The new machine learning based approach for permeability evaluation in horizontal well initial flow rate calculation while drilling

This study focuses on exploring the potential application of logging-while-drilling data to estimate permeability, which is essential for calculating the initial production rate and profile of horizontal wells with the intention to optimize well trajectory and improve its economic efficiency in real time. The dataset comprises information from horizontal wells drilled by Rosneft Oil Company between 2022 and 2024, including las-files with logging-while-drilling records and history-matched on actual initial production rates averaged along the well horizontal part absolute permeability values. To tackle this challenge, several machine learning models were developed, employing both classification and regression techniques. History-matched on actual initial production rates average well permeability was selected as the target variable, with number of averaged logging curve features serving as input parameters. The best-performing model demonstrated the ability to predict permeability required for estimating the initial production rate of horizontal wells with a mean error of 63 % and a median error of 24 %. As the result of the study, it is clearly shown that this approach to history-matched eligible for initial production rate calculation permeability estimation while drilling, utilising machine learning tools is promising. Furthermore, even at this stage the developed methodology already shows potential for enhancing technological and economic efficiency of Rosneft’s Oil Company new wells drilling program.

References

1. Nwulu E.O. et al., Machine learning applications in predictive maintenance: Enhancing efficiency across the oil and gas industry, International Journal of Engineering Research Updates, 2023, no. 5(1), pp. 13–27, DOI: https://doi.org/10.53430/ijeru.2023.5.1.0017

2. Lawal A. et al., Machine learning in oil and gas exploration: A review, IEEE Access, 2024, no. 12, pp. 19035–19058, DOI: https://doi.org/10.1109/access.2023.3349216

3. Choubey S., Karmakar G.P., Artificial intelligence techniques and their application in oil and gas industry, Artificial Intelligence Review, 2020, V. 54(04), pp. 1–19,

DOI: https://doi.org/10.1007/s10462-020-09935-1

4. Elkatatny S., Mahmoud M., Tariq Z., Abdulraheem A., New insights into the prediction of heterogeneous carbonate reservoir permeability from well logs using artificial intelligence network, Neural Comput & Applic., 2017, DOI: https://doi.org/10.1007/s00521-017-2850-x

5. Azbukhanov A.F., Kostrigin I.V., Bondarenko K.A. et al., Selection of wells for hydraulic fracturing based on mathematical modeling using machine learning methods

(In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2019, no. 11, pp. 38–42, DOI: https://doi.org/10.24887/0028-2448-2019-11-38-42

6. Yige Sun et al., Deep learning versus conventional methods for missing data imputation: A review and comparative study, Expert Systems with Applications, 2023,

V. 227, DOI: https://doi.org/10.1016/j.eswa.2023.120201

7. Breiman L., Friedman J.H., Olshen R.A., Stone C.J., Classification and regression trees, New York: Chapman and Hall/CRC, 1984, 368 r.,

DOI: https://doi.org/10.1201/9781315139470

8. Quinlan J.R., Induction of decision trees. Machine Learning, Kluwer Academic Publishers, 1986, No. 1, pp. 81–106, DOI: https://doi.org/10.1023/A:1022643204877

9. Kudashov K.V., Antsupov V.E., Akimova D.A., New wells production rate forecast quality improvement due to calculated values achievement for objects with different geological, geophysical and technological parameters (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2023, no. 8, pp. 128–130, DOI: https://doi.org/10.24887/0028-2448-2023-8-128-130

10. Helle H.B., Bhatt A., Ursin B., Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study, Geophysical Prospecting, 2001, no. 49, pp. 431–444, DOI: https://doi.org/10.1046/j.1365-2478.2001.00271.x

11. Shokir E.M., Permeability estimation from well log responses, Journal of Canadian Petroleum Technology, 2006, V. 45(11), pp. 41–46,

DOI: https://doi.org/10.2118/06-11-05