Model development for shear sonic velocity using geophysical log data: Sensitivity analysis and statistical assessment

doi:10.1016/j.jngse.2020.103778

Journal of Natural Gas Science and Engineering

Volume 88, April 2021, 103778

https://doi.org/10.1016/j.jngse.2020.103778 Get rights and content

Highlights

•
Hybrid models are implemented to assess V_s prediction performance.
•
Both regression analysis and LSSVM-CSA are applied to rank the log variables.
•
V_p and $\emptyset_{N}$ are the most significant parameters affecting V_s magnitude.
•
Predictions and variable ranking are in excellent agreement with the field log data.
•
A new correlation is introduced to predict V_s for clastic formation cases.

Abstract

Reservoir geomechanical parameters play a vital role in evaluating sanding potential, wellbore stability, and drilling performance for making decisions on development strategies of oil and gas fields. Compared to expensive experimental/laboratory investigations, the petrophysical log data-driven shear sonic velocity models are proven to be cost effective and quick to estimate geomechanical formation properties. The objective of this study is to introduce a new deterministic model using the most contributing predictor variables and to rank the variables according to their relative importance while predicting dynamic shear sonic wave velocity for clastic sedimentary rocks. The least-squares support vector machine with a global optimization technique is used to construct a data-driven model for shear sonic velocity estimation. A new regressive correlation is also proposed. The model performance is assessed using statistical parameters to ensure the model accuracy and reliability. Based on the relative contribution of log variables, the influential variables are ranked (higher to lower order) as follows: acoustic compressional sonic velocity, porosity, and bulk density. The developed model is validated; the predictions are compared with real data and existing log-based correlations for clastic sedimentary rocks using data from two real fields, namely the North Sea and the Niger Delta basins. The developed shear wave velocity model will assist geomechanists and drilling engineers in obtaining accurate formation elastic properties and rock strength, which are required for wellbore failure analysis as well as assessment of sanding occurrence during exploration and drilling.

Section snippets

Credit author statement

Mohammad Islam Miah: Conceptualization, Data curation, Methodology, Modeling/Simulation, Validation, Writing – original draft; Salim Ahmed: Supervision, Technical Discussion, Writing- Reviewing and Editing; Sohrab Zendehboudi: Supervision, Technical Discussion, Writing- Reviewing and Editing

Fundamentals of wireline logs and quality assurance of log data

The formation properties such as rock porosity, permeability, and acoustic travel time can be obtained through experimental core analysis and/or using petrophysical wireline log data. The logs are reliable to estimate in-situ porosity, permeability, and acoustic velocities in the absence of core data. The most common petrophysical logs are natural gamma-ray (GR), resistivity, sonic, neutron, and density logs, which are generally used for formation analysis and geomechanical properties

Data quality

The wireline log data such as gamma-ray, formation bulk density, neutron porosity, acoustic compressional, and shear sonic velocities are considerably changed with formation depth due to the complex behavior and heterogeneity of shaly sand sedimentary rocks. For the data set under study, the statistical information on the log data samples are presented in Table 2 for the entire depth of the formation. The correlation matrix between V_s and other formation properties (e.g., gamma-ray, neutron

Conclusions

In this study, Gaussian radial basis kernel function (RBF) - least square support vector machine (LSSVM) linked with coupled simulated annealing (CSA) optimization technique is employed to obtain the dynamic shear sonic velocity (V_s) using real field petrophysical log data such as gamma-ray, neutron porosity, bulk density, and compressional sonic velocity. The LSSVM approach also finds the most influential parameters and ranks them based on their relative significance. In the current research,

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We thank Equinor Canada Ltd., Natural Sciences and Engineering Research Council of Canada (NSERC), Memorial University, and InnovateNL for providing financial support, and also the Chittagong University of Engineering and Technology, Bangladesh for research support to accomplish the study.

