Rock Mechanics
Farhad Mollaei; Ali Moradzadeh; Reza Mohebian
Abstract
The important aspects of this study are to estimate the mechanical parameters of reservoir rock including Uniaxial Compressive Strength (UCS) and friction (FR) angle using well log data. The aim of this research is to estimate the UCS and FR angle (φ) using new deep learning (DL) methods including ...
Read More
The important aspects of this study are to estimate the mechanical parameters of reservoir rock including Uniaxial Compressive Strength (UCS) and friction (FR) angle using well log data. The aim of this research is to estimate the UCS and FR angle (φ) using new deep learning (DL) methods including Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and CNN + LSTM (CL) by well log and core test data of one Iranian hydrocarbon field. As only 12 UCS and 6 FR core tests of single well in this field were available, they were firstly calculated, and then generalized to other depths using two newly derived equations and relevant logs. Next, the effective input logs' data for predicting these parameters have been selected by an auto-encoder DL method, and finally, the values of UCS and φ angle were predicted by the MLP, LSTM, CNN, and CL networks. The efficiency of these four prediction models was then evaluated using a blind dataset, and a range of statistical measures applied to training, testing, and blind datasets. Results show that all four models achieve satisfactory prediction accuracy. However, the CL model outperformed the others, yielding the lowest RMSE of 1.0052 and the highest R² of 0.9983 for UCS prediction, along with an RMSE of 0.0201 and R² of 0.9917 for φ angle prediction on the blind dataset. These findings highlight the high accuracy of deep learning algorithms, particularly the CL algorithm, which demonstrates superior precision compared to the MLP method.
Exploration
Shaghayegh Esmaeilzadeh; Ali Moradzadeh; omid Asghari; Reza Mohebian
Abstract
Seismic inversion is a critical technique for estimating the spatial distribution of petro-elastic properties in the subsurface, based on the seismic reflection data. This work introduces an iterative geostatistical seismic inversion method, designed to address challenges in complex geological settings ...
Read More
Seismic inversion is a critical technique for estimating the spatial distribution of petro-elastic properties in the subsurface, based on the seismic reflection data. This work introduces an iterative geostatistical seismic inversion method, designed to address challenges in complex geological settings by incorporating self-updating local variogram models. Unlike the conventional approaches that rely on a single global variogram or fixed local variograms, the proposed method dynamically updates the spatial continuity models at each iteration using automatic variogram modeling and clustering of variogram parameters. The optimal number of clusters is determined using three cluster validity indices: Silhouette Index (SI), Davies-Bouldin Index (DB), and Calinski-Harabasz Index (CH). The method’s effectiveness was evaluated using a three-dimensional non-stationary synthetic dataset, demonstrating robust convergence when employing the SI and CH indices, with both achieving a high global correlation coefficient of 0.9 between the predicted and true seismic data. Among these, the CH index provided the best balance between the computational efficiency and inversion accuracy. The results highlight the method’s ability to effectively capture local spatial variability, while maintaining a reasonable computational cost, making it a promising approach for seismic inversion in complex sub-surface environments.
Y. Asgari Nezhad; A. Moradzadeh
Abstract
One of the most essential factors involved in unconventional gas reserves for drilling and production is a suitable quality facies determination. The direct core and geochemical analyses are the most common methods used for studying this quality. Due to the lack of this data and the high cost, the researchers ...
Read More
One of the most essential factors involved in unconventional gas reserves for drilling and production is a suitable quality facies determination. The direct core and geochemical analyses are the most common methods used for studying this quality. Due to the lack of this data and the high cost, the researchers have recently resorted to the indirect methods that use the common data of the reservoir (including petro-physical logs and seismic data). One of the major problems in using these methods is that the complexities of these reproducible repositories cannot be accurately modeled. In this work, the quality of facies in shale gas is zoned using the deep learning technique. The applied method is long short-term memory (LSTM) neural network. In this scheme, the features required for zoning are automatically extracted and used to model the reservoir complexities properly. The results of this work show that zoning is done with an appropriate accuracy (86%) using the LSTM neural network, while it is 78% for a conventional intelligent MLP network. This specifies the superior accuracy of the deep learning method.
F. Moradpouri; A. Moradzadeh; R. Cruz Pestana; M. Soleimani Monfared
Abstract
In this paper, first the limitations of the ray-based method and the one-way wave-field extrapolation migration (WEM) in imaging steeply dipping structures are discussed by some examples. Then a new method of the reverse time migration (RTM), used in imaging such complex structures is presented. The ...
