Complex fracture development related to stratigraphic architecture: Challenges for structural deformation prediction, Tensleep Sandstone at the Alcova anticline, Wyoming

The creation of large complex fracture networks by hydraulic fracturing is imperative for enhanced oil recovery from tight sand or shale reservoirs, tight gas extraction, and Hot-Dry-Rock (HDR) geothermal systems to improve the contact area to the rock matrix. Although conventional fracturing treatments may result in bi-wing fractures, there is evidence by microseismic mapping that fracture networks can develop in many unconventional reservoirs, especially when natural fracture systems are present and the differences between the principle stresses are low. However, not much insight is gained about fracture development as well as fluid and proppant transport in naturally fractured tight formations. In order to clarify the relationship between rock and treatment parameters, and resulting fracture properties, numerical simulations were performed using a commercial Discrete Fracture Network (DFN) simulator. A comprehensive sensitivity analysis is presented to identify typical fracture network patterns resulting from massive water fracturing treatments in different geological conditions. It is shown how the treatment parameters influence the fracture development and what type of fracture patterns may result from different treatment designs. The focus of this study is on complex fracture network development in different natural fracture systems. Additionally, the applicability of the DFN simulator for modeling shale gas stimulation and HDR stimulation is critically discussed. The approach stated above gives an insight into the relationships between rock properties (specifically matrix properties and characteristics of natural fracture systems) and the properties of developed fracture networks. Various simulated scenarios show typical conditions under which different complex fracture patterns can develop and prescribe efficient treatment designs to generate these fracture systems. Hydraulic stimulation is essential for the production of oil, gas, or heat from ultratight formations like shales and basement rocks (mainly granite). If natural fracture systems are present, the fracturing process becomes more complex to simulate. Our simulation results reveal valuable information about main parameters influencing fracture network properties, major factors leading to complex fracture network development, and differences between HDR and shale gas/oil shale stimulations.

Download Full-text

Numerical Investigation of Complex Hydraulic Fracture Development in Naturally Fractured Reservoirs

10.2118/spe-173326-ms ◽

2015 ◽

Author(s):

Wu Kan ◽

Jon E. Olson

Keyword(s):

Fracture Propagation ◽

Injection Pressure ◽

Natural Fracture ◽

Hydraulic Fractures ◽

Differential Stress ◽

Complex Fracture ◽

Fracture Patterns ◽

Fracture Development ◽

Fracture Complexity ◽

Shale Reservoirs

Abstract Complex fracture networks have become more evident in shale reservoirs due to the interaction between pre-existing natural and hydraulic fractures. Accurate characterization of fracture complexity plays an important role in optimizing fracturing design, especially for shale reservoirs with high-density natural fractures. In this study, we simulated simultaneous multiple fracture propagation within a single fracturing stage using a complex hydraulic fracture development model. The model was developed to simulate complex fracture propagation by coupling rock mechanics and fluid mechanics. A simplified three-dimensional displacement discontinuity method was implemented to more accurately calculate fracture displacements and fracture-induced dynamic stress changes than our previously developed pseudo-3d model. The effects of perforation cluster spacing, differential stress (SHmax - Shmin) and various geometry natural fracture patterns on injection pressure and fracture complexity were investigated. The single stage simulation results shown that (1) higher differential stress suppresses fracture length and increases injection pressure; (2) there is an optimal choice for the number of fractures per stage to maximize effective fracture surface area, beyond which increasing the number of fractures actually decreases effective fracture area; and (3) fracture complexity is a function of natural fracture patterns (various regular pattern geometries were investigated). Natural fractures with small relative angle to hydraulic fractures are more likely to control fracture propagation path. Also, natural fracture patterns with more long fractures tend to increase the likelihood to dominate the preferential fracture trend of fracture trajectory. Our numerical model can provide a physics-based complex fracture network that can be imported into reservoir simulation models for production analysis. The overall sensitivity results presented should serve as guidelines for fracture complexity analysis.

Download Full-text

Numerical Simulation of Complex Fracture Network Development by Hydraulic Fracturing in Naturally Fractured Ultratight Formations

Journal of Energy Resources Technology ◽

10.1115/1.4028690 ◽

2014 ◽

Vol 136 (4) ◽

Cited By ~ 32

Author(s):

Hannes Hofmann ◽

Tayfun Babadagli ◽

Günter Zimmermann

Keyword(s):

