Abstract
Pressure transient response (PTR) of horizontal well in naturally fractured reservoirs (NFR) has a particular characteristic shape. This PTR is often used to estimate parameters of NFRs and detect their wellbore and boundary regimes. Interporosity flow coefficient (λ) and storativity ratio (ω) are two important parameters of the NFR that often estimated by matching process on the PTR. Since the matching techniques’ results are not often unique, in this study, the multi-output least squares support vector regression (MLS-SVR) is employed for simultaneous estimation of λ and ω. A databank of 500 PTRs for horizontal wells in naturally fractured reservoirs is generated by the finite element method, converted to the pressure derivative (PD) curves, and then used to develop and evaluate this auto-characterization paradigm. The predictive accuracy of the model is checked and validated by both smooth and noisy PTRs. The proposed model predicts ω and λ with overall absolute average relative deviations (AARD) of 0.186% and 3.754%, respectively. The correlation coefficients (R2) of 1 and 0.99992 are obtained for the prediction of ω and λ, respectively. The Leverage outlier detection technique justified that only less than 6% of the predictions are within the suspect region. This MLS-SVR model can be simply integrated with commercial pressure transient analysis (PTA) packages for accurate prediction of ω and λ even from the noisy PTRs.
Similar content being viewed by others
Abbreviations
- V :
-
bulk volume
- c :
-
compressibility
- k :
-
permeability
- L :
-
half-length of the horizontal well
- p :
-
pressure
- t :
-
time
- R :
-
risk function
- w :
-
weight vector
- b :
-
intercept of the SVR regression function
- k (x i, x):
-
kernel function
- N :
-
number of data records
- H :
-
Hat matrix
- X :
-
matrix of standardized residuals
- H ⁎ :
-
warning leverage value
- n :
-
number of model parameters
- m :
-
number of data records
- λ :
-
interporosity flow coefficient
- ω :
-
storativity ratio
- ϕ :
-
porosity
- α :
-
shape factor
- Δ :
-
difference
- ε :
-
error
- :
-
slack variable
- α i :
-
Lagrange multiplier
- γ :
-
width of the RBF
- NFR:
-
naturally fractured reservoirs
- PTR:
-
pressure transient response
- MLS-SVR:
-
multi-output least squares support vector regression
- PD:
-
pressure derivative
- AARD%:
-
absolute average relative deviations
- R 2 :
-
regression coefficients
- PTA:
-
pressure transient analyses
- ANN:
-
artificial neural networks
- RBF:
-
radial basis function
- MSE:
-
mean square error
- PV:
-
parameter value
- SR:
-
cutoff value for standardized residuals
- hr:
-
hour
- f:
-
fracture
- m:
-
matrix
- t:
-
total
- act.:
-
actual value
- cal.:
-
calculated value
- T:
-
transpose
- −1:
-
Inverse
- D:
-
dimensionless
References
Abdel-Fattah M, Gameel M, Awad S, Ismaila A (2015) Seismic interpretation of the aptian alamein dolomite in the razzak oil field, western desert, Egypt. Arab J Geosci 8:4669–4684
Alajmi MN, Ertekin T (2007) The development of an artificial neural network as a pressure transient analysis tool for applications in double-porosity reservoirs. In: Asia Pacific oil and gas conference and exhibition. Society of Petroleum Engineers, 30th October–1 November, Jakarta, Indonesia
Barenblatt GI, Zheltov YP (1960) Fundamental equations for the flow of homogeneous fluids through fissured rocks. In: Doklady Akademii Nauk. Russ Acad Sci 132:545–548
Bian XQ, Han B, Du ZM, Jaubert JN, Li MJ (2016) Integrating support vector regression with genetic algorithm for CO2-oil minimum miscibility pressure (MMP) in pure and impure CO2 streams. Fuel 182:550–557
Bourdet D (2002) Well test analysis: the use of advanced interpretation models. Elsevier, Amsterdam
