Machine and deep learning for estimating the permeability of complex carbonate rock from X-ray micro-computed tomography

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 1460 - 1472

Published: March 11, 2021

Accurate estimation of permeability is critical for oil and gas reservoir development management, as it controls production rate. After assessing numerical techniques ranging from pore network modeling (PNM) to the lattice Boltzmann method (LBM), an AI-based workflow developed a quick accurate complex carbonate rock its X-ray micro-computed tomography (micro-CT) image. Following features engineering using both image processing PNM, we trained tested on thousands segmented 3D micro-CT images shallow deep learning algorithms assess permeability. A broad variety supervised are implemented tested, including linear regression, support vector improved gradient boosting, convolutional neural networks. Additionally, explored hybrid physics-driven that takes into account petrophysical properties. Finally, found predicted by machine (ML) agrees very well with more computationally-intensive voxel-based direct simulation. In addition, ML model here provides substantial reduction in computation time roughly three orders magnitude compared LBM. This paper highlights crucial role played predicting properties learning. The proposed framework, integrating diverse algorithms, imaging, modeling, has potential quickly accurately estimate aid simulation characterization.

Language: Английский

---

Accurate prediction of indicators for engineering failures in complex mining environments

Engineering Failure Analysis, Journal Year: 2023, Volume and Issue: 155, P. 107736 - 107736

Published: Oct. 20, 2023

Language: Английский

---

A novel geophysical method for fractures mapping and risk zones identification in a coalmine, Northeast, China

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 3785 - 3804

Published: June 25, 2021

A novel approach for mapping induced fractures and associated lineaments has been introduced based on microseismic data, thus their contribution towards identification, prediction of the risk zones which may lead to a collapse or rockburst in coal mine. By taking Tashan Coal Mine (TCM) Datong coalfield China as an example, real-time data was acquired period 7 months. The processed using standard processing schemes followed by source location parameters estimation. proposed utilizes events, trace them map fractures, network interpret lineaments. Based configuration, geometry, network, thousands were identified vary length. Regional have interpreted orientation these fractures. distribution further correlated with computed energy distribution. spatial helped identification geological discontinuity (fault) that clearly separates study area into northern southern parts. reactivation this fault pose significant threats mass stability, life, property. half is characterized more short length while prolonged fewer upper dominated high (≥50, 000 J), lower part variable pattern (22,000 40, J). It inferred from energy-lineaments correlation figure, describes relatively stronger rock type comparison suggesting different deformational behavior. converging single point define where future can be expected. This provides strong basis catastrophic hazards mines. benefit global underground engineering researchers establishing early warning systems.

Language: Английский

---

Geophysical Characterization of Mining-Induced Complex Geological Deformations in a Deep Coalmine

Lithosphere, Journal Year: 2022, Volume and Issue: 2021(Special 4)

Published: Feb. 1, 2022

Abstract In this paper, a new way of characterizing the mining-driven complex geological deformations is presented by utilizing continuous real-time microseismic (MS) data acquired through passive seismological monitoring. The was processed, and source parameters were computed encompassing over 60,000 MS events. correlation between events regional stratigraphy reveals that Upper Middle Xishanyao Formations substantially contributed to induced seismicity. Most high-energy (>3000 J) are located 1200 1400 m in coal seams above, indicative formation macrofracture roof rock layers perturbing limits required for equilibrium state structure. Taking advantage recorded events, large number fractures mapped interpreted associated lineaments their deformation mechanism discussed. with energy showed an extremely erratic complicated pattern revealing different levels accumulated stress. Based on orientation geometry, classified into two broad systems provided guidance determining propagation directions deformation. in-depth analysis lineaments, fault identified length>700 various splays oriented serving as tool engineering. An upsurged shear stress decreased normal along may lead enhanced Furthermore, total 5 vulnerable zones example offering insights worldwide geohazard assessments.

Language: Английский

---

Application of Machine Learning Method of Data-Driven Deep Learning Model to Predict Well Production Rate in the Shale Gas Reservoirs

Energies, Journal Year: 2021, Volume and Issue: 14(12), P. 3629 - 3629

Published: June 18, 2021

Reservoir modeling to predict shale reservoir productivity is considerably uncertain and time consuming. Since we need simulate the physical phenomenon of multi-stage hydraulic fracturing. To overcome these limitations, this paper presents an alternative proxy model based on data-driven deep learning model. Furthermore, study not only proposes development process a model, but also verifies using field data for 1239 horizontal wells from Montney formation in Alberta, Canada. A neural network (DNN) multi-layer perceptron was applied cumulative gas production as dependent variable. The independent variable largely divided into four types: well information, completion fracturing data. It found that prediction performance better when principal component with contribution 85% analysis extracts important information multivariate data, predicting DNN 6 variables calculated through importance analysis. Hence, develop reliable sensitivity hyperparameters performed determine one-hot encoding, dropout, activation function, rate, hidden layer number neuron number. As result, best mean absolute percentage error improved at least 0.2% up 9.1%. novel approach can be other formations. useful guide economic future plans nearby reservoirs.