References (97)

A.K. Abbas et al.
Estimating rock mechanical properties of the Zubair shale formation using a sonic wireline log and core analysis
J. Nat. Gas Sci. Eng.
(2018)
K.O. Akande et al.
Investigating the effect of correlation-based feature selection on the performance of support vector machines in reservoir characterization
J. Nat. Gas Sci. Eng.
(2015)
K.O. Akande et al.
Investigating the effect of correlation-based feature selection on the performance of neural network in reservoir characterization
J. Nat. Gas Sci. Eng.
(2015)
M.S. Ameen et al.
Predicting rock mechanical properties of carbonates from wireline logs (A case study: arab-D reservoir, Ghawar field, Saudi Arabia)
Mar. Petrol. Geol.
(2009)
M. Amiri et al.
Water saturation estimation in tight shaly gas sandstones by application of Progressive Quasi-Static (PQS) algorithm–A case study
J. Nat. Gas Sci. Eng.
(2015)
M. Anemangely et al.
Shear wave travel time estimation from petrophysical logs using ANFIS-PSO algorithm: a case study from Ab-Teymour Oilfield
J. Nat. Gas Sci. Eng.
(2017)
M. Anemangely et al.
Machine learning technique for the prediction of shear wave velocity using petrophysical logs
J. Petrol. Sci. Eng.
(2019)
F.A. Anifowose et al.
Non-linear feature selection-based hybrid computational intelligence models for improved natural gas reservoir characterization
J. Nat. Gas Sci. Eng.
(2014)
F.A. Anifowose et al.
Ensemble model of non-linear feature selection-based extreme learning machine for improved natural gas reservoir characterization
J. Nat. Gas Sci. Eng.
(2015)
F. Anifowose et al.
Non-linear feature selection based hybrid computational intelligence models for improved natural gas reservoir characterization
J. Nat. Gas Sci. Eng.
(2014)

K. Aoyagi et al.

Clay minerals in carbonate reservoir rocks and their significance in porosity studies

Sediment. Geol.

(1972)

M. Arabloo et al.

SVM modeling of the constant volume depletion (CVD) behavior of gas condensate reservoirs

J. Nat. Gas Sci. Eng.

(2014)

M. Arabloo et al.

A novel modeling approach to optimize oxygen–steam ratios in coal gasification process

Fuel

(2015)

P. Bagheripour et al.

Support vector regression based determination of shear wave velocity

J. Petrol. Sci. Eng.

(2015)

R. Baouche et al.

Distribution of pore pressure and fracture pressure gradients in the paleozoic sediments of Takouazet field, Illizi basin, Algeria

J. Afr. Earth Sci.

(2020)

R. Baouche et al.

Integrated reservoir characterization of the paleozoic and mesozoic sandstones of the el ouar field, Algeria

J. Petrol. Sci. Eng.

(2020)

R. Baouche et al.

Characterization of Pore Pressure, Fracture Pressure, Shear Failure and its Implications for Drilling, Wellbore Stability and Completion Design–A Case Study from the Takouazet Field, Illizi Basin, Algeria

(2020)

Y. Cai et al.

A data mining approach to finding relationships between reservoir properties and oil production for CHOPS

Comput. Geosci.

(2014)

R.D. Carroll

(2020)

D. Moos et al.

Comprehensive wellbore stability analysis utilizing quantitative risk assessment

J. Petrol. Sci. Eng.

(2003)

A. Movahhed et al.

Introducing a method for calculating water saturation in a carbonate gas reservoir

J. Nat. Gas Sci. Eng.

(2019)

M. Nikravesh et al.

(2019)

P. Wang et al.

On a new method of estimating shear wave velocity from conventional well logs

J. Petrol. Sci. Eng.

(2019)

Cited by (12)