Bourdet D, Ayoub JA, Pirard YM (1989) Use of pressure derivative in well test interpretation. SPE Form Eval 4:293–302
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Discovery 2:121–167
Cao B, Dong W, Lv Z, Gu Y, Singh S, Kumar P (2020a) Hybrid microgrid many-objective sizing optimization with fuzzy decision. IEEE Trans Fuzzy Syst 28:2702–2710
Cao B, Zhao J, Gu Y, Ling Y, Ma X (2020b) Applying graph-based differential grouping for multiobjective large-scale optimization. Swarm Evol Comput 53:100626
Cao Y, Li Y, Zhang G, Jermsittiparsert K, Nasseri M (2020c) An efficient terminal voltage control for PEMFC based on an improved version of whale optimization algorithm. Energy Rep 6:530–542
Chao L, Zhang K, Li Z, Zhu Y, Wang J, Yu Z (2018) Geographically weighted regression based methods for merging satellite and gauge precipitation. J Hydrol 558:275–289
Chen Y, He L, Guan Y, Lu H, Li J (2017) Life cycle assessment of greenhouse gas emissions and water-energy optimization for shale gas supply chain planning based on multi-level approach: case study in Barnett, Marcellus, Fayetteville, and Haynesville shales. Energy Convers Manag 134:382–398
Chen H, Qiao H, Xu L, Feng Q, Cai K (2019) A fuzzy optimization strategy for the implementation of RBF LSSVR model in vis–NIR analysis of Pomelo maturity. IEEE Trans Ind Inform 15:5971–5979
Chen H, Chen A, Xu L, Xie H, Qiao H, Lin Q, Cai K (2020a) A deep learning CNN architecture applied in smart near-infrared analysis of water pollution for agricultural irrigation resources. Agric Water Manag 240:106303
Chen H, Zhang G, Fan D et al (2020b) Nonlinear lamb wave analysis for microdefect identification in mechanical structural health assessment. Measurement 108026
Chen Y, Li J, Lu H, Yan P (2021) Coupling system dynamics analysis and risk aversion programming for optimizing the mixed noise-driven shale gas-water supply chains. J Clean Prod 278:123209
Cheng Y (2004) Pressure transient testing and productivity analysis for horizontal wells, (Doctoral dissertation, Texas A&M University).
Cheng X, He L, Lu H, Chen Y, Ren L (2016) Optimal water resources management and system benefit for the Marcellus shale-gas reservoir in Pennsylvania and West Virginia. J Hydrol 540:412–422
Ciftci E (2018) A practical approach for the interpretation of flowing well tests. Arab J Geosci 11:761
Daolun LI, Xuliang LIU, Wenshu ZHA et al (2020) Automatic well test interpretation based on convolutional neural network for a radial composite reservoir. Pet Explor Dev 47:623–631
Deng Y, Zhang T, Sharma BK, Nie H (2019) Optimization and mechanism studies on cell disruption and phosphorus recovery from microalgae with magnesium modified hydrochar in assisted hydrothermal system. Sci Total Environ 646:1140–1154
Egya D, Geiger S, Corbett P, March R (2017) The effect of fracture skin, network connectivity, and network size on well-test responses in naturally fractured reservoirs. In: 79th EAGE Conference and Exhibition 2017. European Association of Geoscientists & Engineers, 12-15 June, Paris, France
Engler T, Tiab D (1996) Analysis of pressure and pressure derivative without type curve matching, 4. Naturally fractured reservoirs. J Pet Sci Eng 15:127–138
Escobar FH, Zhao Y-L, Urazán C, Trujillo CM (2018) Pressure and pressure derivative interpretation for horizontal wells in compressible formations. J Geophys Eng 15:1551–1560
Eslamloueyan R, Vaferi B, Ayatollahi S (2010) Fracture characterizations from well testing data using artificial neural networks. In: 72nd European Association of Geoscientists and Engineers Conference and Exhibition 2010: A New Spring for Geoscience. Incorporating SPE EUROPEC 2010, 14-17 June, Barcelona, Spain
Ghaffarian N, Eslamloueyan R, Vaferi B (2014) Model identification for gas condensate reservoirs by using ANN method based on well test data. J Pet Sci Eng:123