Language: Английский

---

Sand production control mechanisms during oil well production and construction

Petroleum Research, Journal Year: 2021, Volume and Issue: 6(4), P. 361 - 367

Published: March 7, 2021

Sand production is considered as one of the significant issues that significantly reduce wellbore productivity. The process sand or solids in operations crucial operational inefficiencies can lead to wells collapsing. Besides, drilling mud might erupt through formation. Therefore, it essential properly determine what types are produced correctly predict efficient control mechanisms. This paper aimed compare different mechanisms and how minimize production. Moreover, we consider injection pressure moisture on rate. According this study’s findings, increase had caused increase, which should be performances. Furthermore, chemical such resin hydrogel usually has methods. An expandable screen an three-layer component driven into well expanded.

Language: Английский

---

A fully multifield coupling model of gas extraction and air leakage for in-seam borehole

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 1293 - 1305

Published: Feb. 26, 2021

Air leakage around the in-seam borehole leads to a poor quality of gas extracted, thus outburst danger is difficult eliminate but also may cause other disasters, such as coal spontaneous combustion and explosion. Moreover, low concentration discharged into atmosphere always possesses an intense greenhouse effect. To investigate multiphase multifield coupling problem extraction, dual medium model was developed in this study describe gas–air mixture flow by considering influences matrix shrinkage effective stress on permeability. And calculated 3D numerical simulation presented ideal matching result with field data. Then, key factors affecting extraction were investigated model. It found that elevated increase sealing length diameter, reduced augmentation negative pressure Therefore, we improved technology, which can fully seal various fractures for improving extraction. The test indicated technology 1.91 times traditional method, still maintained relatively high level subsequent stage superior application results work help optimizing quality, great significance CBM exploitation utilization, reduce emission ensure safe mining.

Language: Английский

---

Numerical simulation study on rock-breaking process and mechanism of compound impact drilling

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 3137 - 3148

Published: June 3, 2021

Improving the rock-breaking efficiency of hard rock formation has always been concern in drilling. Compound impact drilling technology is a new method to achieve efficient crushing by exerting axial and torsional impacts on bit. In this study, investigate dynamic process mechanism under compound impact, numerical model was developed based cohesive elements. Dynamic tests sandstone were conducted verify effectiveness method. Then, simulated using proposed model. Finally, effects parameters cyclic loading performance systematically studied. The results indicate that stereoscopic effect can be achieved impact. amount broken increases with increase load amplitude times, whereas both angle duration have an optimal range. When 30°, 2.5 ms, maximum. A novel tool designed analysis mechanism, laboratory field carried out optimized.

Language: Английский

---

Theoretical and experimental investigation on fracture response of coal impacted by high-velocity water jet

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 3210 - 3224

Published: June 5, 2021

This paper adopted theoretical derivation and experiments to investigate the stress wave propagation characteristics in rock under water jet impact fracture response of coal. The evolution model displacement was presented analyze formation, attenuation during fragmentation. failure criterion for shear tensile dynamic loads, crack quasi-static pressure were established. And broken pit range formed by component damage caused obtained. Based on scanning electron microscopy binarization method, impinging coal various velocities conducted study patterns It is indicated that water-hammer its unloading stage, radial tangential decrease first then increase, accompanying occurrences stresses peaks maximum compression displacement. In subsequent stagnation trend be stable compressive stress, respectively. peak values at different stages all decay exponentially with distance. Moreover, surrounded annular cracks when velocity exceeds threshold value, while split will occur high velocity. rock-breaking specific energy consumption decreases increases increasing Besides, scope verified close experimental value before splitting Combined morphology debris sizes, two microscopic damage-failure modes impacted jets are revealed, namely failure.

Language: Английский

---

Characteristics of the fracture geometry and the injection pressure response during near-wellbore diverting fracturing

Energy Reports, Journal Year: 2021, Volume and Issue: 7, P. 491 - 501

Published: Jan. 19, 2021

Near-wellbore diverting fracturing (NWDF) can successfully generate new fractures and dramatically enhance the stimulation effects for vertical wells. Investigating fracture geometry injection pressure response is necessary candidate well selection effect evaluation. This work establishes a fully fluid–solid coupling model to simulate process of NWDF under various parameter combinations, including horizontal stress contrast, Young's modulus, rock tensile strength, permeability, pore pressure, rate. The verified against reported experimental results. Moreover, following key methods were applied in this study. Extended finite element method (XFEM) arbitrary propagation path, which mesh-free simulation method; initiation tip elements are characterized by cohesive zone model; truss propping proppants; plug proposed plugging tight plug. results show that be improved selecting low high strength. However, has no impact on NWDF. Enhancing fluid rate improve while should optimized certain well. During field operation NWDF, large difference between peak value steady denotes curvature bad result. No obvious straight perfect

Language: Английский

---