☆ 4.6 Article

Evolutionary-assisted reinforcement learning for reservoir real-time production optimization under uncertainty

PETROLEUM SCIENCE (2023)

Related references

Note: Only part of the references are listed.

Article Energy & Fuels

Training effective deep reinforcement learning agents for real-time life-cycle production optimization

Kai Zhang et al.

Summary: The paper introduces a novel production optimization method that models the life-cycle production optimization problem as a finite-horizon Markov decision process. By training deep reinforcement learning agents, it achieves maximizing net present value and realtime adjustment of well control schemes.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2022)

Add to Collection

Article Energy & Fuels

Safe reinforcement learning for real-time automatic control in a smart energy-hub

Dawei Qiu et al.

Summary: This paper proposes a model-free safe deep reinforcement learning method for optimal control of a renewable-based energy hub. It demonstrates superior performance in reducing energy cost, carbon emission, and computational time compared to state-of-the-art deep reinforcement learning and optimization-based approaches, through training and testing on a real-world dataset.

APPLIED ENERGY (2022)

Add to Collection

Article Engineering, Petroleum

Deep Reinforcement Learning for Generalizable Field Development Optimization

Jincong He et al.

Summary: This study develops an artificial intelligence using deep reinforcement learning (DRL) to provide optimized field development plans for new scenarios within seconds. The AI is trained on a wide range of geological structures, rock and fluid properties, operational constraints, and economic conditions, making it highly applicable to different scenarios.

SPE JOURNAL (2022)

Add to Collection

Article Automation & Control Systems

Affine Transformation-Enhanced Multifactorial Optimization for Heterogeneous Problems

Xiaoming Xue et al.

Summary: Evolutionary multitasking (EMT) is a new research topic that aims to improve convergence across multiple optimization tasks by facilitating knowledge transfer. Existing EMT algorithms are limited to homogeneous problems, and little effort has been made to generalize EMT for solving heterogeneous problems. This article proposes a novel rank loss function to achieve superior intertask mapping and derive an analytical solution for affine transformation. The proposed technique can seamlessly integrate with EMT paradigms, and its effectiveness is demonstrated through experiments on synthetic multitasking and many-tasking benchmark problems.

IEEE TRANSACTIONS ON CYBERNETICS (2022)

Add to Collection

Article Computer Science, Information Systems

Efficient hierarchical surrogate-assisted differential evolution for high-dimensional expensive optimization

Guodong Chen et al.

Summary: The study proposes a novel and efficient hierarchical surrogate-assisted differential evolution (EHSDE) algorithm for high-dimensional expensive optimization problems. By balancing exploration and exploitation using a hierarchical framework and utilizing global and local surrogate models to accelerate convergence speed, the algorithm demonstrates effectiveness and efficiency on benchmark functions and production optimization problems.

INFORMATION SCIENCES (2021)

Add to Collection

Article Engineering, Petroleum

Multifidelity Genetic Transfer: An Efficient Framework for Production Optimization

Faliang Yin et al.

Summary: This paper presents a novel production optimization framework based on genetic transfer learning to accelerate convergence and improve the quality of optimal solution. By combining results from high- and low-fidelity tasks, a multifidelity optimization model is established, and the concept of relative fidelity is introduced. The framework includes multitasking and sequential transfer modes to balance computing resources for tasks with different fidelity levels.

SPE JOURNAL (2021)

Add to Collection

Article Energy & Fuels

A double-model differential evolution for constrained waterflooding production optimization

Kai Zhang et al.

Summary: In this paper, a novel doublemodel differential evolution (CSDE) algorithm is proposed for waterflooding production optimization with nonlinear inequality constraints, which integrates support vector machine (SVM) and radial basis function (RBF) into evolutionary algorithm. Experimental results demonstrate that the CSDE algorithm can handle constraints more efficiently and achieve higher net present values compared with original evolutionary algorithms and other single-model algorithms.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2021)

Add to Collection

Article Multidisciplinary Sciences

Embodied intelligence via learning and evolution

Agrim Gupta et al.