Gramatica P (2007) Principles of QSAR models validation: internal and external. QSAR Comb Sci 26:694–701
Guo H, Li X, Zhu Q, Zhang Z, Liu Y, Li Z, Wen H, Li Y, Tang J, Liu J (2020a) Imaging nano-defects of metal waveguides using the microwave cavity interference enhancement method. Nanotechnology 31:455203
Guo T, Tang S, Liu S et al (2020b) Physical Simulation of Hydraulic Fracturing of Large-Sized Tight Sandstone Outcrops. SPE J
He L, Chen Y, Zhao H, Tian P, Xue Y, Chen L (2018) Game-based analysis of energy-water nexus for identifying environmental impacts during shale gas operations under stochastic input. Sci Total Environ 627:1585–1601
Huang S, Yao Y, Zhang S, Ji J, Ma R (2018) Pressure transient analysis of multi-fractured horizontal wells in tight oil reservoirs with consideration of stress sensitivity. Arab J Geosci 11:285
Khan M, Arif M, Ali N, Yaseen M, Ahmed A, Siyar SM (2016) Petrophysical parameters and modelling of the Eocene reservoirs in the Qadirpur area, Central Indus Basin, Pakistan: implications from well log analysis. Arab J Geosci 9:425
Khandelwal M, Kankar PK (2011) Prediction of blast-induced air overpressure using support vector machine. Arab J Geosci 4:427–433
Landa JL, Horne RN, Kamal MM, Jenkins CD (2000) Reservoir characterization constrained to well-test data: a field example. SPE Reserv Eval Eng 3:325–334
Li T, Xu M, Zhu C, Yang R, Wang Z, Guan Z (2019) A deep learning approach for multi-frame in-loop filter of HEVC. IEEE Trans Image Process 28:5663–5678
Li Z, Liu H, Dun Z, Ren L, Fang J (2020) Grouting effect on rock fracture using shear and seepage assessment. Constr Build Mater 242:118131
Liu E, Lv L, Yi Y, Xie P (2019a) Research on the steady operation optimization model of natural gas pipeline considering the combined operation of air coolers and compressors. IEEE Access 7:83251–83265
Liu Y, Chen S, Guan B, Xu P (2019b) Layout optimization of large-scale oil–gas gathering system based on combined optimization strategy. Neurocomputing 332:159–183
Liu X, Li D, Yang J, Zha W, Zhou Z, Gao L, Han J (2020) Automatic well test interpretation based on convolutional neural network for infinite reservoir. J Pet Sci Eng 195:107618
Lv Z, Qiao L (2020) Deep belief network and linear perceptron based cognitive computing for collaborative robots. Appl Soft Comput 92:106300
Moosavi SR, Qajar J, Riazi M (2018a) A comparison of methods for denoising of well test pressure data. J Pet Explor Prod Technol 8:1519–1534
Moosavi SR, Vaferi B, Wood DA (2018b) Applying orthogonal collocation for rapid and reliable solutions of transient flow in naturally fractured reservoirs. J Pet Sci Eng 162:166–179
Moosavi SR, Vaferi B, Wood DA (2020) Auto-detection interpretation model for horizontal oil wells using pressure transient responses. Adv Geo-Energy Res 4:305–316
Nategh M, Vaferi B, Riazi M (2019) Orthogonal collocation method for solving the diffusivity equation: Application on dual porosity reservoirs with constant pressure outer boundary. ASME J Energy Resour Technol 141:042001
Qian J, Feng S, Li Y, Tao T, Han J, Chen Q, Zuo C (2020a) Single-shot absolute 3D shape measurement with deep-learning-based color fringe projection profilometry. Opt Lett 45:1842–1845
Qian J, Feng S, Tao T, Hu Y, Li Y, Chen Q, Zuo C (2020b) Deep-learning-enabled geometric constraints and phase unwrapping for single-shot absolute 3D shape measurement. APL Photonics 5:46105
Qu S, Han Y, Wu Z, Raza H (2020) Consensus modeling with asymmetric cost based on data-driven robust optimization. Gr Decis Negot 19:1–38
Quan Q, Hao Z, Xifeng H, Jingchun L (2020) Research on water temperature prediction based on improved support vector regression. Neural Comput Applic 28:1–10
