Estimation of SARA composition of crudes purely from density and viscosity using machine learning based models

Abstract

Accurate characterization of crude oils by determining the composition of saturates, aromatics, resins and asphaltenes (SARA) has always been a challenging task in the petroleum industry. However, conventional experimental methods for determination of SARA composition are labour intensive, time-consuming and expensive. In the present study, artificial neural network (ANN) models were developed to predict the SARA composition from easily measurable parameters like density and viscosity. A dataset of 216 crude oil samples covering wide range of geographical locations was compiled from various literature sources. The ANN models with one hidden layer and six neurons are trained, tested and validated using MATLAB neural network toolbox. Results obtained on analysis revealed reasonably good accuracy of prediction of SARA components except for aromatics. The performance of developed ANN models was compared with various correlations reported in literature and found to be better in terms of mean squared error and coefficient of determination. The developed models hence provide a cost-effective and time-efficient alternative to the conventional SARA characterization techniques.

Keywords

SARA analysis / Crude oil / Artificial neural network / Predictive models / Density / Viscosity

CRediT authorship contribution statement

Anand D. Kulkarni: Writing -original draft, Software, Project administration, Methodology, Investigation, Formal analysis. Pratiksha D. Khurpade: Writing -review & editing, Validation, Methodology, Investigation, Formal analysis, Data curation. Somnath Nandi: Writing -review & editing, Visualization, Supervision, Resources, Formal analysis, Conceptualization.

Declaration of competing interest

This is to confirm that there are no relevant financial or non-financial competing interests to the research work reported.

References

[1] M.Z. Hasanvand, M.A. Ahmadi, R.M. Behbahani, F. Feyzi, Developing grey-box model model to diagnose asphaltene stability in crude oils: application of refractive index, Petroleum 2 (2016) 369-380, https://doi.org/10.1007/s11696-023-03242-z.

[2] A.A. Sulaimon, H. A/L Rajan, A. Qasim, N.P. Christiana, P.I. Murungi, Developing new correlations for asphaltene deposition involving SARA fractions and colloidal instability index, J. Petrol. Sci. Eng. 220 (2023) 111143, https://doi.org/10.1016/j.petrol.2022.111143.

[3] A. Karevan, M. Zirrahi, H. Hassanzadeh, Standardized high-performance liquid chromatography to replace conventional methods for determination of saturate, aromatic, resin, and asphaltene (SARA) fractions, ACS Omega 7 (22) (2022) 18897-18903, https://doi.org/10.1021/acsomega.2c01880.

[4] S. Rudyk, Relationships between SARA fractions of conventional oil, heavy oil, natural bitumen and residues, Fuel 216 (2018) 330-340, https://doi.org/10.1016/j.fuel.2017.12.001.

[5] J.M. Hidalgo-Herrador, A. Vráblík, P. Jísa, et al., Hydrovisbreaking of vacuum residue from Russian export blend: influence of brown coal, light cycle oil, or naphtha addition, Chem. Pap. 69 (2015) 1075-1083, https://doi.org/10.1515/chempap-2015-0119.

[6] Y. Hongfu, C. Xiaoli, L. Haoran, X. Yupeng, Determination of multi-properties of residual oils using mid-infrared attenuated total reflection spectroscopy, Fuel 85 (12-13) (2006) 1720-1728, https://doi.org/10.1016/j.fuel.2006.02.003.

[7] B.N. Barman, V.L. Cebolla, L. Membrado, Chromatographic techniques for petroleum and related products, Crit. Rev. Anal. Chem. 30 (2-3) (2000) 75-120, https://doi.org/10.1080/10408340091164199.

[8] P.L. Grizzle, D.M. Sablotny, Automated liquid chromatographic compound class group-type separation of crude oils and bitumens using chemically bonded aminosilane, Anal. Chem. 58 (12) (1986) 2389-2396, https://doi.org/10.1021/ac00125a009.

[9] C. Jiang, S.R. Larter, K.J. Noke, L.R. Snowdon, TLC-FID (Iatroscan) analysis of heavy oil and tar sand samples, Org. Geochem. 39 (8) (2008) 1210-1214, https://doi.org/10.1016/j.orggeochem.2008.01.013.

[10] J.M. Chaffin, M.S. Lin, M. Liu, R.R. Davison, C.J. Glover, J.A. Bullin, The use of HPLC to determine the saturate content of heavy petroleum products, J. Liq. Chromatogr. Relat. Technol. 19 (10) (1996) 1669-1682, https://doi.org/10.1080/10826079608005500.

[11] J.C. Suatoni, R.E. Swab, Rapid hydrocarbon group-type analysis by high performance liquid chromatography, J. Chromatogr. Sci. 13 (1975) 361-366, https://doi.org/10.1093/chromsci/13.8.361.

