.

JOURNAL OF HYDROLOGY (2021)

Article Energy & Fuels

Machine learning based rate optimization under geologic uncertainty

Deepthi Sen et al.

Summary: The novel approach proposed uses machine learning for rate optimization over multiple geologic realizations, improving injection sweep efficiency. By parsimoniously representing geologic heterogeneity and combining it with ML for forward modeling, the approach is scalable for large field applications.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2021)

Article Engineering, Petroleum

AI-Based Estimation of Hydraulic Fracturing Effect

A. S. Erofeev et al.

Summary: The study utilized a gradient-boosting machine-learning algorithm to forecast oil and total liquid production after hydraulic fracturing, showing promising results. Data preprocessing and feature engineering were conducted, and the trained model accurately predicted production rates for different time periods.

SPE JOURNAL (2021)

Article Energy & Fuels

An improved data space inversion method to predict reservoir state fields via observed production data

Deng Liu et al.

Summary: The DSI method, based on Bayesian theory, is used for reservoir history matching and production prediction, obtaining accurate flow predictions without inverting geological parameters. By expanding the data space and efficient history matching, this method accurately predicts oil saturation and pressure profiles of real reservoir models, showing high computational efficiency and accuracy.

PETROLEUM SCIENCE (2021)

Article Mathematics, Interdisciplinary Applications

Data-Space Inversion With a Recurrent Autoencoder for Naturally Fractured Systems

Su Jiang et al.

Summary: The paper introduces the use of Data-space inversion (DSI) with recurrent autoencoder (RAE) parameterization for flow simulation in a naturally fractured reservoir. RAE proves to be more accurate in data reconstruction and captures more important correlations in time-series data compared to traditional methods.

FRONTIERS IN APPLIED MATHEMATICS AND STATISTICS (2021)

Review Energy & Fuels

Multiphase Flow in Highly Fractured Shale Gas Reservoirs: Review of Fundamental Concepts for Numerical Simulation

Maxian B. Seales

JOURNAL OF ENERGY RESOURCES TECHNOLOGY-TRANSACTIONS OF THE ASME (2020)

Article Energy & Fuels

Data-driven model for hydraulic fracturing design optimization: focus on building digital database and production forecast

Anton D. Morozov et al.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2020)

Article Energy & Fuels

Production Performance Evaluation of Multifractured Horizontal Wells in Shale Oil Reservoirs: An Analytical Method

Xingbang Meng et al.

JOURNAL OF ENERGY RESOURCES TECHNOLOGY-TRANSACTIONS OF THE ASME (2019)

Article Engineering, Chemical

Economic Model-Based Controller Design Framework for Hydraulic Fracturing To Optimize Shale Gas Production and Water Usage

Kaiyu Cao et al.

INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH (2019)

Article Computer Science, Interdisciplinary Applications

Deterministic ensemble smoother with multiple data assimilation as an alternative for history-matching seismic data

Alexandre A. Emerick

COMPUTATIONAL GEOSCIENCES (2018)

Article Computer Science, Interdisciplinary Applications

Modeling of hydraulic fracturing and designing of online pumping schedules to achieve uniform proppant concentration in conventional oil reservoirs

Prashanth Siddhamshetty et al.

COMPUTERS & CHEMICAL ENGINEERING (2018)

Article Environmental Sciences

Tuning Fractures With Dynamic Data

Mengbi Yao et al.

WATER RESOURCES RESEARCH (2018)

Article Engineering, Multidisciplinary

A Stochastic Simplex Approximate Gradient (StoSAG) for optimization under uncertainty

Rahul Rahul-Mark Fonseca et al.

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING (2017)

Article Geosciences, Multidisciplinary

A New Data-Space Inversion Procedure for Efficient Uncertainty Quantification in Subsurface Flow Problems

Wenyue Sun et al.

MATHEMATICAL GEOSCIENCES (2017)

Article Energy & Fuels

A comprehensive overview on recent developments in refracturing technique for shale gas reservoirs

Maunish Shah et al.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2017)

Article Computer Science, Interdisciplinary Applications

Production forecasting and uncertainty quantification for naturally fractured reservoirs using a new data-space inversion procedure

Wenyue Sun et al.

COMPUTATIONAL GEOSCIENCES (2017)

Article Water Resources

Projection-based Embedded Discrete Fracture Model (pEDFM)

Matei Tene et al.

ADVANCES IN WATER RESOURCES (2017)

Article Computer Science, Interdisciplinary Applications

Optimal pumping schedule design to achieve a uniform proppant concentration level in hydraulic fracturing

Seeyub Yang et al.

COMPUTERS & CHEMICAL ENGINEERING (2017)

Article Engineering, Chemical

Optimization models for planning shale gas well refracture treatments

Diego C. Cafaro et al.

AICHE JOURNAL (2016)

Article Computer Science, Interdisciplinary Applications

Surrogate-based optimization of hydraulic fracturing in pre-existing fracture networks

Mingjie Chen et al.

COMPUTERS & GEOSCIENCES (2013)

Article Energy & Fuels

Factors Affecting Hydraulically Fractured Well Performance in the Marcellus Shale Gas Reservoirs

Tunde Osholake et al.

JOURNAL OF ENERGY RESOURCES TECHNOLOGY-TRANSACTIONS OF THE ASME (2013)

Article Geosciences, Multidisciplinary

Fracture development in shale and its relationship to gas accumulation

Wenlong Ding et al.

GEOSCIENCE FRONTIERS (2012)

Article Engineering, Electrical & Electronic

Non-dominated sorting genetic algorithm-II for robust multi-objective optimal reactive power dispatch

L. Zhihuan et al.

IET GENERATION TRANSMISSION & DISTRIBUTION (2010)

Article Computer Science, Interdisciplinary Applications

History matching of facies distribution with the EnKF and level set parameterization

Haibin Chang et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2010)

Article Geosciences, Multidisciplinary

Improving the Ensemble Estimate of the Kalman Gain by Bootstrap Sampling

Yanfen Zhang et al.

MATHEMATICAL GEOSCIENCES (2010)