Fracture Network ◽

Natural Fracture ◽

Fracture Networks ◽

Network Development ◽

Complex Fracture ◽

Fracture Patterns ◽

Fracture Systems ◽

Fracture Development ◽

Complex Fracture Network ◽

Treatment Designs

The creation of large complex fracture networks by hydraulic fracturing is imperative for enhanced oil recovery from tight sand or shale reservoirs, tight gas extraction, and hot-dry-rock (HDR) geothermal systems to improve the contact area to the rock matrix. Although conventional fracturing treatments may result in biwing fractures, there is evidence by microseismic mapping that fracture networks can develop in many unconventional reservoirs, especially when natural fracture systems are present and the differences between the principle stresses are low. However, not much insight is gained about fracture development as well as fluid and proppant transport in naturally fractured tight formations. In order to clarify the relationship between rock and treatment parameters, and resulting fracture properties, numerical simulations were performed using a commercial discrete fracture network (DFN) simulator. A comprehensive sensitivity analysis is presented to identify typical fracture network patterns resulting from massive water fracturing treatments in different geological conditions. It is shown how the treatment parameters influence the fracture development and what type of fracture patterns may result from different treatment designs. The focus of this study is on complex fracture network development in different natural fracture systems. Additionally, the applicability of the DFN simulator for modeling shale gas stimulation and HDR stimulation is critically discussed. The approach stated above gives an insight into the relationships between rock properties (specifically matrix properties and characteristics of natural fracture systems) and the properties of developed fracture networks. Various simulated scenarios show typical conditions under which different complex fracture patterns can develop and prescribe efficient treatment designs to generate these fracture systems. Hydraulic stimulation is essential for the production of oil, gas, or heat from ultratight formations like shales and basement rocks (mainly granite). If natural fracture systems are present, the fracturing process becomes more complex to simulate. Our simulations suggest that stress state, in situ fracture networks, and fluid type are the main parameters influencing hydraulic fracture network development. Major factors leading to more complex fracture networks are an extensive pre-existing natural fracture network, small fracture spacings, low differences between the principle stresses, well contained formations, high tensile strength, high Young’s modulus, low viscosity fracturing fluid, and large fluid volumes. The differences between 5 km deep granitic HDR and 2.5 km deep shale gas stimulations are the following: (1) the reservoir temperature in granites is higher, (2) the pressures and stresses in granites are higher, (3) surface treatment pressures in granites are higher, (4) the fluid leak-off in granites is less, and (5) the mechanical parameters tensile strength and Young’s modulus of granites are usually higher than those of shales.

Download Full-text

A Deep Learning Prediction Model for Structural Deformation Based on Temporal Convolutional Networks

Computational Intelligence and Neuroscience ◽

10.1155/2021/8829639 ◽

2021 ◽

Vol 2021 ◽

pp. 1-12

Author(s):

Xianglong Luo ◽

Wenjuan Gan ◽

Lixin Wang ◽

Yonghong Chen ◽

Enlin Ma

Keyword(s):

Prediction Model ◽

Structural Engineering ◽

Orthogonal Experiment ◽

Deformation Monitoring ◽

Structural Deformation ◽

Long Term Memory ◽

Model Parameters ◽

Deformation Prediction ◽

Convolutional Networks ◽

Proposed Model

The structural engineering is subject to various subjective and objective factors, the deformation is usually inevitable, the deformation monitoring data usually are nonstationary and nonlinear, and the deformation prediction is a difficult problem in the field of structural monitoring. Aiming at the problems of the traditional structural deformation prediction methods, a structural deformation prediction model is proposed based on temporal convolutional networks (TCNs) in this study. The proposed model uses a one-dimensional dilated causal convolution to reduce the model parameters, expand the receptive field, and prevent future information leakage. By obtaining the long-term memory of time series, the internal time characteristics of structural deformation data can be effectively mined. The network hyperparameters of the TCN model are optimized by the orthogonal experiment, which determines the optimal combination of model parameters. The experimental results show that the predicted values of the proposed model are highly consistent with the actual monitored values. The average RMSE, MAPE, and MAE with the optimized model parameters reduce 44.15%, 82.03%, and 66.48%, respectively, and the average running time is reduced by 45.41% compared with the results without optimization parameters. The average RMSE, MAE, and MAPE reduce by 26.88%, 62.16%, and 40.83%, respectively, compared with WNN, DBN-SVR, GRU, and LSTM models.

Download Full-text

Effect of Flow Rate and Viscosity on Complex Fracture Development in UFM Model

Effective and Sustainable Hydraulic Fracturing ◽

10.5772/56406 ◽

2013 ◽

Cited By ~ 23

Author(s):

Olga Kresse ◽

Xiaowei Weng ◽

Dimitry Chuprakov ◽

Romain Prioul ◽

Charles Cohe

Keyword(s):

Flow Rate ◽

Complex Fracture ◽

Fracture Development

Download Full-text

Thermal monitoring and thermal deformation prediction for spherical machine tool spindles

Thermal Science ◽

10.2298/tsci1904271f ◽

2019 ◽

Vol 23 (4) ◽

pp. 2271-2279 ◽

Cited By ~ 1

Author(s):

Cheng-Biao Fu ◽

An-Hong Tian ◽

Her-Terng Yau ◽

Mao-Chin Hoang

Keyword(s):

Finite Element Analysis ◽

Finite Element ◽

Machine Tool ◽

Thermal Deformation ◽

Thermal Error ◽

Structural Deformation ◽

Thermal Monitoring ◽

Element Analysis ◽

Deformation Prediction ◽

Thermal Error Model

Machine tool operations and processing can cause temperature changes in various components because of internal and external thermal effects. Thermal deformations caused by thermal effect in machine tools can result in errors in processing size or shape and decrease processing precision. Thus, this paper focuses on the analysis of heating during machine tool spindle?s high speed operation, which is the heat source that causes component and structural deformation. In this paper, thermal monitoring was used to build a thermal error prediction model. Temperature change around the spindle was measured with a DS18B20, then multiple regression analysis was used to establish the relationship between thermal deformation quantity and temperature fields at specific points. Finally, finite element analysis was used to build the thermal error model. A solution for the correlation coefficient was obtained using the least squares method. The result of this study verified that finite element analysis can predict front bearing and rear bearing temperature rise, and is consistent with laboratory results. The error in thermal steady-state deformation prediction was less than 2 ?m. This information can be used by the controller to effectively compensate the processing and improve processing precision.