Early warning of drillstring faulty conditions based on multi-model fusion in geological drilling processes
2023, Journal of Process Control
The early warning of drilling faults is of paramount importance to prevent or reduce environmental and property damages. In view of the complex drillstring kinematic characteristics and changeable formation environment, this paper proposes a new fault early warning method for the drillstring system based on a multi-model fusion and self-updating strategy. The major contributions are twofold: (1) A hybrid drillstring early warning method is proposed to identify faulty conditions based on the fusion of multi-dimensional prediction models of the drillstring system; (2) An event-triggered model self-updating strategy is proposed to mitigate the prediction performance degradation caused by formation uncertainties. The effectiveness and practicability of the proposed method are demonstrated by industrial case studies, and the results indicate that the proposed method outperforms other approaches.
P-wave prediction method under multi-source spatiotemporal feature fusion and physics-informed neural network
2023, Geoenergy Science and Engineering
Compressional wave velocity (P-wave) plays an important role in seismic amplitude versus offset (AVO) forward and inversion modeling and reservoir prediction. However, due to the high cost and difficulty of P-wave logging, the actual acquired P-wave data is limited, which generally does not suffice for practical applications. To provide reliable data support for oil and gas reservoir exploration, a P-wave prediction method under multi-source spatiotemporal feature fusion and physics-informed neural network (MSTFPIN) is proposed. First, a spatiotemporal hybrid model (STHN) is constructed by combining convolutional neural network (CNN) and bidirectional long short-term memory (BLSTM) neural network to accurately express the spatial and temporal characteristics of the logging curves, which utilizes the local perception ability of CNN and long short-term memory function of BLSTM to mine their local shape and depth trends. Then, seismic data are introduced as an additional input of multi-source information, which fills the gap of inter-well geological information, and feature fusion is used to fully exploit the information from multiple sources to enhance the available data. Finally, a physics-informed neural network (PINN) is constructed based on the idea of seismic forward modeling in petroleum engineering to guide the training process of the prediction model, which ensures that the prediction results are consistent with geological laws. P-wave prediction experiments are conducted using real logging and seismic data to verify the effectiveness and generalization ability of the proposed method compared with other methods. The experimental results show that the proposed MSTFPIN model exhibit better prediction performance, and the correlation coefficient and coefficient of determination reach 0.9619 and 0.9234, respectively. Compared with traditional empirical and regression models, the proposed MSTFPIN exhibited higher accuracy and generalization. Overall, our results demonstrate the proposed method can be implemented in real oil fields to improve the accuracy of reservoir prediction.
Improved prediction of shear wave velocity for clastic sedimentary rocks using hybrid model with core data
2021, Journal of Rock Mechanics and Geotechnical Engineering
Citation Excerpt :
The most commonly used supervised machine learning tools are artificial neural network (ANN), convolution neural network (CNN), general regression neural network (GRNN), fuzzy logic (FL), neuro-fuzzy (NF), support vector machine (SVM), genetic algorithm (GA), gene expression programming (GEP), and random forest (RF). Additionally, hybrid connectionist models such as adaptive neuro-fuzzy inference system (ANFIS), least squares support vector machine (LSSVM) with optimization techniques of particle swarm optimization (PSO), cuckoo optimization algorithm (COA) or coupled simulated annealing (CSA), ANN-GA, ANN with wavelet transform (WT), SVM-GA and ant colony–fuzzy inference system (ACOFIS) are also applied to different disciplines such as rock mechanics (Yang and Zhang, 1997; Meulenkamp and Grima, 1999; Yilmaz and Yuksek, 2009; Cranganu and Bautu; 2010; Ocak and Seker, 2012; Khandelwal and Monjezi, 2013; Kumar et al., 2013; Maleki et al., 2014; Yurdakul et al., 2014; Anemangely et al., 2019; Dumke and Berndt, 2019; Lawal and Kwon, 2021; Miah et al., 2021), mining (Zhang et al., 2020), and geotechnical engineering (Zhang et al., 2021a; Zheng et al., 2021). Due to the considerations of computational time and cost-effective manner, above tools are broadly used to develop the data-driven model for obtaining rock mechanical properties including wave velocities related to rock elastic constants, brittleness performance, rock failure analysis for safe drilling operations, as well as rock mass characterization (Gaviglio, 1989; Hoek and Brown, 1997; Sonmez et al., 2004; Saadat et al., 2014; Kim et al., 2017; Zoveidavianpoor, 2017; Onalo, 2019; Bukar et al., 2019; Mews et al., 2019; Sulaimon and Teng, 2020; Miah et al., 2020a,b; Onalo et al., 2020; Rezaee et al., 2020; Zhang et al., 2021b).