Read More
In this paper, first the limitations of the ray-based method and the one-way wave-field extrapolation migration (WEM) in imaging steeply dipping structures are discussed by some examples. Then a new method of the reverse time migration (RTM), used in imaging such complex structures is presented. The proposed method uses a new wave-field extrapolator called the Leapfrog-Rapid Expansion Method (L-REM) for wave-field extrapolation. This improved method also includes a new imaging condition based on Poynting vector for wave-field separation and calculating the reflection angles. Afterwards, the results obtained for the application of the new RTM method are compared with those obtained by the harmonic-source method as a delay shot or plane wave RTM. Finally, the efficiency of these imaging methods is tested using the BP 2004 2D seismic dataset. The results obtained indicate the superiority of the presented RTM method in imaging such steep dip structures in comparison with the other imaging procedures.
H. Sabeti; A. Moradzadeh; F. Doulati Ardejani; A. Soares
Abstract
Stochastic seismic inversion is a family of inversion algorithms in which the inverse solution was carried out using geostatistical simulation. In this work, a new 3D stochastic seismic inversion was developed in the MATLAB programming software. The proposed inversion algorithm is an iterative procedure ...
Read More
Stochastic seismic inversion is a family of inversion algorithms in which the inverse solution was carried out using geostatistical simulation. In this work, a new 3D stochastic seismic inversion was developed in the MATLAB programming software. The proposed inversion algorithm is an iterative procedure that uses the principle of cross-over genetic algorithms as the global optimization technique. The model perturbation towards the objective function is performed recurring to direct sequential simulation and co-simulation. This new algorithm was applied to a synthetic dataset with and without noise. The results obtained for the inverted impedance were satisfactory in both cases. In addition, a real dataset was chosen to be applied by the algorithm. Good results were achieved regarding the real dataset. For the purpose of validation, blind well tests were done for both the synthetic and real datasets. The results obtained showed that the algorithm was able to produce inverted impedance that fairly matched the well logs. Furthermore, an uncertainty analysis was performed for both the synthetic and real datasets. The results obtained indicate that the variance of acoustic impedance is increased in areas far from the well location.
M. Rezaie; A. Moradzadeh; A. Nejati Kalate
Abstract
One of the most remarkable basis of the gravity data inversion is the recognition of sharp boundaries between an ore body and its host rocks during the interpretation step. Therefore, in this work, it is attempted to develop an inversion approach to determine a 3D density distribution that produces a ...
Read More
One of the most remarkable basis of the gravity data inversion is the recognition of sharp boundaries between an ore body and its host rocks during the interpretation step. Therefore, in this work, it is attempted to develop an inversion approach to determine a 3D density distribution that produces a given gravity anomaly. The subsurface model consists of a 3D rectangular prisms of known sizes and positions and unknown density contrasts that are required to be estimated. The proposed inversion scheme incorporates the Cauchy norm as a model norm that imposes sparseness and the depth weighting of the solution. A physical-bound constraint is enforced using a generic transformation of the model parameters. The inverse problem is posed in the data space, leading to a smaller dimensional linear system of equations to be solvedand a reduction in the computation time. For more efficiency, the low-dimensional linear system of equations is solved using a fast iterative method such as Lanczos Bidiagonalization. The tests carried out on the synthetic data show that the sparse data-space inversion produces blocky and focused solutions. The results obtained for the 3D inversion of the field gravity data (Mobrun gravity data) indicate that the sparse data-space inversion could produce the density models consistent with the true structures.
A. Khojamli; F. Doulati Ardejani; A. Moradzadeh; A. Nejati Kalateh; A. Roshandel Kahoo; S. Porkhial
Abstract
The Ardabil geothermal area is located in the northwest of Iran, which hosts several hot springs. It is situated mostly around the Sabalan Mountain. The Sabalan geothermal area is now under investigation for the geothermal electric power generation. It is characterized by its high thermal gradient and ...
Read More
The Ardabil geothermal area is located in the northwest of Iran, which hosts several hot springs. It is situated mostly around the Sabalan Mountain. The Sabalan geothermal area is now under investigation for the geothermal electric power generation. It is characterized by its high thermal gradient and high heat flow. In this study, our aim is to determine the fractal parameter and top and bottom depths of the magnetic sources. A modified spectral analysis technique named “de-fractal spectral depth method” is developed and used to estimate the top and bottom depths of the magnetized layer. A mathematical relationship is used between the observed power spectrum (due to fractal magnetization) and an equivalent random magnetization power spectrum. The de-fractal approach removes the effect of fractal magnetization from the observed power spectrum, and estimates the parameters of the depth to top and depth to bottom of the magnetized layer using the iterative forward modelling of the power spectrum. This approach is applied to the aeromagnetic data of the Ardebil province. The results obtained indicated variable magnetic bottom depths ranging from 10.4 km in the northwest of Sabalan to about 21.1 km in the north of the studied area. In addition, the fractal parameter was found to vary from 3.7 to 4.5 within the studied area.
