Abstract
In the well-log data processing, the principal advantage of the nuclear magnetic resonance (NMR) method is the measurement of fluid volume and pore size distribution without resorting to parameters such as rock resistivity. Preliminary processing of the well-log data allowed first to have the petrophysical parameters and then to evaluate the performances of the transverse relaxation time T 2 NMR. Petrophysical parameters such as the porosity of the formation as well as the effective permeability can be estimated without having recourse the fluid type. The well-log data of five wells were completed during the construction of intelligent models in the Saharan oil field Oued Mya Basin in order to assess the reliability of the developed models. Data processing of NMR combined with conventional well data was performed by artificial intelligence. First, the support vector regression method was applied to a sandy clay reservoir with a model based on the prediction of porosity and permeability. NMR parameters estimated using intelligent systems, i.e., fuzzy logic (FL) model, back propagation neural network (BP-NN), and support vector machine, with conventional well-log data are combined with those of NMR, resulting in a good estimation of porosity and permeability. The results obtained during the processing are then compared to the FL and NN regression models performed by the regression method during the validation stage. They show that the correlation coefficients R 2 estimated vary between 0.959 and 0.964, corresponding to the root mean square error values of 0.20 and 0.15.
Similar content being viewed by others
Abbreviations
- T 1 :
-
Longitudinal relaxation time, seconds
- T 2 :
-
Transverse relaxation time, seconds
- T 2LM :
-
Logarithmic mean of T 2 distribution spectrum
- T 2_DIST:
-
T 2 distribution of all fluids
- T 2gm :
-
Geometric mean relaxation time
- Bin1:
-
Clay-bound fluid volume
- Bin2:
-
Capillary-bound fluid volume
- Bin3:
-
Very small pore fluid volume
- Bin4:
-
Small pore fluid volume
- Bin5:
-
Medium pore fluid volume
- Bin6:
-
Medium-large pore fluid volume
- Bin7:
-
Large pore fluid volume
- Bin8:
-
Very large pore fluid volume
- NMR:
-
Nuclear magnetic resonance log
- FL:
-
Fuzzy logic
- NN:
-
Neural network
- SVR:
-
Support vector regression
- ANNs:
-
Artificial neural networks
- SCG:
-
Scaled conjugate gradient
- CMEF:
-
Committee machine with empirical formula
- GA:
-
Genetic algorithm
- FFP:
-
Free fluid porosity
- FFV:
-
Free fluid volume
- BFV:
-
Bulk fluid volume
- Swir :
-
Irreducible water saturation
- PE:
-
Absorption factor
- MAE:
-
Mean absolute error
- AAE:
-
Average absolute error
- LOO:
-
Leave-one-out cross-validation
- σ :
-
Constant parameter of the kernel function
- ε :
-
Insensitive loss function
References
Aïfa T, Baouche R, Baddari K (2014a) Neuro-fuzzy system to predict permeability and porosity from well log data: a case study of Hassi R’Mel gas field, Algeria. J Pet Sci Eng 123:217–229
Aïfa T, Ali Zerrouki A, Baddari K, Geraud Y (2014b) Magnetic susceptibility and its relation with fractures and petrophysical parameters in the tight sand oil reservoir of Hamra quartzites, southwest of the Hassi Messaoud oil field, Algeria. J Pet Sci Eng 123:120–137
Al-Anazi AF, Gates ID (2010) Support vector regression for porosity prediction in a heterogeneous reservoir: a comparative study. Comput Geosci 36(12):1494–1503. https://doi.org/10.1016/j.cageo.2010.03.022
Bakhorji A, Schmitt D (2008) Velocity of P- and S-waves in Arab-D and WCSB (Western Canada Sedimentary Basin, Canada) carbonates. In: Proceedings of the Canadian Society of Petroleum Geology (CSPG), Canadian Society of Exploration Geophysicists (CSEG) and Canadian Well-logging Society (CWLS) Convention, Calgary, pp 368–372
Bhatt A, Helle HB (2002) Committee neural networks for porosity and permeability prediction from well logs. Geophys Prospect 50:645–660
Boudjemaa A (1987) Evolution Structurale du Bassin Triasique dans le Nord Est du Sahara (Algérie). PhD Thesis, Université Paris-Sud, Orsay, 279p
Chiu SL (1994) Fuzzy model identification based on cluster estimation. J Intell Fuzzy Syst 2(3):267–278
Coates GR, Xiao L, Prammer MG (1999) NMR logging. Principles and applications. Gulf Publishing Company, Halliburton Energy Services, Houston 234p