Rousseeuw PJ, Leroy AM (2005) Robust regression and outlier detection. John wiley & sons, New York
Şahin AU, Çiftçi E (2016) An integration-based estimation approach for the determination of slug test parameters under various flow geometries. Arab J Geosci 9:639
Saidi AM (1983) Simulation of naturally fractured reservoirs. SPE Res Eng 3(02):638–648
Shi K, Wang J, Zhong S, Tang Y, Cheng J (2020) Non-fragile memory filtering of TS fuzzy delayed neural networks based on switched fuzzy sampled-data control. Fuzzy Sets Syst 394:40–64
Sierra O (1986) Fundamentals of well-log interpretation: The Interpretation of Logging Data, Elsevier, Amsterdam
Smola AJ, Schölkopf B (2004) A tutorial on support vector regression. Stat Comput 14:199–222
Su Z, Liu E, Xu Y, Xie P, Shang C, Zhu Q (2019) Flow field and noise characteristics of manifold in natural gas transportation station. Oil Gas Sci Technol d’IFP Energies Nouv 74:70
Tiab D (1989) Direct type-curve synthesis of pressure transient tests. In: Low Permeability Reservoirs Symposium. Society of Petroleum Engineers
Tiab D (1994) Analysis of pressure and pressure derivative without type-curve matching: Vertically fractured wells in closed systems. J Pet Sci Eng 11:323–333
Tiab D, Donaldson EC (2015) Petrophysics: theory and practice of measuring reservoir rock and fluid transport properties, 3rd ed. Gulf professional publishing, Amsterdam
Tikhamarine Y, Souag-Gamane D, Kisi O (2019) A new intelligent method for monthly streamflow prediction: hybrid wavelet support vector regression based on grey wolf optimizer (WSVR–GWO). Arab J Geosci 12:540
Vaferi B, Eslamloueyan R, Ayatollahi S (2011) Automatic recognition of oil reservoir models from well testing data by using multi-layer perceptron networks. J Pet Sci Eng 77:254–262
Vaferi B, Eslamloueyan R, Ghaffarian N (2016) Hydrocarbon reservoir model detection from pressure transient data using coupled artificial neural network-Wavelet transform approach. Appl Soft Comput J 47:63–75
Wang F (2016) Continuous reservoir model calibration with time-dependent reservoir properties diagnosed from long-term down-hole transient pressure data. Arab J Geosci 9:254
Warren JE, Root PJ (1963) The behavior of naturally fractured reservoirs. Soc Pet Eng J 3:245–255
Xu X, Chen L (2019) Projection of long-term care costs in China, 2020–2050: based on the Bayesian quantile regression method. Sustainability 11:3530
Xu S, An X, Qiao X, Zhu L, Li L (2013) Multi-output least-squares support vector regression machines. Pattern Recogn Lett 34:1078–1084
Xu M, Li T, Wang Z, Deng X, Yang R, Guan Z (2018) Reducing complexity of HEVC: a deep learning approach. IEEE Trans Image Process 27:5044–5059
Yan J, Pu W, Zhou S, Liu H, Greco MS (2020) Optimal resource allocation for asynchronous multiple targets tracking in heterogeneous radar networks. IEEE Trans Signal Process 68:4055–4068
Yang S, Deng B, Wang J et al (2019) Scalable digital neuromorphic architecture for large-scale biophysically meaningful neural network with multi-compartment neurons. IEEE Trans Neural Netw Learn Syst 31:148–162
Yang Y, Liu J, Yao J, Kou J, Li Z, Wu T, Zhang K, Zhang L, Sun H (2020) Adsorption behaviors of shale oil in kerogen slit by molecular simulation. Chem Eng J 387:124054
Yue H, Wang H, Chen H et al (2020) Automatic detection of feather defects using Lie group and fuzzy Fisher criterion for shuttlecock production. Mech Syst Signal Process 141:106690
Zhang X, Wang Y, Chen X et al (2018) Decentralized adaptive neural approximated inverse control for a class of large-scale nonlinear hysteretic systems with time delays. IEEE Trans Syst Man, Cybern Syst 49:2424–2437
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author (s) declare that they have no competing interests.
