Skip to main content

Advertisement

SpringerLink
Log in

Group Method of Data Handling (GMDH) Neural Network for Estimating Total Organic Carbon (TOC) and Hydrocarbon Potential Distribution (S1, S2) Using Well Logs

  • Original Paper
  • Published:
Natural Resources Research Aims and scope Submit manuscript

Abstract

Recent advancement in computing capabilities has brought to light the application of machine learning methods in estimating geochemical data from well logs. The widely employed artificial neural network (ANN) has intrinsic problems in its application. Therefore, the objective of this study was to present a group method of data handling (GMDH) neural network as an improved alternative in predicting total organic carbon (TOC), S1, and S2 from well logs. The study used bulk density, sonic travel time, deep lateral resistivity log, gamma-ray, spontaneous potential, neutron porosity well logs as input variables to predict TOC, S1, and S2 of the Nondwa, Mbuo, and Mihambia Formations in the Triassic to mid-Jurassic of the Mandawa Basin in southeast Tanzania. The TOC prediction results indicated that the GMDH model trained well while generalizing better across the testing data than both ANN and ΔlogR. Specifically, the GMDH provided TOC testing predictions having the least errors of 0.40 and 0.45 for mean square error (MSE) and mean absolute error (MAE), respectively, as compared to 1.27 and 0.81, 0.68 and 0.7, 1.4 and 0.89 obtained by backpropagation neural network (BPNN), radial basis function neural network (RBFNN), and ΔlogR, respectively. For S1 and S2, the ANN models performed excellently during training but were unable to produce similar results when tested on the completely unseen well data. This represents a clear case of over-fitting by ANN. During testing, the GMDH avoided over-fitting and outperformed ANN by obtaining the least MSE of 0.04 and 1.16 and MAE of 0.07 for S1 and S2, respectively, while BPNN achieved MSE and MAE of 0.08 and 0.17 for S1, 1.96, and 0.9 for S2, and RBFNN obtained MSE and MAE of 0.15 and 0.25 for S1 and 1.4 and 0.87 for S2. Hence, the improved generalization performance of the GMDH makes it an improved form of a neural network for TOC, S1, and S2 prediction. The proposed model was further adopted to predict the geochemical data and determine the source rock quality for the East Lika-1 well, which has no core data.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10

Similar content being viewed by others

References

  • Amiri Bakhtiar, H., Telmadarreie, A., Shayesteh, M., Heidari Fard, M., Talebi, H., & Shirband, Z. (2011). Estimating total organic carbon content and source rock evaluation, applying ΔlogR and neural network methods: Ahwaz and Marun oilfields, SW of Iran. Petroleum Science and Technology, 29(16), 1691–1704.

    Article  Google Scholar 

  • Anastasakis, L., & Mort, N. (2001). The development of self-organization techniques in modelling: a review of the group method of data handling (GMDH). Research Report-University of Sheffield Department of Automatic Control and Systems Engineering, 47, 191–204.

  • Armaghani, D. J., Momeni, E., & Asteris, P. (2020). Application of group method of data handling technique in assessing deformation of rock mass. Metaheuristic Computing Applied, 1, 1–18.

    Google Scholar 

  • Asante-Okyere, S., Shen, C., Ziggah, Y. Y., Rulegeya, M. M., & Zhu, X. (2020). A novel hybrid technique of integrating gradient-boosted machine and clustering algorithms for lithology classification. Natural Resources Research, 29(4), 2257–2273.

    Article  Google Scholar 

  • Bai, Y., & Tan, M. (2020). Dynamic committee machine with fuzzy-c-means clustering for total organic carbon content prediction from wireline logs. Computers & Geosciences, 146, 104626.

  • Cappuccio, F., Porreca, M., Omosanya, K. O., Minelli, G., & Harishidayat, D. (2020). Total organic carbon (TOC) enrichment and source rock evaluation of the Upper Jurassic-Lower Cretaceous rocks (Barents Sea) by means of geochemical and log data. International Journal of Earth Sciences, 7, 1–12.

    Google Scholar 

  • Caracciolo, L., Andò, S., Vermeesch, P., Garzanti, E., McCabe, R., Barbarano, M., et al. (2020). A multidisciplinary approach for the quantitative provenance analysis of siltstone: Mesozoic Mandawa Basin, southeastern Tanzania. Geological Society, London, Special Publications, 484(1), 275–293.

    Article  Google Scholar 

  • Carvajal-Ortiz, H., & Gentzis, T. (2015). Critical considerations when assessing hydrocarbon plays using Rock-Eval pyrolysis and organic petrology data: Data quality revisited. International Journal of Coal Geology, 152, 113–122.

