Uso de Métodos de Aprendizado de Máquina e Algoritmo Genético para Predição de TOC e Classificação de Litologia

Juliana da Costa Cabral; Clovis Antonio da Silva; Grazione de Souza; Camila Martins Saporetti

doi:10.63595/vetor.v35i1.18357

Authors

Juliana da Costa Cabral Instituto Politécnico - Universidade do Estado do Rio de Janeiro
Clovis Antonio da Silva Instituto Politécnico - Universidade do Estado do Rio de Janeiro
Grazione de Souza Instituto Politécnico - Universidade do Estado do Rio de Janeiro https://orcid.org/0000-0002-4840-4472
Camila Martins Saporetti Instituto Politécnico - Universidade do Estado do Rio de Janeiro https://orcid.org/0000-0002-8145-7074

DOI:

https://doi.org/10.63595/vetor.v35i1.18357

Keywords:

Machine Learning, Genetic Algorithm, Lithology, Total Organic Carbon

Abstract

Oil and gas are the main sources of primary energy in the world. From these resources, derivatives and petrochemicals are obtained that feed the production of energy, services and various products. Among the crucial stages of oil production are the classification of reservoirs, drilling and analysis of geological data to determine the feasibility of extraction. However, these processes are often done manually by experts or using methods that are expensive, inaccurate and time-consuming. In this context, this work aims to classify lithologies and predict the total organic carbon rate through the application of machine learning techniques, employing a genetic algorithm with exhaustive search to optimize regression/classification methods. The database used refers to a well in Campo Marlim, Campos Basin. The results show that Extreme Gradient Boosting (XGB) performed well in the experiments carried out, with average accuracy = 0.941 and RMSE = 0.150 in the test set, being an alternative to assist specialists in the task of lithology classification and rate prediction. total carbon.

Downloads

Download data is not yet available.

References

[1] N. Ezekwe, Petroleum Reservoir Engineering Practice. Westford, USA: Prentice Hall, 2010.

[2] N. Al-Mohannadi, “Simulation of horizontal well tests,” Tese de doutorado, Colorado School of Mines, Golden, USA, 2004.

[3] A. J. Rosa, R. S. Carvalho, e J. A. D. Xavier, Engenharia de Reservatórios de Petróleo. Interciência, 2006.

[4] M. Mohammadpour, H. Roshan, M. Arashpour, e H. Masoumi, “Machine learning assisted Kriging to capture spatial variability in petrophysical property modelling,” Marine and Petroleum Geology, vol. 167, p. 106967, 2024. Disponível em: https://doi.org/10.1016/j.marpetgeo.2024.106967

[5] C. M. Saporetti, L. Goliatt, e E. Pereira, “Neural network boosted with differential evolution for lithology identification based on well logs information,” Earth Science Informatics, vol. 14, no. 1, pp. 133–140, 2021. Disponível em: https://doi.org/10.1007/s12145-020-00533-x

[6] E. M. Vasini, A. Battistelli, P. Berry, S. Bonduà, V. Bortolotti, C. Cormio, e L. Pan, “Interpretation of production tests in geothermal wells with T2Well-EWASG,” Geothermics, vol. 73, pp. 158–167, 2018. Disponível em:https://doi.org/10.1016/j.geothermics.2017.06.005

[7] G. van Graas, T. Viets, J. De Leeuw, e P. Schenck, “A study of the soluble and insoluble organic matter from the Livello Bonarelli, a cretaceous black shale deposit in the Central Apennines, Italy,” Geochimica et Cosmochimica Acta, vol. 47, no. 6, pp. 1051–1059, 1983. Disponível em: https://doi.org/10.1016/0016-7037(83)90235-1

[8] H. Yang, H. Pan, H. Ma, A. A. Konaté, J. Yao, e B. Guo, “Performance of the synergetic wavelet transform and modified K-means clustering in lithology classification using nuclear log,” Journal of Petroleum Science and Engineering, vol. 144, pp. 1–9, 2016. Disponível em: https://doi.org/10.1016/j.petrol.2016.02.031

[9] S. Elkatatny, “A self-adaptive artificial neural network technique to predict total organic carbon (TOC) based on well logs,” Arabian Journal of Science and Engineering, vol. 44, pp. 6127–6137, 2018. Disponível em: https://doi.org/10.1007/s13369-018-3672-6

[10] Y. Xie, C. Zhu, W. Zhou, Z. Li, X. Liu, e M. Tu, “Evaluation of machine learning methods for formation lithology identification: A comparison of tuning processes and model performances,” Journal of Petroleum Science and Engineering, vol. 160, pp. 182–193, 2018. Disponível em: https://doi.org/10.1016/j.petrol.2017.10.028

