Optimization of Recovery by Huff and Puff Gas Injection in Shale Oil Reservoirs Using the Climbing Swarm Derivative Free Algorithm
Bibliographic record
Abstract
Abstract Recent improved and enhanced oil recovery (IOR and EOR) methods in shale reservoirs use huff and puff gas injection (H&P). Investigating the technical and economic impact of this technology for one well is challenging and time consuming. Even more so when the petroleum company is planning H&P and refracturing (RF) jobs in multiple wells. Thus, in this paper we present an original methodology to learn how to perform these tasks faster and at lower cost to improve oil recovery. The procedure is explained with the use of an actual H&P gas injection pilot horizontal well in the Eagle Ford shale whose performance is matched using the methodology developed in this paper. The methodology includes use of an original Climbing Swarm (CS) derivative-free algorithm that drives, without human intervention, computer or laptop material balance (MatBal) and net present value (NPV) calculations. The code was written in Python. Following history match, the methodology demonstrates that significant improvements in oil recovery can be obtained by injecting gas at larger rates during shorter periods of time (as opposed to injecting gas at smaller rates during longer periods of time). Once oil recovery improvement in a pilot horizontal well is demonstrated, the methodology is extended to the analysis of H&P gas injection and refracturing in horizontal wells and shale reservoirs that have not yet been developed or are in initial stages of development; this provides a preliminary assessment of H&P and refracturing potential. Results indicate that oil recovery and NPV from multiple wells can be improved significantly by a strategic combination of H&P gas injection and refracturing. Combination of derivative-free optimization algorithms, MatBal calculations and net present value permits optimizing when to start the H&P gas injection project, the optimum gas injection rates and time-span of injection, reservoir pressure at which gas injection should be initiated in each cycle, and the time-span during which the well should produce oil, previous to starting a new cycle of gas injection. The development strategy of shale oil reservoirs could be improved significantly if the possibility of H&P gas injection is considered previous to field development. This could be the case of the Eagle Ford shale in Mexico, La Luna shale in Colombia and Venezuela, Vaca Muerta shale in Argentina and other shale oil reservoirs worldwide. The paper contributes the development of an original methodology, which includes use of a derivative free algorithm we call "Climbing Swarm (CS)." CS drives the computer or laptop to perform MatBal and NPV calculations, without human intervention, once the optimization process is started. The methodology improves oil recovery and NPV from a single horizontal well or from multiple horizontal wells operating under H&P gas injection.
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 | 0.001 |
| 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".