Courel L, Aït Salem H, Ben Ismail H, El Mostaïne M, Fekirine B, Kamoun F, Mami L, Oujidi M, Soussi M (2000) An overview of the epicontinental, Triassic series of the Maghreb (NW Africa). In: Bachmann GH & Lerche I (eds): Epicontinental Triassic. Zbl Geol Paläont 2(9–10):1145–1166
Dominic S, Das R, Whitley D, Anderson C (1991) Genetic reinforcement learning for neural networks. IJCNN-91-Seattle International Joint Conference on Neural Networks, Washington, Seattle, USA. IEEE Neural Netw 2:71–76
Eastwood RL, Castagna JP (1987) Interpretation of Vp/Vs ratios from sonic logs. Shear-wave exploration, SHDanbom& SN Domenico(eds), SEG, Geophysical Development Series no. 1. Tulsa, pp 139–153
Fung C, Wong K, Eren H (1997) Modular artificial neural network for prediction of petrophysical properties from well log data. IEEE Trans Instrum Meas 46(6):1295–1299. https://doi.org/10.1109/19.668276
Galeazzi S, Point O, Haddadi N, Mather J, Druesne D (2012) The Illizi and BerkineBasins in Southern Algeria [abstract]. In: Roberts DG, Bally AW (eds) Regional geology and tectonics: phanerozoic passive pargins, cratonic basins and global tectonic maps. Elsevier, Amsterdam, pp 662–729
Hahn EL (1950) Spin Echoes. Phys Rev 80(4):580–594. https://doi.org/10.1103/PhysRev.80.580
Hastie T, Tibshirani R, Friedman J (2001) The elements of statistical learning: data mining, inference, and prediction. Springer, New York. https://doi.org/10.1007/978-0-387-21606-5
Hawkins DM (2004) The problem of overfitting. J Chem Inf Comput Sci 44(1):1–12. https://doi.org/10.1021/ci0342472
Helle H, Ursin B (2001) Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study. Geophys Prospect 49(4):431–444. https://doi.org/10.1046/j.1365-2478.2001.00271.x
Huang Z, Shimeld J, Williamson M, Katsube J (1996) Permeability prediction with artificial neural network modelling in the Venture gas field, offshore Eastern Canada. Geophysics 61(2):422–436. https://doi.org/10.1190/1.1443970
Huang Y, Gedeon TD, Wong PM (2001) An integrated neural-fuzzy-genetic- algorithm using hyper-surface membership functions to predict permeability in petroleum reservoirs. Eng Appl Artif Intell 14(1):15–21. https://doi.org/10.1016/S0952-1976(00)00048-8
Hurlimann MD (2001) Diffusion and relaxation effects in general stray field NMR experiments. J Magn Reson 148(2):367–378. https://doi.org/10.1006/jmre.2000.2263
Kadkhodaie-Ilkhchi A, Rezaee MR, Moallemi SA (2006) A fuzzy logic approach for the estimation of permeability and rock types from conventional well log data: an example from the Kangan reservoir in Iran offshore gas field, Iran. J Geophys Eng 3(4):356–369. https://doi.org/10.1088/1742-2132/3/4/007
Kadkhodaie-Ilkhchi A, Rahimpour-Bonab H, Rezaee MR (2008) A committee machine with intelligent systems for estimation of total organic carbon content from petrophysical data: an example from Kangan and Dalan reservoirs in South Pars gas field, Iran. Comput Geosci 35:459–474
Karimpouli S, Malehmir A (2015) Neuro-Bayesian facies inversion of prestack seismic data from a carbonate reservoir in Iran. J Pet Sci Eng 131:11–17. https://doi.org/10.1016/j.petrol.2015.04.024
Katahara KW (1995) Gamma ray log response in shaly sands. Log Anal 36(4):50–56
Kenyon WE (1997) Petrophysical principles of applications of NMR logging. Log Anal 38(2):21–43
Kenyon WE, Day PI, Straley C, Willemsen JF (1988) A three-part study of NMR longitudinal relaxation properties of water-saturated sandstones. SPE Form Eval 3(3):622–636 SPE-15643-PA
Malki HA, Baldwin J (2002) A neuro-fuzzy based oil/gas producibility estimation method. IEEE Int Joint Conf Neural Netw 1:896–901
Mamdani EH (1976) Advances in the linguistic synthesis of fuzzy controllers. Int J Man-Mach Stud 8(6):669–678. https://doi.org/10.1016/S0020-7373(76)80028-4
Masoudi P, NadjarAraabi B, Aïfa T, Memarian M (2016) Clustering as an efficient tool for assessing fluid content and movability by resistivity logs. In: Numerical Proceedings of the 4th International Mine and Mining Industries Congress and Exposition, Tehran, 9p
Masoudi P, Aïfa T, Memarian M, Tokhmechi B (2017a) Uncertainty assessment of volume of investigation to enhance the vertical resolution of well-logs. J Pet Sci Eng 154:252–276
Masoudi P, Aïfa T, Memarian M, Tokhmechi B (2017b) Uncertainty assessment of porosity and permeability by clustering algorithm and fuzzy arithmetic. J Pet Sci Eng. https://doi.org/10.1016/j.petrol.2017.11.018
