Skip Nav Destination
Close Modal
Update search
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
Filter
- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number
NARROW
Format
Subjects
Date
Availability
1-20 of 339
Keywords: machine learning
Close
Follow your search
Access your saved searches in your account
Would you like to receive an alert when new items match your search?
Sort by
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0036
.... machine learning well logging Artificial Intelligence Reservoir Characterization neural network abdulraheem transit time wave velocity prediction correlation gamma ray log analysis Upstream Oil & Gas empirical correlation compressional resistivity wave transit time sonic wave transit...
Abstract
ABSTRACT: Geomechanics plays a vital role in reducing drilling problems. Parameters such as Young's modulus and Poisson's ratio are calculated from acoustic logs (sonic wave transit time). Occasionally, these logs are not recorded because of many factors such as time saving or cost cutting. In such cases, empirical correlations are used to calculate the sonic transit times. But none of these empirical correlations are universally acceptable. As a result, inaccurate values can potentially raise major concerns throughout the life of the well. The objective of this paper is to develop a robust empirical model for compressional and shear wave transit times using artificial intelligence techniques for unconventional reservoirs. For this purpose, well logs data was used from a tight sandstone formation to predict the transit times. Artificial neural networks (ANN) was used in this study. The ANN models predicted the sonic waves with very high accuracy with a correlation coefficient (CC) up to 0.96 and an average absolute percentage errors (AAPE) as low as 2%. The shear wave transit time prediction from the new model was also validated using available sandstone empirical equations.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0023
...; fluid type; hydraulic horsepower (HHP) per stage; lb/ft 2 of proppants per stage; number of stages, and lateral length (completed interval) of horizontal wells,. fracturing materials hydraulic fracturing machine learning Completion Installation and Operations Artificial Intelligence...
Abstract
ABSTRACT: This work offers a predictive model as a success-of-completion strategy (treatment) for a major successful shale play (Wolfcamp). The predictive model may be used to evaluate spacing between fracture clusters and the number of clusters and perforations, and to guide future selective optimum completion for the shale play. Many important parameters that control behavior of producing wells have been analyzed, including number of days on production; depth; fluids in bbl; horizontal well completion configurations; stages per well; fracture type; average water requirement; proppant type; fluid type; hydraulic horsepower (HHP) per stage; lb/ft 2 of proppants per stage; number of stages, and lateral length (completed interval) of horizontal wells,.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0106
... density inversion MEQ machine learning hydraulic fracturing stimulation fluid pressure seismicity probability Artificial Intelligence Fluid Dynamics Earthquake genetic algorithm riffault Dempsey cloud 1 1. INTRODUCTION Hydraulic stimulation is applied to wells to improve formation...
Abstract
ABSTRACT: It is commonly assumed that the dimensions of an induced microseismic cloud around an Enhanced Geothermal System (EGS) well delineate the “stimulated volume” in which permeability has been increased. This interpretation derives from a self-propping fracture model, wherein shear slip results in some permanent increase of hydraulic aperture. Or, more simply, earthquakes equal permeability. In contrast, here, we interpret microearthquakes (MEQs) directly in terms of the stress and pressure conditions that control earthquake triggering. In many cases, triggering is dominated by fluid pressure changes within a fracture, although increasingly poroelastic and earthquake interaction effects are seen as important.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0092
... failure criteria. A machine learning algorithm is developed to generate cement parameters, such as cohesion and friction angle. The statistic value, Misfits, is used for the quantitative comparison of the accuracy of different failure criteria. Results show that using only the conventional triaxial...
Abstract
ABSTRACT: The accurate prediction of cement failure is significant to maintain wellbore integrity. In the downhole situations, the failure criteria widely used for rock are not guaranteed to be suitable for cement. Till now, seldom researchers have paid attention to the right choice of failure criterion for cement and most of them simply used the Mohr-Coulomb criterion. In this work, comprehensive research is performed to collect published conventional and true triaxial compression strength data of cement. All the accessible data are used to evaluate the accuracy of six different failure criteria. A machine learning algorithm is developed to generate cement parameters, such as cohesion and friction angle. The statistic value, Misfits, is used for the quantitative comparison of the accuracy of different failure criteria. Results show that using only the conventional triaxial strength data, the influence of intermediate stress cannot be reflected and Mohr-Coulomb could provide similar accuracy compared to the three stress dependent criteria. However, when the true triaxial compression strength data is applied, the Mohr-Coulomb could generate significant errors. Three stress dependent criteria should be used for predicting cement failure. This study fills the gap of finding the right failure criterion for cement and provides guidance for wellbore integrity.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0034
... discussed for each model showing the fine differences between the models. Artificial Intelligence exponent hydraulic conductivity 2-parameter model machine learning Fluid Dynamics pressure parameter sorptivity test Upstream Oil & Gas correspond sorptivity capillary pressure curve...