Download Full-text

A Prediction Model of Structural Settlement Based on EMD-SVR-WNN

Advances in Civil Engineering ◽

10.1155/2020/8831965 ◽

2020 ◽

Vol 2020 ◽

pp. 1-11

Author(s):

Xianglong Luo ◽

Wenjuan Gan ◽

Lixin Wang ◽

Yonghong Chen ◽

Xue Meng

Keyword(s):

Prediction Model ◽

Prediction Accuracy ◽

Deformation Monitoring ◽

Monitoring Data ◽

Structural Deformation ◽

Random Fluctuation ◽

Structural Safety ◽

Support Vector ◽

Foundation Pit ◽

Deformation Prediction

Timely and accurate prediction of structural settlement is of great significance to eliminate the hidden danger of structural and prevent structural safety accidents. Since the deformation monitoring data usually is nonstationary and nonlinear, the deformation prediction is a difficult problem in the structural monitoring research. Aiming at the problems in the structural deformation prediction model and considering the internal characteristics of deformation monitoring data and the influence of different components in the data on the prediction accuracy, a combined prediction model based on the Empirical Mode Decomposition, Support Vector Regression, and Wavelet Neural Network (EMD-SVR-WNN) is proposed. EMD model is used to decompose the structure settlement monitoring data, and the settlement data can be effectively divided into relatively stable trend terms and residual components of random fluctuation by energy matrix. According to the different characteristics of random items and trend items, WNN and SVR methods are, respectively, used for prediction, and the final settlement prediction is obtained by integrating the prediction results. The measured ground settlement data of foundation pit in subway construction is used to test the performance of the model, and the test results show that the prediction accuracy of the combined prediction model proposed in this paper reaches 99.19%, which is 77.30% higher than the traditional SVR, WNN, and DBN-SVR models. The experimental results show that the proposed prediction model is an effective model of structural settlement.

Download Full-text

Characterize Fracture Development Through Strain Rate Measurements by Distributed Acoustic Sensor DAS

10.2118/205267-ms ◽

2022 ◽

Author(s):

Jin Tang ◽

Ding Zhu

Keyword(s):

Hydraulic Fracturing ◽

Strain Rate ◽

Fiber Optic ◽

Fracture Propagation ◽

Acoustic Sensor ◽

Observation Well ◽

Complex Fracture ◽

Multistage Hydraulic Fracturing ◽

Fracture Development ◽

Fracture Growth

Abstract In multistage hydraulic fracturing treatments, the combination of extreme large-scale pumping (high rate and volume) and the high heterogeneity of the formation (because of large contact area) normally results in complex fracture growth that cannot be simply modeled with conventional fracture models. Lack of understanding of the fracturing mechanism makes it difficult to design and optimize hydraulic fracturing treatments. Many monitoring, testing and diagnosis technologies have been applied in the field to describe hydraulic fracture development. Strain rate measured by distributed acoustic sensor (DAS) is one of the tools for fracture monitoring in complex completion scenarios. DAS measures far-field strain rate that can be of assistance for fracture characterization, cross-well fracture interference identification, and well stimulation efficiency evaluation. Many field applications have shown DAS responses on observation wells or surrounding producers when a well in the vicinity is fractured. Modeling and interpreting DAS strain rate responses can help quantitatively map fracture propagation. In this work, a methodology is developed to generate the simulated strain-rate responds to assumed fracture systems. The physical domain contains a treated well that the generate strain variation in the domain because of fracturing, and an observation well that has fiber-optic sensor installed along it to measure the strain rate responses to the fracture propagation. Instead of using a complex fracture model to forward simulate fracture propagation, this work starts from a simple 2D fracture propagation model to provide hypothetical fracture geometries in a relatively reasonable and acceptable range for both single fracture case and multiple fracture case. Displacement discontinuity method (DDM) is formulated to simulate rock deformation and strain rate responds on fiber-optic sensors. At each time step, fracture propagation is first allowed, then stress, displacement and strain field are estimated as the fracture approaches to the observation well. Afterward, the strain rate is calculated as fracture growth to generate patterns as fracture approaching. Extended simulation is conducted to monitor fracture propagation and strain rate responses. The patterns of strain rate responses can be used to recognize fracture development. Examples of strain rate responses for different fracturing conditions are presented in this paper. The relationship of injection rate distribution and strain rate responses is investigated to show the potential of using DAS measurements to diagnose multistage hydraulic fracturing treatments.

Download Full-text