Hybrid machine learning algorithms to predict condensate viscosity in the near wellbore regions of gas condensate reservoirs

Abstract Gas condensate reservoirs display unique phase behavior and are highly sensitive to reservoir pressure changes. This makes it difficult to determine their PVT characteristics, including their condensate viscosity, which is a key variable in determining their flow behavior. In this study, a novel machine learning approach is developed to predict condensate viscosity in the near wellbore regions (μc) from five input variables: pressure (P), temperature (T), initial gas to condensate ratio (RS), gas specific gravity (γg), and condensate gravity (API). Due to the absence of accurate recombination methods for determining μc machine learning methods offer a useful alternative approach. Nine machine learning and hybrid machine learning algorithms are evaluated including novel multiple extreme learning machine (MELM), least squares support vector machine (LSSVM) and multi-layer perceptron (MLP) and each hybridized with a particle swarm optimizer (PSO) and genetic algorithm (GA). The new MELM algorithm has some advantages including 1) rapid execution, 2) high accuracy, 3) simple training, 4) avoidance of overfitting, 5) non-linear conversion during training, 6) no trapping at local optima, 6) fewer optimization steps than SVM and LSSVM. Combining MELM with PSO, to find the best control parameters, further improves its performance. Analysis indicates that the MELM-PSO model provides the highest μc prediction accuracy achieving a root mean squared error (RMSE) of 0.0035 cP and a coefficient of determination (R2) of 0.9931 for a dataset of 2269 data records compiled from gas-condensate fields around the world. The MELM-PSO algorithm generates no outlying data predictions. Spearman correlation coefficient analysis identifies that P, γg and Rs are the most influential variables in terms of condensate viscosity based on the large dataset studied.

[1]  S. Heddam,et al.  Lithofacies prediction in non-cored wells from the Sif Fatima oil field (Berkine basin, southern Algeria): A comparative study of multilayer perceptron neural network and cluster analysis-based approaches , 2020, Journal of African Earth Sciences.

[2]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[3]  Ehsan Khamehchi,et al.  Modeling viscosity of light and intermediate dead oil systems using advanced computational frameworks and artificial neural networks , 2020 .

[4]  David Bienvenido-Huertas,et al.  Estimating Adaptive Setpoint Temperatures Using Weather Stations , 2019, Energies.

[5]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[6]  Shifei Ding,et al.  Multi layer ELM-RBF for multi-label learning , 2016, Appl. Soft Comput..

[7]  Zelimir Schmidt,et al.  New Correlations For Crude Oil Physical Properties , 1991 .

[8]  Yoshua Bengio,et al.  No Unbiased Estimator of the Variance of K-Fold Cross-Validation , 2003, J. Mach. Learn. Res..

[9]  Abouzar Choubineh,et al.  Improved predictions of wellhead choke liquid critical-flow rates: Modelling based on hybrid neural network training learning based optimization , 2017 .

[10]  A. Hassanpouryouzband,et al.  CO2 Capture by Injection of Flue Gas or CO2-N2 Mixtures into Hydrate Reservoirs: Dependence of CO2 Capture Efficiency on Gas Hydrate Reservoir Conditions. , 2018, Environmental science & technology.

[11]  H. D. Beggs,et al.  Estimating the Viscosity of Crude Oil Systems , 1975 .

[12]  G. Falcone,et al.  Measurement of Gas Viscosity at High Pressures and High Temperatures , 2009 .

[13]  Xiao-Liang Shen,et al.  A hybrid particle swarm optimization algorithm using adaptive learning strategy , 2018, Inf. Sci..

[14]  Johan A. K. Suykens,et al.  Weighted least squares support vector machines: robustness and sparse approximation , 2002, Neurocomputing.

[15]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[16]  Guang-Bin Huang,et al.  What are Extreme Learning Machines? Filling the Gap Between Frank Rosenblatt’s Dream and John von Neumann’s Puzzle , 2015, Cognitive Computation.

[17]  Abdolhossein Hemmati-Sarapardeh,et al.  Modeling relative permeability of gas condensate reservoirs: Advanced computational frameworks , 2020, Journal of Petroleum Science and Engineering.

[18]  Mehdi Mehrpooya,et al.  A novel multi-hybrid model for estimating optimal viscosity correlations of Iranian crude oil , 2016 .

[19]  B. Dindoruk,et al.  Machine learning augmented dead oil viscosity model for all oil types , 2020 .

[20]  E Marubini,et al.  Bravais-Pearson and Spearman correlation coefficients: meaning, test of hypothesis and confidence interval. , 2002, The International journal of biological markers.