    Article  Google Scholar 

  • Charsky, A., & Herron, S. (2013). Accurate, direct total organic carbon (TOC) log from a new advanced geochemical spectroscopy tool: Comparison with conventional approaches for TOC estimation. In AAPG annual convention and exhibition, Pittsburg, PA, May, 2013 (pp. 19–22).

  • Curiale, J. A., & Curtis, J. B. (2016). Organic geochemical applications to the exploration for source-rock reservoirs—A review. Journal of Unconventional Oil and Gas Resources, 13, 1–31.

    Article  Google Scholar 

  • Delvaux, D. (2001). Karoo rifting in western Tanzania: Precursor of Gondwana break-up. Contributions to geology and paleontology of Gondwana in honor of Helmut Wopfner: Cologne, Geological Institute, University of Cologne, 111–125.

  • Einvik-Heitmann, V., Dypvik, H., Hou, G., Fossum, K., Nerbraten, K., Karega, A., et al. (2015). The early cretaceous Kihuluhulu formation of the Mandawa Basin. In First EAGE Eastern Africa petroleum geoscience forum, 2015 (Vol. 2015, pp. 1–1, Vol. 1). European Association of Geoscientists & Engineers.

  • Emanuel, A., Kasanzu, C., & Kagya, M. (2020). Geochemical characterization of hydrocarbon source rocks of the Triassic-Jurassic time interval in the Mandawa basin, southern Tanzania: Implications for petroleum generation potential. South African Journal of Geology, 17, 121–134.

    Google Scholar 

  • Evenick, J. C. (2020). Late Cretaceous (Cenomanian and Turonian) organofacies and TOC maps: Example of leveraging the global rise in public-domain geochemical source rock data. Marine and Petroleum Geology, 111, 301–308.

    Article  Google Scholar 

  • Fossum, K., Dypvik, H., Haid, M. H., Hudson, W. E., & van den Brink, M. (2020). Late Jurassic and Early Cretaceous sedimentation in the Mandawa Basin, coastal Tanzania. Journal of African Earth Sciences, 104013.

  • Fossum, K., Morton, A. C., Dypvik, H., & Hudson, W. E. (2019). Integrated heavy mineral study of Jurassic to Paleogene sandstones in the Mandawa Basin, Tanzania: Sediment provenance and source-to-sink relations. Journal of African Earth Sciences, 150, 546–565.

    Article  Google Scholar 

  • Hakimi, M. H., Abdullah, W. H., Lashin, A. A., Ibrahim, E.-K.H., & Makeen, Y. M. (2020). Hydrocarbon generation potential of the organic-rich Naifa Formation, Say’un–Masila Rift Basin, Yemen: Insights from geochemical and palynofacies analyses. Natural Resources Research, 29(4), 2687–2715.

    Article  Google Scholar 

  • Hood, A., Gutjahr, C., & Heacock, R. (1975). Organic metamorphism and the generation of petroleum. AAPG Bulletin, 59(6), 986–996.

    Google Scholar 

  • Hou, G. (2015). Late Cretaceous Sedimentation (Mavuji Group) in Mandawa Basin, Tanzania.

  • Hudson, W. (2011). The geological evolution of the petroleum prospective Mandawa Basin southern coastal Tanzania. Trinity College Dublin

  • Hudson, W., & Nicholas, C. (2014). The Pindiro Group (Triassic to Early Jurassic Mandawa Basin, southern coastal Tanzania): Definition, palaeoenvironment, and stratigraphy. Journal of African Earth Sciences, 92, 55–67.

    Article  Google Scholar 

  • Kagya, M. L. (1996). Geochemical characterization of Triassic petroleum source rock in the Mandawa basin, Tanzania. Journal of African Earth Sciences, 23(1), 73–88.

    Article  Google Scholar 

  • Liu, C., Zhao, W., Sun, L., Wang, X., Sun, Y., Zhang, Y., et al. (2020). Geochemical assessment of the newly discovered oil-type Shale in the Shuangcheng area of the northern Songliao Basin, China. Journal of Petroleum Science and Engineering, 196, 107755.

  • Mahmoud, A. A., Elkatatny, S., Ali, A., Abdulraheem, A., & Abouelresh, M. (2020). Estimation of the total organic carbon using functional neural networks and support vector machine. In International petroleum technology conference, 2020. International Petroleum Technology Conference.

  • Mahmoud, A. A., Elkatatny, S., Ali, A. Z., Abouelresh, M., & Abdulraheem, A. (2019). Evaluation of the total organic carbon (TOC) using different artificial intelligence techniques. Sustainability, 11(20), 5643.