Accurate measurement of acoustic velocities of sedimentary rocks is essential for prediction of rock elastic constants and well failure analysis during drilling operations. Direct measurement by advanced logging tools such as dipole sonic imager is not always possible. For older wells, such data are not available in most cases. Therefore, it is an alternate way to develop a reliable correlation to estimate the shear wave velocity from existing log and/or core data. The objective of this research is to investigate the nature of dependency of different reservoir parameters on the shear wave velocity (V_s) of clastic sedimentary rocks, and to identify the parameter/variable which shows the highest level of dependency. In the study, data-driven connectionist models are developed using machine learning approach of least square support vector machine (LSSVM). The coupled simulated annealing (CSA) approach is utilized to optimize the tuning and kernel parameters in the model development. The performance of the simulation-based model is evaluated using statistical parameters. It is found that the most dependency predictor variable is the compressional wave velocity, followed by the rock porosity, bulk density and shale volume in turn. A new correlation is developed to estimate V_s, which captures the most influential parameters of sedimentary rocks. The new correlation is verified and compared with existing models using measured data of sandstone, and it exhibits a minimal error and high correlation coefficient (R² = 0.96). The hybridized LSSVM-CSA connectionist model development strategy can be applied for further analysis to predict rock mechanical properties. Additionally, the improved correlation of V_s can be adopted to estimate rock elastic constants and conduct wellbore failure analysis for safe drilling and field development decisions, reducing the exploration costs.
Hybrid mathematical modelling of three-phase flow in porous media: Application to water-alternating-gas injection
2021, Journal of Natural Gas Science and Engineering
Citation Excerpt :
In another work, Esmaeli et al. (2019) used supervised (learning) LSSVM model to predict the effect of temperature on two-phase oil-water relative permeability data. The LSSVM model has also been used for other oil and gas related applications such as predicting the phase equilibrium conditions of clathrate hydrate (Sinehbaghizadeh et al., 2019), modelling freezing point depression of electrolyte solutions (Yarveicy et al., 2014), calculating minimum miscibility pressure for CO2-oil systems (Shokrollahi et al., 2013), and predicting shear sonic wave velocity for clastic sedimentary rocks (Miah et al., 2021). The fuzzy neural network models benefit from the learning capability in the ANNs and knowledge demonstration capability of the fuzzy logic (Salleh et al., 2017).
Machine learning algorithms are extensively used to reduce the complexity of applied problems in various fields, including energy. Accurate prediction of the performance of water alternating gas (WAG) injection as an enhanced oil recovery (EOR) process is of great importance in the optimal management of hydrocarbon resources. In the current work, a hybrid mathematical model is proposed for the near-immiscible WAG injection process. We use data-driven sub-models, including least square support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS) in series with an empirical model (EM) and a first principle model (FPM) to study three-phase flow in porous media. The LSSVM and ANFIS sub-models predict the two-phase water-oil, gas-oil, and gas-water relative permeabilities. The outputs from these models are supplied to the empirical models (EMs) to estimate the three-phase relative permeabilities for oil, gas, and water phases. The model developed using LSSVM shows a better prediction performance in estimating the relative permeabilities, compared to that using ANFIS. The relative importance parameter analysis reveals that for the LSSVM sub-model, water saturation is the most influencing input parameter for the gas-water and oil-water systems while for the gas-oil system, gas saturation is the most important input parameter. Using the models proposed in this work, some hybrid models are developed to forecast the ultimate recovery factor (RF) in the testing phase. The predicted ultimate RF values are 92.0%, 91.6%, and 82.9% for the correlation-based EM-FPM, LSSVM-EM-FPM, and ANFIS-EM-FPM hybrid models, respectively, in comparison to the measured ultimate RF value of 93.6% after three cycles of water and gasinjection. Among the proposed hybrid models, the LSSVM-EM-FPM model significantly removes the non-linearity of the two-phase relative permeabilities. In general, the LSSVM-EM-FPM hybrid model possesses the same level of accuracy as that of the EM-FPM hybrid model, but with significantly less model complexity and non-linearity. Thus, the LSSVM-EM-FPM hybrid model can be used in demanding applications such as optimization and control of the WAG injection process, leading to a better resource management.
A sandwich resonant bar method with transfer matrices for measuring the elastic parameters of rock at low frequency
2023, Journal of Geophysics and Engineering
Synthesizing Missing Travel Time of P-Wave and S-Wave: A Two-Stage Evolutionary Modeling Approach
2023, IEEE Sensors Journal

View all citing articles on Scopus

View full text