Additional information
Responsible Editor: Santanu Banerjee
Appendix 1. Multi-output least squares support vector regression
Appendix 1. Multi-output least squares support vector regression
Support vector regression maps nonlinear patterns into higher dimensional feature space that can approach infinite dimensions (Khandelwal and Kankar 2011; Tikhamarine et al. 2019; Quan et al. 2020). It then applies linear regression to the mapped feature space (Chao et al. 2018; Xu and Chen 2019). Considering a dataset of N records existing in multi-dimensional feature space, its ith data record or element can be expressed as:
Here, xi and yi represent an actual and predicted value of the ith data record. For such a data set, the support vector regression can be expressed as follows (Xu et al. 2013):
where <,> denotes the dot product of the matrix elements involving all the x data records, w is the weight vector of the SVR regression function, and b is an intercept of the SVR regression function.
Accepting a certain level of error (ε), the objective of SVR is to establish a function F(x) that estimates the values of y from x data for a training dataset that maintains deviations at or less than the value specified for ε. The risk function (R) can then be solved using appropriate optimization techniques (Chen et al. 2017; Deng et al. 2019; Liu et al. 2019a; Cao et al. 2020a, b, c; Qu et al. 2020).
where:
Equation (4) expresses an error insensitive loss function. ε determines the regression’s precision by essentially defining the radius of a cylinder surrounding the regression function, f(x), within which acceptable values may exist. By substitution of Eq. (3) into Eq. (2), it is possible to determine functions that fit the records of the SVR training subset with deviations of no more than ε.
The acceptable error components associated with SVR minimization can be further defined as follows (Smola and Schölkopf 2004; Chen et al. 2019):
This minimization involves the constraints listed in Eq. (5):
where C is a positive regularization constant, and are positive slack variables.
C is a metric that establishes a trade-off between a solution’s ability to be generalized across all the elements of a data subset (e.g., the SVR training subset) versus achieving acceptable levels of accuracy (expressed in terms of error tolerance by ε). and quantify the distance from the values to theboundaries values of the error cylinder defined by ε.
SVR can then be expressed as a dual problem in the form of Eq. (6) to be maximized (Burges 1998; Bian et al. 2016):
Here, αi and \( {\upalpha}_{\mathrm{i}}^{\ast } \) are Lagrange multipliers derived from quadratic equation solutions.
The SVR single space function can then be mathematically expressed by Eq. (8):
The SVR dual space regression function is expressed by Eq. (9):
where k (xi, x) indicates the kernel function that satisfies Mercer’s conditions.
Those data records in the dataset determined by Eq. (9) to have non-zero coefficients are the support vectors. A kernel function commonly applied in Eq. (9) is the radial basis function (RBF) expressed as Eq. (10):
where γ shows the width of the RBF.
The critical variables in establishing acceptable SVR optimization solutions (regression functions) and the complexity of those solutions are ε (Eq. 3), C (Eq. 4), and γ (Eq. 10). SVR optimization, therefore, focuses upon optimizing the variables ε, C, and γ.
In MLS-SVR, each data record in the dataset has multiple independent-dependent variables (Yan et al. 2020). For the case of our study, the digitized pressure derivative curve with 28 points is the independent variable, while ω and λ constitute the target vector.
Rights and permissions
About this article
Cite this article
Moosavi, S.R., Vaferi, B. & Wood, D.A. Auto-characterization of naturally fractured reservoirs drilled by horizontal well using multi-output least squares support vector regression. Arab J Geosci 14, 545 (2021). https://doi.org/10.1007/s12517-021-06559-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12517-021-06559-9