[12] C.A. Islas-Flores, E. Buenrostro-Gonzalez, C. Lira-Galeana, Comparisons between open column chromatography and HPLC SARA fractionations in petroleum, Energy Fuels 19 (5) (2005) 2080-2088, https://doi.org/10.1021/ef050148+.

[13] H. Bisht, M. Reddy, M. Malvanker, R.C. Patil, A. Gupta, B. Hazarika, A.K. Das, Efficient and quick method for saturates, aromatics, resins, and asphaltenes analysis of whole crude oil by thin-layer chromatography-flame ionization detector, Energy Fuels 27 (6) (2013) 3006-3013, https://doi.org/10.1021/ef4002204.

[14] T. Fan, J.S. Buckley, Rapid and accurate SARA analysis of medium gravity crude oils, Energy Fuels 16 (6) (2002) 1571-1575, https://doi.org/10.1021/ef0201228.

[15] N. Aske, H. Kallevik, J. Sjöblom, Determination of saturate, aromatic, resin, and asphaltenic (SARA) components in crude oils by means of infrared and nearinfrared spectroscopy, Energy Fuels 15 (5) (2001) 1304-1312, https://doi.org/10.1021/ef010088h.

[16] F.S. Falla, C. Larini, G.A.C. Le Roux, F.H. Quina, L.F.L. Moro, C.A.O. Nascimento, Characterization of crude petroleum by NIR, J. Petrol. Sci. Eng. 51 (1-2) (2006) 127-137, https://doi.org/10.1016/j.petrol.2005.11.014.

[17] S. Akmaz, O. Iscan, M.A. Gurkaynak, M. Yasar, The structural characterization of saturate, aromatic, resin, and asphaltene fractions of Batiraman crude oil, Petrol. Sci. Technol. 29 (2) (2011) 160-171, https://doi.org/10.1080/10916460903330361.

[18] A. Permanyer, D. Azevedo, C. Rebufa, J. Kister, F. Gonçalves, Characterization of Brazilian oils by FTIR and SUVF spectroscopy. A comparison with GC/MS results, Geogaceta 38 (2005) 139-141.

[19] V. Bansal, G.J. Krishna, A. Chopra, A.S. Sarpal, Detailed hydrocarbon characterization of RFCC feed stocks by NMR spectroscopic techniques, Energy Fuels 21 (2) (2007) 1024-1029, https://doi.org/10.1021/ef060268x.

[20] A.M. Kharrat, J. Zacharia, V.J. Cherian, A. Anyatonwu, Issues with comparing SARA methodologies, Energy Fuels 21 (6) (2007) 3618-3621, https://doi.org/10.1021/ef700393a.

[21] T. Fan, J. Wang, J.S. Buckley, Evaluating crude oils by SARA analysis, in: SPE/DOE Improved Oil Recovery Symposium (p. SPE-75228-MS), 2002, https://doi.org/10.2118/75228-MS.

[22] L.V. Meléndez, A. Lache, J.A. Orrego-Ruiz, Z. Pachón, E. Mejía-Ospino, Prediction of the SARA analysis of Colombian crude oils using ATR-FTIR spectroscopy and chemometric methods, J. Petrol. Sci. Eng. 90e91 (2012) 56-60, https://doi.org/10.1016/j.petrol.2012.04.016.

[23] M. Mohammadi, M.K. Khorrami, A. Vatani, H. Ghasemzadeh, H. Vatanparast, A. Bahramian, A. Fallah, Genetic algorithm based support vector machine regression for prediction of SARA analysis in crude oil samples using ATR-FTIR spectroscopy, Spectrochim. Acta A Mol. Biomol. Spectrosc. 245 (2021) 118945, https://doi.org/10.1016/j.saa.2020.118945.

[24] A. Chamkalani, Correlations between SARA fractions, density, and RI to investigate the stability of asphaltene, ISRN Anal. Chem. (I) (2012) 1-6, https://doi.org/10.5402/2012/219276.

[25] D. Stratiev, I. Shishkova, R. Dinkov, S. Nenov, S. Sotirov, E. Sotirova, I. Kolev, V. Ivanov, S. Ribagin, K. Atanassov, D. Stratiev, D. Yordanov, D. Nedanovski, Prediction of petroleum viscosity from molecular weight and density, Fuel 331 (2023), https://doi.org/10.1016/j.fuel.2022.125679.

[26] D. Stratiev, S. Sotirov, E. Sotirova, S. Nenov, R. Dinkov, I. Shishkova, I.V. Kolev, D. Yordanov, S. Vasilev, K. Atanassov, S. Simeonov, G.N. Palichev, Prediction of molecular weight of petroleum fluids by empirical correlations and artificial neuron networks, Processes 11 (2) (2023), https://doi.org/10.3390/pr11020426.

