Implementing Simulation and Artificial Intelligence Tools To Optimize the Performance of the CO2 Sequestration in Coalbed Methane Reservoirs
Bibliographic record
Abstract
Abstract CO2 capture and sequestration is inevitable. The concentration of the CO2 in the atmosphere is increasing continuously which will cause global warming among other consequences. Among storage options, the underground storage in depleted oil and gas reservoirs and unminable coals are considered the most economical storage options. On the other hand, natural gas consumption, which is considered to be a clean fuel, has increased significantly during the past years. Therefore seeking for new unconventional energy resources, especially gas seems to be inevitable. This goal is followed not only because of economical benefits but also because of environmental issues we are encountering these days. The purpose of this study is to develop an Artificial Neural network (ANN) tool to predict the important performance indicators such as methane recovered and CO2 injected, which are critical in CO2 storage projects in coal seams. We have combined the simulation method with artificial intelligence tools to predict the complex behavior of coal bed methane (CBM) reservoirs. In the first step a simulation is done using CMG software. A dual porosity model, which accounts for the optimum conditions during CO2 sequestration and consequently the optimum methane recovery from coal bed reservoirs was developed. Then the data extracted from the simulated CBM reservoir was employed to train the ANN model. Different parameters related to the coal seam such as porosity, permeability, initial pressure, thickness, temperature and initial water saturation are considered as the input for the network. The outputs are the CO2 injected and the recovered methane, which show the performance of the CO2 injection project. The Back-Propagation learning algorithm was used and different transfer functions and numbers of hidden layers were tried to find the best model with the least error. The tested neural network predictions were plotted versus the real data available and also different error analyses were carried out to prove the accuracy of the model. The R-Squared for the predicted values for the CO2 injected and the recovered methane were 0.92 and 0.94; the average percent arithmetic deviations were 4.8% and 4.5% respectively. INTRODUCTION In the carbon dioxide enhanced coal bed methane production/sequestration process, CO2 is injected into a coal seam to drive methane out of the bulk matrix. Because coal seams have proven to store large quantities of sorbed gases for geologic time, they exhibit significant potential for sequestration of carbon dioxide for the indefinite future. [1] There are two important parameters to consider when evaluating future CO2 sequestration in CBM reservoirs: the amount of gas that the reservoir can store, and, the potential to transport large quantities throughout the reservoir. [2] The increase in greenhouse gases in the atmosphere is one of the most important environmental issues, which leads into global warming. Increasing the efficiency of power plants or switching from coal to much more environmentally friendly fossil fuels such as natural gas are among the ways to reduce the carbon dioxide emission. [3] However, sequestration of CO2 in geological formations for an extended period of time can be one of the most promising technologies for mitigating the atmospheric CO2 concentration. Since CO2 can be naturally stored on coal surfaces, so the coal seams can be used as safe and reliable geological repositories. Coal seams are widespread and exist in many areas within the close proximity of power plants, so they are good choices for storing CO2. In recent years the attention given to the use of unmineable coal seams for sequestration purposes has progressively increased because the simultaneous recovery of natural gas helps to decrease the cost of the CO2 sequestration project. [4]
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| 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 | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".