    Article  Google Scholar 

  • Mahmoud, A. A. A., Elkatatny, S., Mahmoud, M., Abouelresh, M., Abdulraheem, A., & Ali, A. (2017). Determination of the total organic carbon (TOC) based on conventional well logs using artificial neural network. International Journal of Coal Geology, 179, 72–80.

    Article  Google Scholar 

  • Mani, D., Kalpana, M., Patil, D., & Dayal, A. (2017). Organic matter in gas shales: Origin, evolution, and characterization. In Shale Gas (pp. 25–54). Elsevier.

  • Najafzadeh, M., & Azamathulla, H. M. (2013). Group method of data handling to predict scour depth around bridge piers. Neural Computing and Applications, 23(7–8), 2107–2112.

    Article  Google Scholar 

  • Nerbråten, K. B. (2014). Petrology and sedimentary provenance of Mesozoic and Cenozoic sequences in the Mandawa Basin.

  • Passey, Q., Creaney, S., Kulla, J., Moretti, F., & Stroud, J. (1990). A practical model for organic richness from porosity and resistivity logs. AAPG Bulletin, 74(12), 1777–1794.

    Google Scholar 

  • Passey, Q. R., Bohacs, K., Esch, W. L., Klimentidis, R., & Sinha, S. (2010) From oil-prone source rock to gas-producing shale reservoir-geologic and petrophysical characterization of unconventional shale gas reservoirs. In International oil and gas conference and exhibition in China, 2010. Society of Petroleum Engineers

  • Qian, N. (1999). On the momentum term in gradient descent learning algorithms. Neural Networks, 12(1), 145–151.

    Article  Google Scholar 

  • Rui, J., Zhang, H., Ren, Q., Yan, L., Guo, Q., & Zhang, D. (2020). TOC content prediction based on a combined Gaussian process regression model. Marine and Petroleum Geology, 104429.

  • Schmoker, J. W. (1979). Determination of organic content of Appalachian Devonian shales from formation-density logs: Geologic notes. AAPG Bulletin, 63(9), 1504–1509.

    Google Scholar 

  • Shalaby, M. R., Jumat, N., Lai, D., & Malik, O. (2019). Integrated TOC prediction and source rock characterization using machine learning, well logs and geochemical analysis: Case study from the Jurassic source rocks in Shams Field, NW Desert, Egypt. Journal of Petroleum Science and Engineering, 176, 369–380.

    Article  Google Scholar 

  • Shalaby, M. R., Malik, O. A., Lai, D., Jumat, N., & Islam, M. A. (2020). Thermal maturity and TOC prediction using machine learning techniques: Case study from the Cretaceous-Paleocene source rock, Taranaki Basin, New Zealand. Journal of Petroleum Exploration and Production Technology, 10, 2175–2193.

    Article  Google Scholar 

  • Shanmuganathan, S. (2016). Artificial neural network modelling: An introduction. In Artificial neural network modelling (pp. 1–14). Springer.

  • Shen, C., Asante-Okyere, S., Yevenyo Ziggah, Y., Wang, L., & Zhu, X. (2019). Group method of data handling (GMDH) lithology identification based on wavelet analysis and dimensionality reduction as well log data pre-processing techniques. Energies, 12(8), 1509.

    Article  Google Scholar 

  • Shi, X., Wang, J., Liu, G., Yang, L., Ge, X., & Jiang, S. (2016). Application of extreme learning machine and neural networks in total organic carbon content prediction in organic shale with wireline logs. Journal of Natural Gas Science and Engineering, 33, 687–702.

    Article  Google Scholar 

  • Smelror, M., Fossum, K., Dypvik, H., Hudson, W., & Mweneinda, A. (2018). Late Jurassic-Early Cretaceous palynostratigraphy of the onshore Mandawa Basin, southeastern Tanzania. Review of Palaeobotany and Palynology, 258, 248–255.

    Article  Google Scholar 

  • Tan, M., Song, X., Yang, X., & Wu, Q. Z. (2015). Support-vector-regression machine technology for total organic carbon content prediction from wireline logs in organic shale: A comparative study. Journal of Natural Gas Science and Engineering., 26, 792–802.

    Article  Google Scholar 

  • Tariq, Z., Mahmoud, M., Abouelresh, M., & Abdulraheem, A. (2020). Data-driven approaches to predict thermal maturity indices of organic matter using artificial neural networks. ACS Omega.

