Toward accurate prediction of carbon dioxide (CO2) compressibility factor using tree-based intelligent schemes (XGBoost and LightGBM) and equations of state
Bibliographic record
Abstract
Enhancing efficiency and boosting output from oil reservoirs has consistently captured the attention of engineers and industrialists within the energy sector. In recent years, there has been a notable increase in the application of enhanced oil recovery (EOR) techniques. EOR methods refer to operations which are designed in order to maximize the oil recovery factor. Among various gas mixtures that are proposed as candidates to be injected into mature oil reservoirs, CO2 gas attains miscibility with the resident hydrocarbon fluid at a reasonable pressure and increases the oil recovery factor. CO2 injection, as an EOR method, has the potential of being coupled with CO2 sequestration and reducing the emission of greenhouse gas. To design a successful CO2 injection process, it is very important to have precise knowledge about the compressibility factor (Z-factor) of CO2 as it directly affects material balance calculations, pipeline design, design of surface facilities, and CO2 compression. Z-factor, also defined as the gas deviation factor, is mathematically explained as the ratio of actual gas volume to that of an ideal gas at a given temperature and pressure. In this study, two powerful and robust tree-based machine learning (ML) algorithms, namely light gradient boosted machine (LightGBM) and extreme gradient boosting (XGBoost) were utilized to precisely estimate CO2 Z-factor. To this end, a comprehensive databank from the literature is employed, which contains 2118 data points over extensive ranges of pressures and temperatures. The proposed models predict the CO2 Z-factor with respect to reduced temperature (Tr) and reduced pressure (Pr). Moreover, the results of the developed techniques were compared with those of Patel-Teja (PT) and Peng-Robinson (PR) equations of state (EoSs) applying various graphical and statistical error tests. The results demonstrated that the LightGBM intelligent technique has the highest accuracy with the lowest error value of 0.42 % and R2 of 0.999. The trend analysis illustrated that the LightGBM model could verify the actual variation of CO2 Z-factor with pressure (direct relationship) and accurately forecast the physical behavior of the CO2 Z-factor variation. Lately, outlier detection utilizing the Leverage approach illustrated that nearly all data points, except only 39 points, were statistically reliable and located in the valid zone. The results of this research can extremely help for better understanding of CO2 sequestration, decreasing the greenhouse gas emission, and exploring EOR techniques especially CO2 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 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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| 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".