[21]  Ali Naseri,et al.  Toward reservoir oil viscosity correlation , 2013 .

[22]  J. Sheng,et al.  Optimization of huff-n-puff gas injection in a shale gas condensate reservoir , 2016 .

[23]  David A. Wood,et al.  Adaptive neuro-fuzzy algorithm applied to predict and control multi-phase flow rates through wellhead chokes , 2020 .

[24]  Ali Naseri,et al.  Reservoir oil viscosity determination using a rigorous approach , 2014 .

[25]  Pejman Tahmasebi,et al.  Hybrid geological modeling: Combining machine learning and multiple-point statistics , 2020, Comput. Geosci..

[26]  B. Tohidi,et al.  Viscosity, density, interfacial tension and compositional data for near critical mixtures of methane + butane and methane + decane systems at 310.95K , 2005 .

[27]  Jacob Benesty,et al.  Noise Reduction in Speech Processing , 2009 .

[28]  David A. Wood,et al.  Performance comparison of bubble point pressure from oil PVT data: Several neurocomputing techniques compared , 2019, Experimental and Computational Multiphase Flow.

[29]  Jamshid Moghadasi,et al.  Development of a New Comprehensive Model for Choke Performance Correlation in Iranian Oil Wells , 2014 .

[30]  Keun-Chang Kwak,et al.  Short-Term Electricity-Load Forecasting Using a TSK-Based Extreme Learning Machine with Knowledge Representation , 2017 .

[31]  Tao Yang,et al.  LBC Viscosity Modeling of Gas Condensate to Heavy Oil , 2007 .

[32]  Ibrahim Berkan Aydilek.  A hybrid firefly and particle swarm optimization algorithm for computationally expensive numerical problems , 2018, Appl. Soft Comput..

[33]  Adel M. Elsharkawy,et al.  Models for predicting the viscosity of Middle East crude oils , 1999 .

[34]  B. Tohidi,et al.  Viscosity of binary and multicomponent hydrocarbon fluids at high pressure and high temperature conditions: Measurements and predictions , 2013 .

[35]  Chiranth Hegde,et al.  Rate of Penetration (ROP) Modeling Using Hybrid Models: Deterministic and Machine Learning , 2018 .

[36]  L. Myers,et al.  Spearman Correlation Coefficients, Differences between , 2004 .

[37]  Hamzeh Ghorbani,et al.  Rheological and filtration characteristics of drilling fluids enhanced by nanoparticles with selected additives: an experimental study , 2018 .

[38]  Amir H. Mohammadi,et al.  Efficient estimation of natural gas compressibility factor using a rigorous method , 2014 .

[39]  Jamshid Moghadasi,et al.  Prediction of Bubble Point Pressure Using New Hybrid Computationail Intelligence Models , 2021 .

[40]  Jamshid Moghadasi,et al.  Comparison of accuracy and computational performance between the machine learning algorithms for rate of penetration in directional drilling well , 2021 .

[41]  Alireza Bahadori,et al.  An accurate model to predict drilling fluid density at wellbore conditions , 2016 .

[42]  David A. Wood,et al.  Prediction performance advantages of deep machine learning algorithms for two-phase flow rates through wellhead chokes , 2021, Journal of Petroleum Exploration and Production.

[43]  D. B. Bennion,et al.  Gas Condensate Reservoir Performance , 2009 .

[44]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[45]  David A. Wood,et al.  Prediction of oil flow rate through an orifice flow meter: Artificial intelligence alternatives compared , 2020 .

[46]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[47]  Tarek Helmy,et al.  Hybrid computational models for the characterization of oil and gas reservoirs , 2010, Expert Syst. Appl..

[48]  B. Dabir,et al.  Modeling gas/vapor viscosity of hydrocarbon fluids using a hybrid GMDH-type neural network system , 2017 .

[49]  David A. Wood,et al.  Shear modulus prediction of embedded pressurized salt layers and pinpointing zones at risk of casing collapse in oil and gas wells , 2020 .

[50]  Sohrab Zendehboudi,et al.  Prediction of Condensate-to-Gas Ratio for Retrograde Gas Condensate Reservoirs Using Artificial Neural Network with Particle Swarm Optimization , 2012 .

[51]  L. K. Thomas,et al.  Oil and Gas Relative Permeabilities Determined From Rate-Time Performance Data , 1986 .

[52]  Zhangxin Chen,et al.  AI based mechanistic modeling and probabilistic forecasting of hybrid low salinity chemical flooding , 2020 .