Abstract
ABSTRACT: The capillary pressure curve is a property of rocks whose determination requires elaborate petrophysical measurements. The utility of the capillary pressure curve can be extended to the evaluation of the fluid flow properties of rocks which provides another method for characterization of rocks. In this work we study 1-, 2- and 3-parameter capillary curves models through their effect on the imbibition curves. The 3-parameter model is the usual Van Genuchten family of capillary curves. Each model produces imbibition curves with specific characteristics, encapsulating the behavior of the rock during absorption with increasing detail. Hence, the characteristics of the imbibition curves are associated with suitable regimes of the parameters of each model, thereby providing enough classifying information to determine a suitable capillary curve from imbibition data through back analysis. In order to prove the effectiveness of the methodology we have performed an imbibition test on sandstone and compared it against a series of produced results from finite element analysis for all types of capillary curve models. The degree of coincidence with the experimental imbibition curve is discussed for each model showing the fine differences between the models.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0101
... that many of these equations were developed for only one type of sandstone and tend to generalize poorly to the broader database. machine learning Artificial Intelligence wellbore integrity reservoir geomechanics strength zorlu Upstream Oil & Gas Reservoir Characterization Wellbore...
Abstract
ABSTRACT: The mechanical properties of rock, such as unconfined compressive strength, are important to many engineering problems such as slope stability analysis, underground excavation, and drilling mechanics. Many previous studies have proposed relationships that can be used to estimate rock strength properties based on petrographic indices that are easier to measure. In this study, a comprehensive literature review was performed to build a database containing petrographic and mechanical properties for sedimentary rock types. A subset of this database containing only sandstone rock units was used for statistical analyses to develop predictive models. Linear and nonlinear bivariate models were constructed to relate each petrographic and physico-mechanical index variable to each mechanical property. Additionally, multiple regression models were developed using a subset of the petrographic and predictor variables. Contrary to prior studies using smaller data sets, very few correlations were identified, and those that were (for example dry density and mean grain size as predictors of UCS) tended to be for cases where limited data were available for the variables in question. A separate literature review was then performed to find previously-studied relationships that predict the strength properties of sandstones based on petrographic properties. 59 equations for sandstones were found and were tested against the broader sandstones database compiled in this study. It was found that many of these equations were developed for only one type of sandstone and tend to generalize poorly to the broader database.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0130
... orientation stimulation microseismic data microseismicity DPA Reservoir Characterization baig geomechanical model dynamic parameter analysis dynamic parameter deformation machine learning Upstream Oil & Gas Simulation prediction evolution fracture geometry hydraulic fracturing...
Abstract
ABSTRACT: In this paper, we integrate hydraulic fracture and geomechanical modelling with microseismic data to quantify deformation and stress field evolution as a predictive tool, as well as to provide a calibration and microseismic workflow. Our approach utilizes simulated fractures as well as moment tensor solutions to represent localized discrete rupture zones. The geomechanical algorithm evaluates the Green's functions for each rupture, and subsequently calculates the co- and post-seismic deformation using linear superposition. These data are then combined with dynamic parameter analysis to define an integrated view of the reservoir. We apply the approach on a data set from a hydraulic fracture stimulation in the Marcellus and show that integrating fracture models with geomechanical simulation can forecast zones of microseismic activity, identified by regions of high deformation and high Coulomb stresses. We compare dynamic parameter analysis from the observed microseismic data with geomechanical predictions from elastic and visco-elastic models and note the behavioral sensitivity of dynamic parameters to the geomechanical Earth model. The visco-elastic models better represent the dynamic or collective behavior of the microseismicity while maintaining reasonable model rigidity to produce consistent stress drops. Furthermore, the elastic models are sufficiently accurate to forecast potential volumes and timing of microseismicity.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0243
... better than existing correlations using different technics in optimization and agreement study. machine learning Reservoir Characterization reservoir geomechanics neural network correlation Artificial Intelligence neural network model lab test test data normal distribution agreement...
Abstract
ABSTRACT: Elastic properties of rocks including Young's modulus and Poisson's ratio, in specific, are the main parameters, which are needed for design of hydraulic fracturing geometry, estimation of stimulated hydraulic fracture volume, sanding prediction and design of gravel pack and many other applications. Conventionally, the elastic properties are estimated from the compression and shear sonic and density logs, due to the intrinsic relationship of the sonic velocity and density with rock stiffness, then, calibrated against the core data taken at some depths. However, the use of any correlation is subjected to several shortcomings, including the fact that they cannot be generalized and in many cases quality cores are not available to conduct the calibration. In this study, we used the Artificial neural network (ANN) to estimate the elastic properties of the Bakken Formation. A total of 240 core samples from eight wells drilled into Upper, Middle and Lower Bakken were used. Bulk density, compressional and shear velocities, and porosity were the main parameters used for training purposes. The results indicated that an optimized ANN model is capable of predicting the elastic properties better than existing correlations using different technics in optimization and agreement study.