  • Tenaglia, M., Eberli, G. P., Weger, R. J., Blanco, L. R., Sánchez, L. E. R., & Swart, P. K. (2020). Total organic carbon quantification from wireline logging techniques: A case study in the Vaca Muerta Formation, Argentina. Journal of Petroleum Science and Engineering, 107489.

  • Wang, P., Peng, S., & He, T. H. (2018). A novel approach to total organic carbon content prediction in shale gas reservoirs with well logs data, Tonghua Basin, China. Journal Natural Gas Science and Engineering, 55, 1–15.

    Article  Google Scholar 

  • Wang, H., Wu, W., Chen, T., Dong, X., & Wang, G. (2019a). An improved neural network for TOC, S1 and S2 estimation based on conventional well logs. Journal of Petroleum Science and Engineering, 176, 664–678.

    Article  Google Scholar 

  • Wang, J., Gu, D., Guo, W., Zhang, H., & Yang, D. (2019b). Determination of total organic carbon content in shale formations with regression analysis. Journal of Energy Resources Technology, 141(1).

  • Xiong, H., Wu, X., & Fu, J. (2019). Determination of total organic carbon for organic rich shale reservoirs by means of cores and logs. In SPE annual technical conference and exhibition, 2019. Society of Petroleum Engineers.

  • Yu, H., Rezaee, R., Wang, Z., Han, T., Zhang, Y., Arif, M., & Johnson, L. (2017). A new method for TOC estimation in tight shale gas reservoirs. International Journal of Coal Geology, 179, 269–277.

    Article  Google Scholar 

  • Zhang, B., Chen, C., Ye, Q., Liu, J., & Doermann, D. (2019). Calibrated stochastic gradient descent for convolutional neural networks. In Proceedings of the AAAI conference on artificial intelligence, 2019 (Vol. 33, pp. 9348–9355).

  • Zhu, L., Zhang, C., Zhang, C., Wei, Y., Zhou, X., Cheng, Y., et al. (2018). Prediction of total organic carbon content in shale reservoir based on a new integrated hybrid neural network and conventional well logging curves. Journal of Geophysics and Engineering, 15(3), 1050–1061.

    Article  Google Scholar 

  • Zhu, L., Zhang, C., Zhang, C., Zhang, Z., Nie, X., Zhou, X., et al. (2019a). Forming a new small sample deep learning model to predict total organic carbon content by combining unsupervised learning with semisupervised learning. Applied Soft Computing, 83, 105596.

  • Zhu, L., Zhang, C., Zhang, Z., Zhou, X., & Liu, W. (2019b). An improved method for evaluating the TOC content of a shale formation using the dual-difference ΔlogR method. Marine and Petroleum Geology, 102, 800–816.

    Article  Google Scholar 

  • Zhu, L., Zhang, C., Zhang, C., Zhang, Z., Zhou, X., Liu, W., et al. (2020). A new and reliable dual model-and data-driven TOC prediction concept: A TOC logging evaluation method using multiple overlapping methods integrated with semi-supervised deep learning. Journal of Petroleum Science and Engineering, 188, 10694.

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the Major National Science and Technology Programs in the “Thirteenth Five-Year” Plan period (No. 2017ZX05032-002-004), the Outstanding Youth Funding of Natural Science Foundation of Hubei Province (No. 2016CFA055).

Author information

Authors and Affiliations

  1. Key Laboratory of Tectonics and Petroleum Resources, Ministry of Education, China University of Geosciences, Wuhan, 430074, China

    Alvin K. Mulashani, Chuanbo Shen, Solomon Asante-Okyere & Elieneza N. Abelly

  2. Department of Geoscience and Mining Technology, College of Engineering and Technology, Mbeya University of Science and Technology, Box 131, Mbeya, Tanzania

    Alvin K. Mulashani

  3. Department of Petroleum and Natural Gas Engineering, School of Petroleum Studies, University of Mines and Technology, Tarkwa, Ghana

    Solomon Asante-Okyere

  4. Faculty of Geophysics and Geomatics, China University of Geosciences, Wuhan, 430074, China

    Paulus N. Kerttu

Authors
  1. Alvin K. Mulashani
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Chuanbo Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Solomon Asante-Okyere
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Paulus N. Kerttu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Elieneza N. Abelly
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Chuanbo Shen.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mulashani, A.K., Shen, C., Asante-Okyere, S. et al. Group Method of Data Handling (GMDH) Neural Network for Estimating Total Organic Carbon (TOC) and Hydrocarbon Potential Distribution (S1, S2) Using Well Logs. Nat Resour Res 30, 3605–3622 (2021). https://doi.org/10.1007/s11053-021-09908-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11053-021-09908-3

Keywords

Search

Navigation