[53]  Jayantha Kodikara,et al.  Evaluation of soil-concrete interface shear strength based on LS-SVM , 2016 .

[54]  A. Ramezanzadeh,et al.  Machine learning technique for the prediction of shear wave velocity using petrophysical logs , 2019, Journal of Petroleum Science and Engineering.

[55]  Mohammad Ali Ahmadi,et al.  Evolving smart approach for determination dew point pressure through condensate gas reservoirs , 2014 .

[56]  Alberto Guadagnini,et al.  Analysis of the performance of a crude-oil desalting system based on historical data , 2021, Fuel.

[57]  David A. Wood,et al.  Wellbore stability analysis to determine the safe mud weight window for sandstone layers , 2019, Petroleum Exploration and Development.

[58]  S. Zendehboudi,et al.  Determination of bubble point pressure and oil formation volume factor: Extra trees compared with LSSVM-CSA hybrid and ANFIS models , 2020 .

[59]  John Lohrenz,et al.  Calculating Viscosities of Reservoir Fluids From Their Compositions , 1964 .

[60]  David A. Wood,et al.  Petroleum Well Blowouts as a Threat to Drilling Operation and Wellbore Sustainability: Causes, Prevention, Safety and Emergency Response , 2021, Journal of Construction Materials.

[61]  Viscosity model based on equations of state for hydrocarbon liquids and gases , 1997 .

[62]  L. Lin,et al.  A concordance correlation coefficient to evaluate reproducibility. , 1989, Biometrics.

[63]  David A. Wood,et al.  Predicting Formation Pore-Pressure from Well-Log Data with Hybrid Machine-Learning Optimization Algorithms , 2021, Natural Resources Research.

[64]  Jorge L. D. Pinaya,et al.  Brazilian montane rainforest expansion induced by Heinrich Stadial 1 event , 2019, Scientific Reports.

[65]  Iftikhar Ahmad,et al.  Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000 , 2018 .

[66]  Natarajan Sriraam,et al.  Optimal configuration of multilayer perceptron neural network classifier for recognition of intracranial epileptic seizures , 2017, Expert Syst. Appl..

[67]  Jamshid Moghadasi,et al.  Determination of bubble point pressure & oil formation volume factor of crude oils applying multiple hidden layers extreme learning machine algorithms , 2021, Journal of Petroleum Science and Engineering.

[68]  Akhil Datta-Gupta,et al.  Electrofacies Characterization and Permeability Predictions in Carbonate Reservoirs: Role of Multivariate Analysis and Nonparametric Regression , 1999 .

[69]  H. G. O'Dell,et al.  Successfully Cycling a Low-Permeability, High-Yield Gas Condensate Reservoir , 1967 .

[70]  Salaheldin Elkatatny,et al.  Real-Time Determination of Rheological Properties of Spud Drilling Fluids Using a Hybrid Artificial Intelligence Technique , 2019, Journal of Energy Resources Technology.

[71]  Egberto Pereira,et al.  Hybrid Unsupervised Extreme Learning Machine Applied to Facies Identification , 2021 .

[72]  Guang-Bin Huang,et al.  An Insight into Extreme Learning Machines: Random Neurons, Random Features and Kernels , 2014, Cognitive Computation.

[73]  Jamshid Moghadasi,et al.  Hybrid computing models to predict oil formation volume factor using multilayer perceptron algorithm , 2021 .

[74]  Farhad Gharagheizi,et al.  Toward a predictive model for estimating dew point pressure in gas condensate systems , 2013 .

[75]  T. Monger-Mcclure,et al.  Determination of Relative Permeability and Recovery for North Sea Gas-Condensate Reservoirs , 1999 .

[76]  H. R. Zhang,et al.  Condensate Banking Dynamics in Gas Condensate Fields: Compositional Changes and Condensate Accumulation Around Production Wells , 2000 .

[77]  Mohammad Farsi,et al.  Prediction of oil flow rate through orifice flow meters: Optimized machine-learning techniques , 2021 .

[78]  Ke Chen,et al.  Chaotic dynamic weight particle swarm optimization for numerical function optimization , 2018, Knowl. Based Syst..

[79]  Jamshid Moghadasi,et al.  Prediction of gas flow rates from gas condensate reservoirs through wellhead chokes using a firefly optimization algorithm , 2017 .

[80]  Shifei Ding,et al.  Extreme learning machine and its applications , 2013, Neural Computing and Applications.

[81]  Shahaboddin Shamshirband,et al.  Extreme learning machine assessment for estimating sediment transport in open channels , 2016, Engineering with Computers.

