Imitation-Guided Reinforcement Learning with Adaptive Sampling and Expert Knowledge for Real-Time Well Control Optimization
Bibliographic record
Abstract
Summary Deep reinforcement learning (DRL) has been increasingly applied to real-time well control optimization but remains hindered by low sampling efficiency and weak early-stage exploration. Even though neural networks excel in learning, they lack search capacity, while evolutionary algorithms (EAs) provide strong global search but no learning capability. In this paper, we propose an adaptive dual-experience replay framework that integrates the advantages of EAs and reinforcement learning (RL) to improve the efficiency and accuracy of real-time well control optimization. High-quality samples are generated through EAs and used to guide the policy learning of the DRL agent via imitation learning (IL), effectively realizing the warm start of policy learning. Meanwhile, DRL’s autonomous exploration capability is preserved, enabling a collaborative optimization process that combines experience transfer and self-directed learning. An adaptive sampling strategy further enhances training by dynamically adjusting the contributions of experiences from the EA and DRL buffers, improving policy stability and generalization. This study focuses specifically on injection/production rate and bottomhole pressure (BHP) optimization in benchmark reservoir models. Experimental results show that the proposed method outperforms baseline approaches in net present value (NPV) and convergence speed. Additionally, robustness tests demonstrate that the trained policy maintains strong performance when directly applied to unseen scenarios. Notably, it achieves comparable performance to the state-of-the-art soft actor-critic (SAC) algorithm using only half the training episodes, underscoring its potential for intelligent oilfield development.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".