[11] C. M. Saporetti, L. G. da Fonseca, E. Pereira, e L. C. de Oliveira, “Machine learning approaches for petrographic classification of carbonate-siliciclastic rocks using well logs and textural information,” Journal of Applied Geophysics, vol. 155, pp. 217–225, 2018. Disponível em: https://doi.org/10.1016/j.jappgeo.2018.06.012

[12] S. Asante-Okyere, Y. Y. Ziggah, e S. A. Marfo, “Improved total organic carbon convolutional neural network model based on mineralogy and geophysical well log data,” Unconventional Resources, vol. 1, pp. 1–8, 2021. Disponível em: https://doi.org/10.1016/j.uncres.2021.04.001

[13] C. Saporetti, D. Fonseca, L. Oliveira, E. Pereira, e L. Goliatt, “Hybrid machine learning models for estimating total organic carbon from mineral constituents in core samples of shale gas fields,” Marine and Petroleum Geology, vol. 143, p. 105783, 2022. Disponível em: https://doi.org/10.1016/j.marpetgeo.2022.105783

[14] R. O. Silva, C. M. Saporetti, Z. M. Yaseen, E. Pereira, e L. Goliatt, “An approach for total organic carbon prediction using convolutional neural networks optimized by differential evolution,” Neural Computing and Applications, vol. 35, pp. 20 803–20 817, 2023. Disponível em: https://doi.org/10.1007/s00521-023-08865-7

[15] A. de Carvalho, A. Menezes, e R. Bonidia, Ciência de Dados - Fundamentos e Aplicações. Disponível em: https://books.google.com.br/books?id=HNSn0AEACAAJ LTC, 2024.

[16] K. Faceli, A. C. Lorena, J. Gama, T. A. d. Almeida, e A. C. P. L. F. Carvalho, Inteligência artificial: uma abordagem de aprendizado de máquina. LTC, 2021.

[17] G.-B. Huang, Q.-Y. Zhu, e C.-K. Siew, “Extreme learning machine: theory and applications,” Neurocomputing, vol. 70, no. 1-3, pp. 489–501, 2006. Disponível em: https://doi.org/10.1016/j.neucom.2005.12.126

[18] G.-B. Huang, Q.-Y. Zhu, e C.-K. Siew, “Extreme learning machine: a new learning scheme of feedforward neural networks,” em 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No. 04CH37541), vol. 2. IEEE, 2004, pp. 985–990. Disponível em: https://doi.org/10.1109/IJCNN.2004.1380068

[19] C. Cortes e V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273–297, 1995. Disponível em: https://doi.org/10.1007/BF00994018

[20] V. N. Vapnik, The Nature of Statistical Learning Theory. Springer New York, 2000. Disponível em: https://doi.org/10.1007/978-1-4757-3264-1

[21] H. Drucker, C. J. Burges, L. Kaufman, A. Smola, e V. Vapnik, “Support vector regression machines,” Advances in Neural Information Processing Systems, vol. 9, 1996.

[22] E. Fix e J. L. Hodges, “Discriminatory analysis, nonparametric discrimination,” 1951.

[23] T. Cover e P. Hart, “Nearest neighbor pattern classification,” IEEE Transactions on Information Theory, vol. 13, no. 1, pp. 21–27, 1967. Disponível em: https://doi.org/10.1109/TIT.1967.1053964

[24] R. Souza, R. Lotufo, e L. Rittner, “A comparison between optimum-path forest and k-nearest neighbors classifiers,” em 2012 25th SIBGRAPI Conference on Graphics, Patterns and Images. IEEE, 2012, pp. 260–267. Disponível em: https://doi.org/10.1109/SIBGRAPI.2012.43

[25] T. Chen e C. Guestrin, “Xgboost: A scalable tree boosting system,” em Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794. Disponível em: https://doi.org/10.1145/2939672.2939785

[26] R. E. Schapire, “The strength of weak learnability,” Machine Learning, vol. 5, pp. 197–227, 1990. Disponível em: https://doi.org/10.1007/BF00116037

[27] F. R. Bione, I. M. Venancio, T. P. Santos, A. L. Belem, B. R. Rangel, I. V. Souza, A. L. Spigolon, e A. L. S. Albuquerque, “Estimating total organic carbon of potential source rocks in the Espírito Santo basin, SE Brazil, using XGBoost,” Marine and Petroleum Geology, vol. 162, p. 106765, 2024. Disponível em: https://doi.org/10.2139/ssrn.4631704

[28] R. A. Gómez, GASearchCV, 2021. Disponível em: https://sklearn-genetic-opt.readthedocs.io/en/stable/