Matlab user’s guide (2014) Version 2, Fuzzy Logic Toolbox. Math Works, USA, 235p
Mohaghegh S (2000) Virtual-intelligence applications in petroleum engineering: part I: artificial neural networks. J Pet Technol 52(09):64–73. https://doi.org/10.2118/58046-JPT
Narushima Y, Hiroshi Y (2012) Conjugate gradient methods based on secant conditions that generate descent search directions for unconstrained optimization. J Comput Appl Math 236(17):4303–4317
Nguyen D, Widrow B (1990) Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights. In: Proceedings of the International Joint Conference on Neural Networks 3:21–26
Nikravesh M, Aminzadeh F (2003) Soft computing for intelligent reservoir characterization and modeling. In: Nikravesh M, Aminzadeh F, Zadeh LA (eds) Soft computing and intelligent data analysis, developments in petroleum science series, vol 51. Elsevier, Amsterdam, pp 3–32
Ogilvie SR, Cuddy S, Lindsay C, Hurst A (2002) Novel methods of permeability prediction from NMR tool data. Dialog (magazine of the London Petrophysical Society), September
Petrolog (2010) Advanced log analysis software, V.10. Crocker data processing, Petroleum House, WA 6102, Australia
Rogers SJ, Fang JH, Karr CL, Stanley DA (1995) Determination of lithology from well logs using a neural network. AAPG Bull 76(5):731–739
Sabaou N, Aït Salem H, Zazoun RS (2009) Chemostratigraphy, tectonic setting and provenance of the Cambro-Ordovician clastic deposits of the subsurface Algerian Sahara. Sciences et Technologies des Hydrocarbures 1(1):27–33
Schlumberger (1988) Basic log interpretation, Schlumberger log interpretation charts. Schlumberger Educational Services, Texas
Sherrod PH (2009) DTREG Predictive Modeling Software User’s Manual, Version 9.1, http://www.dtreg.com
Sugeno M (1985) Industrial applications of fuzzy control. Elsevier Science Pub. Co, North Holland
Sugeno M, Kang GT (1988) Structure identification of fuzzy model. Fuzzy Sets Syst 28(1):15–33. https://doi.org/10.1016/0165-0114(88)90113-3
Tahmasebi P, Hezarkhani A (2011) Application of a modular feedforward neural network for grade estimation. Nat Resour Res 20(1):25–32. https://doi.org/10.1007/s11053-011-9135-3
Tahmasebi P, Javadpour F, Sahimi M (2017) Data mining and machine learning for identifying sweet spots in shale reservoirs. Expert Syst Appl 88:435–447. https://doi.org/10.1016/j.eswa.2017.07.015
Takagi T, Sugeno M (1985) Fuzzy identification of systems and its application to modeling and control. IEEE Trans Syst Man Cybern 15(1):116–132
Timur A (1968) An investigation of permeability, porosity, and residual water saturation relationships for sandstone reservoirs. In: SPWA 9th Annual Logging Symposium, New Orleans
Timur A (1969) Effective porosity and permeability of sandstones investigated through nuclear magnetic principles. Log Anal 10(1):3
Wong PM, Nikravesh M (2001) Field applications of intelligent computing techniques. J Pet Geol 24(4):381–387. https://doi.org/10.1111/j.1747-5457.2001.tb00681.x
Xiao L, Mao ZQ, Li GR, Jin Y (2013) Estimation of saturation exponent from nuclear magnetic resonance (NMR) logs in low permeability reservoirs. Appl Magn Reson 44(3):333–347. https://doi.org/10.1007/s00723-012-0366-1
Acknowledgements
We would like to thank Sonatrach Co. for providing the data and the Petrolog software program. A Matlab code was developed to better simulate and predict the porosity and the permeability. We are indebted to an anonymous reviewer for the fruitful comments which helped improve this manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Highlights
- New statistical values of NMR porosity and permeability from core data were applied to Saharan oil field, Oued Mya Basin.
- Conventional well logs were combined with NMR log parameters.
- Fuzzy logic clustering radii for permeability and porosity improved the NMR parameters for further performances.
- SVM shows best performance (R 2 = 0.96) for NMR compared to BP-NN (R 2 ~ 0.94) and FL (R 2 ~ 0.92) either for MPHIE or MPRN.
Rights and permissions
About this article
Cite this article
Baouche, R., Aïfa, T. & Baddari, K. Intelligent methods for predicting nuclear magnetic resonance of porosity and permeability by conventional well-logs: a case study of Saharan field. Arab J Geosci 10, 545 (2017). https://doi.org/10.1007/s12517-017-3344-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12517-017-3344-y