[82]  David A. Wood,et al.  A hybrid nanocomposite of poly(styrene-methyl methacrylate- acrylic acid) /clay as a novel rheology-improvement additive for drilling fluids , 2019, Journal of Polymer Research.

[83]  Durga Toshniwal,et al.  PP-NFR: an improved hybrid learning approach for porosity prediction from seismic attributes using non-linear feature reduction , 2019, Journal of Applied Geophysics.

[84]  Majid Amidpour,et al.  A hybrid artificial neural network and genetic algorithm for predicting viscosity of Iranian crude oils , 2014 .

[85]  A. Hassanpouryouzband,et al.  Geological CO2 Capture and Storage with Flue Gas Hydrate Formation in Frozen and Unfrozen Sediments: Method Development, Real Time-Scale Kinetic Characteristics, Efficiency, and Clathrate Structural Transition , 2019, ACS Sustainable Chemistry & Engineering.

[86]  A. Naseri,et al.  A NEURAL NETWORK MODEL AND AN UPDATED CORRELATION FOR ESTIMATION OF DEAD CRUDE OIL VISCOSITY , 2012 .

[87]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[88]  W. Al-Mudhafar.  Incorporation of Bootstrapping and Cross-Validation for Efficient Multivariate Facies and Petrophysical Modeling , 2016 .

[89]  Ian D. Gates,et al.  The application of supervised machine learning techniques for multivariate modelling of gas component viscosity: A comparative study , 2021 .

[90]  A. Hassanpouryouzband,et al.  Insights into CO2 Capture by Flue Gas Hydrate Formation: Gas Composition Evolution in Systems Containing Gas Hydrates and Gas Mixtures at Stable Pressures , 2018 .

[91]  Pijush Samui,et al.  Predicting permeability of tight carbonates using a hybrid machine learning approach of modified equilibrium optimizer and extreme learning machine , 2021, Acta Geotechnica.

[92]  Milad Arabloo,et al.  Robust Modeling Approach for Estimation of Compressibility Factor in Retrograde Gas Condensate Systems , 2014 .

[93]  Guang-Bin Huang,et al.  Extreme learning machine: a new learning scheme of feedforward neural networks , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[94]  Abdulazeez Abdulraheem,et al.  A hybrid particle swarm optimization and support vector regression model for modelling permeability prediction of hydrocarbon reservoir , 2017 .

[95]  David A. Wood,et al.  A geomechanical approach to casing collapse prediction in oil and gas wells aided by machine learning , 2021 .

[96]  David Bienvenido-Huertas,et al.  Applying an artificial neural network to assess thermal transmittance in walls by means of the thermometric method , 2019, Applied Energy.

[97]  Yuehua Huang,et al.  Short-term wind power prediction based on LSSVM–GSA model , 2015 .

[98]  Dan Simon,et al.  Evolutionary Optimization Algorithms , 2013 .

[99]  T. Yagi,et al.  High-pressure responses of alkali metal hydrogen carbonates, RbHCO3 and CsHCO3: Findings of new phases and unique compressional behavior , 2020 .

[100]  Watheq J. Al-Mudhafar,et al.  Integrating well log interpretations for lithofacies classification and permeability modeling through advanced machine learning algorithms , 2017, Journal of Petroleum Exploration and Production Technology.

[101]  P. L. Chong,et al.  Modelling viscosity of liquid dropout near wellbore region in gas condensate reservoirs using modern numerical approaches , 2020 .

[102]  A. Hassanpouryouzband,et al.  An Experimental Investigation on the Kinetics of Integrated Methane Recovery and CO2 Sequestration by Injection of Flue Gas into Permafrost Methane Hydrate Reservoirs , 2019, Scientific Reports.

[103]  Li Chen,et al.  The lattice Boltzmann method for isothermal micro-gaseous flow and its application in shale gas flow: a review , 2016 .

[104]  Carlos Rubio-Bellido,et al.  Comparison of linear regression and artificial neural networks models to predict heating and cooling energy demand, energy consumption and CO2 emissions , 2017 .

[105]  A. Mohammadi,et al.  Experimental measurement and modeling of saturated reservoir oil viscosity , 2014, Korean Journal of Chemical Engineering.

[106]  Shahaboddin Shamshirband,et al.  Rigorous prognostication of natural gas viscosity: Smart modeling and comparative study , 2018, Fuel.

[107]  A. Pádua,et al.  Viscosity and density of mixtures of methane and n-decane from 298 to 393 K and up to 75 MPa , 2004 .