Summary: The new framework of deep evolutionary reinforcement learning evolves agents with diverse morphologies to learn difficult tasks in complex environments, revealing insights into the relationships between environmental physics, embodied intelligence, and rapid learning evolution. Research indicates that environmental complexity fosters the evolution of morphological intelligence, and that more physically stable and energy efficient morphologies facilitate learning and control. Additionally, a morphological Baldwin effect is demonstrated, where evolution rapidly selects morphologies that learn faster, impacting the behaviors learned and expressed throughout generations.

NATURE COMMUNICATIONS (2021)

Add to Collection

Article Engineering, Petroleum

A Recurrent Neural Network-Based Proxy Model for Well-Control Optimization with Nonlinear Output Constraints

Yong Do Kim et al.

Summary: In this study, a recurrent neural network (RNN)-based proxy model was developed for constrained production optimization problems, which was validated for 2D and 3D production optimization. By incorporating a few additional simulation-based iterations after proxy-based optimization, comparable net present values (NPVs) to simulation-based optimization were achieved with significant speedups. The RNN-based proxy provides full well-rate time sequences, allowing for flexibility in varying optimization constraints and economic parameters without the need for retraining.

SPE JOURNAL (2021)

Add to Collection

Article Energy & Fuels

Optimal carbon storage reservoir management through deep reinforcement learning

Alexander Y. Sun

APPLIED ENERGY (2020)

Add to Collection

Article Energy & Fuels

A comprehensive workflow for real time injection-production optimization based on equilibrium displacement

Huijiang Chang et al.

ADVANCES IN GEO-ENERGY RESEARCH (2020)

Add to Collection

Article Energy & Fuels

A comparative evaluation of global search algorithms in black box optimization of oil production: A case study on Brugge field

Toomaj Foroud et al.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2018)

Add to Collection

Article Engineering, Petroleum

Robust Life-Cycle Production Optimization With a Support-Vector-Regression Proxy

Zhenyu Guo et al.

SPE JOURNAL (2018)

Add to Collection

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)

Add to Collection

Article Engineering, Petroleum

A Physics-Based Data-Driven Numerical Model for Reservoir History Matching and Prediction With a Field Application

Hui Zhao et al.

SPE JOURNAL (2016)

Add to Collection

Article Engineering, Petroleum

Ensemble-Based Optimization of the Water-Alternating-Gas-Injection Process

Bailian Chen et al.

SPE JOURNAL (2016)

Add to Collection

Article Energy & Fuels

Metaheuristic profiling to assess performance of hybrid evolutionary optimization algorithms applied to complex wellbore trajectories

David A. Wood

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2016)

Add to Collection

Article Energy & Fuels

Sperm whale algorithm: An effective metaheuristic algorithm for production optimization problems

A. Ebrahimi et al.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2016)

Add to Collection

Article Multidisciplinary Sciences

Human-level control through deep reinforcement learning

Volodymyr Mnih et al.

NATURE (2015)

Add to Collection

Article Statistics & Probability

KULLBACK-LEIBLER UPPER CONFIDENCE BOUNDS FOR OPTIMAL SEQUENTIAL ALLOCATION

Olivier Cappe et al.

ANNALS OF STATISTICS (2013)

Add to Collection

Article Computer Science, Artificial Intelligence

3D Convolutional Neural Networks for Human Action Recognition

Shuiwang Ji et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2013)

Add to Collection

Article Energy & Fuels

Optimal Parametric Design for Water-Alternating-Gas (WAG) Process in a CO2-Miscible Flooding Reservoir

S. Chen et al.

JOURNAL OF CANADIAN PETROLEUM TECHNOLOGY (2010)

Add to Collection

Article Energy & Fuels

Production optimization with adjoint models under nonlinear control-state path inequality constraints

Pallav Sarma et al.

SPE RESERVOIR EVALUATION & ENGINEERING (2008)

Add to Collection

© Peeref 2019-2024. All rights reserved.