[27] D. Stratiev, S. Nenov, I. Shishkova, S. Sotirov, E. Sotirova, R. Dinkov, D. Yordanov, D. Pilev, K. Atanassov, S. Vasilev, D.D. Stratiev, Prediction of viscosity of blends of heavy oils with diluents by empirical correlations and artificial neural network, Ind. Eng. Chem. Res. 62 (49) (2023) 21449-21463, https://doi.org/10.1021/acs.iecr.3c02472.

[28] G.N. Palichev, D. Stratiev, S. Sotirov, E. Sotirova, S. Nenov, I. Shishkova, R. Dinkov, K. Atanassov, S. Ribagin, D.D. Stratiev, D. Pilev, D. Yordanov, Prediction of refractive index of petroleum fluids by empirical correlations and ANN, Processes 11 (2023) 2328, https://doi.org/10.3390/pr11082328.

[29] P. Goel, K. Saurabh, V. Patil-Shinde, S.S. Tambe, Prediction of API values of crude oils by use of saturates/aromatics/resins/asphaltenes analysis: computational-intelligence-based models, SPE J. 22 (3) (2017) 817-853, https://doi.org/10.2118/184391-pa. Society of Petroleum Engineers (SPE).

[30] A.A. Sulaimon, J.K.M. De Castro, S. Vatsa, S. New, Correlations and deposition envelopes for predicting asphaltene stability in crude oils, J. Petrol. Sci. Eng. 190 (2020) 106782, https://doi.org/10.1016/j.petrol.2019.106782.

[31] M. Vers¸ an Kök, M.A. Varfolomeev, D.K. Nurgaliev, Determination of SARA fractions of crude oils by NMR technique, J. Petrol. Sci. Eng. 179 (2019) 1-6, https://doi.org/10.1016/j.petrol.2019.04.026.

[32] O. Isaac Abiodun, A. Jantan, A. Esther Omolara, K. Victoria Dada, N. Abd Elatif Mohamed, H. Arshad, State-of-the-art in artificial neural network applications: a survey, Heliyon 4 (2018) 938, https://doi.org/10.1016/j.heliyon.2018.

[33] S. Alizadeh, S. Ta, L. Samavedham, A.K. Ray, Application of artificial neural network for prediction of 10 crude oil properties, Can. J. Chem. Eng. 101 (2023) 6203-6214, https://doi.org/10.1002/cjce.24938.

[34] Bruce P. Hollebone, Appendix A. The oil properties data appendix, in: M. Fingas (Ed.), Handbook of Oil Spill Science and Technology, 2014, pp. 575-681, https://doi.org/10.1002/9781118989982.app1.

[35] P. Jokuty, S. Whiticar, M. Fingas, E. Meyer, C. Knobel,Hydrocarbon groups and their relationships to oil properties and behaviour, in: 18th Arctic and Marine Oil Spill Program Technical Seminar, 1995. Edmonton; Canada, https://www. boem.gov/uploadedFiles/BOEM/Oil_and_Gas_Energy_Program/Plans/jokuty_ et_al_1995.pdf.

[36] Government of Canada, Environment Canada crude oil and petroleum product database e ECCC petroleum products database. https://open.canada.ca/data/en/dataset/53c38f91-35c8-49a6-a437-b311703db8c5/resource/518929dae63f-4d72-a47f-c98c8ed23020, 2021.

[37] A. Ali Anwar, M.S. Al-Jawad, A.A. Ali, Asphaltene stability of some Iraqi dead crude oils, J. Eng. 25 (3) (2019) 53-67. https://doi.org/10.31026/j.eng.2019.03.05.

[38] L. Nabzar, M.E. Aguilera, The colloidal approach. A promising route for asphaltene deposition modelling, Oil Gas Sci Technol Rev. IFP 63 (1) (2008) 21-35.

[39] S. Mohaghegh, Virtual-Intelligence applications in petroleum engineering: part 1 e artificial neural networks, J. Petrol. Technol. 52 (9) (2000) 64-73.

[40] J. Ghiasi-Freez, A. Kadkhodaie-Ilkhchi, M. Ziaii, Improving the accuracy of flow units prediction through two committee machine models: an example from the South Pars Gas Field, Persian Gulf Basin, Iran, Comput. Geosci. 46 (2012) 10-23, https://doi.org/10.1016/j.cageo.2012.04.006.

[41] A.H. Fath, A. Pouranfard, P. Foroughizadeh, Development of an artificial neural network model for prediction of bubble point pressure of crude oils, Petroleum 4 (3) (2018) 281-291, https://doi.org/10.1016/j.petlm.2018.03.009.

[42] V.D. da Silva Bispo, C.M. Scheid, L.A.D.C. Calçada, Luiz Augusto da Cruz Meleiro, Development of an ANN-based soft-sensor to estimate the apparent viscosity of water-based drilling fluids, J. Petrol. Sci. Eng. 150 (2017) 69-73, https://doi.org/10.1016/j.petrol.2016.11.030.