Proceedings Papers
L. Gazzola, M. Ferronato, M. Frigo, C. Janna, P. Teatini, C. Zoccarato, M. Antonelli, A. Corradi, M. C. Dacome, S. Mantica
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0208
... improve the effectiveness and reliability of geomechanical reservoir models. Artificial Intelligence Ferronato reservoir geomechanics displacement application Reservoir Characterization teatini covariance matrix machine learning Upstream Oil & Gas hydrocarbon reservoir probability...
Abstract
ABSTRACT: The use of numerical models in geomechanics implicitly assumes a number of approximations and uncertainties, even though they are usually regarded as deterministic tools. Simplifications in the constitutive law, uncertainties in geomechanical parameters values, imposition of boundary conditions are only few examples of the probabilistic factors that affect the modelling process of natural phenomena. Integration of Data Assimilation (DA) techniques in the modelling processing chain can improve the outcome accuracy and reliability by incorporating the available observation data. In this work, three different DA techniques have been integrated into a geomechanical reservoir model with the aim at improving land subsidence prediction over producing hydrocarbon fields. A synthetic test case has been analyzed demonstrating that the proposed approach could be a promising tool to improve the effectiveness and reliability of geomechanical reservoir models.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0325
.... machine learning classification abutment angle knowledge management US government Artificial Intelligence calculation new abutment angle equation colwell stress measurement database fracability metals & mining proceedings pillar stability abutment angle equation Case History abutment...
Abstract
ABSTRACT: The National Institute for Occupational Safety and Health (NIOSH) first developed the Analysis of Retreat Mining Pillar Stability (ARMPS) program to help the U.S. coal mining industry to design retreat room and pillar panels. Similar to other pillar design methodologies, ARMPS determines the adequacy of the design by comparing the estimated in situ and mining induced loads to the load bearing capacity of the pillars. ARMPS calculates magnitude of the in situ and mining induced loads by using geometrical computations and empirical rules. The program uses the “abutment angle” concept in calculating the magnitude of the mining induced loads on pillars adjacent to a gob. The value of the abutment angle for coal mines in the United States was derived by back analysis of field measurements, and ARMPS2010 engineering design criterion was derived from the statistical analysis of the databases of more than 640 retreat mining case histories from various U.S. coal mines. In this study, stress measurements from U.S. and Australian coal mines were back analyzed using the square decay stress distribution method, and the abutment angles are investigated. The results of the analyses indicated that for shallow mines with overburden depths of less than 200 m, empirical derivation of 21° abutment angle used in ARMPS2010 was supported by the case histories. However, at depths greater than 200 m, the abutment angle was found to be significantly less than 21°. A new equation employing the overburden depth to panel width ratio was constructed for the calculation of abutment angle for deep cover cases. Finally, the new abutment angle equation was tested using 336 deep cover cases from the ARMPS2010 database. The new abutment angle equation was found to perform a good classification compared to using 21°. It was also apparent that, for deep cover cases (deeper than 200 m), the barrier pillar stability factors were the governing parameters in classification of failed cases and the results can be considered an indicator for the importance of barrier pillars in deep cover retreat mines.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0268
... analysis, a description method for the mechanical properties of carbonates with fractures and holes was established. The analysis results can be used to rock mechanics calculation of deep fractured carbonate rocks. Artificial Intelligence carbonate reservoir machine learning complex reservoir...
Abstract
ABSTRACT: Deep carbonate rocks generally belong to dense sedimentary rocks, but the cracks and holes of different scales in the formation will affect the mechanical properties. In the borehole stability analysis of drilling and completion engineering of deep carbonate reservoir, it is necessary to determine the formation mechanical properties through laboratory experiments and use them as basic parameters for calculation. However, limited core experiments are difficult to fully reflect the formation characteristics, and affect the accuracy of the results. Therefore, based on experimental data, Hoek-Brown criterion and well logging data, a comprehensive classification method was established to describe the characteristics of deep carbonate rocks. At first, the cores of a fractured carbonate reservoir with the vertical depth of 7350-7370m were used to carry out indoor unconfined and confined compressive strength experiments, and the core mechanical parameters such as strength, elastic modulus and Poisson's ratio were obtained. Then, based on the experimental results and logging data, the relevant parameters of Hoek-Brown criterion in different section were obtained. Based on the above analysis, a description method for the mechanical properties of carbonates with fractures and holes was established. The analysis results can be used to rock mechanics calculation of deep fractured carbonate rocks.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0362
... strong tendency to open the natural fracture. The scenarios predicted by the neural networks is in agreement with the rock mechanics concepts and with the expected tendencies, which gives more reliability to the neural network. neural network Simulation machine learning Upstream Oil & Gas...
Abstract
ABSTRACT: The geometry of hydraulic fractures is a major concern, especially in highly fractured reservoirs, once it can severely affect oil productivity. Furthermore, interaction of hydraulic fractures with natural fractures can affect locally the in-situ stress state that may cause slippage of the natural fracture. In some cases, stress shadowing effects can even close the hydraulic fracture. Stress rotation around the natural fracture can make hydraulic fracture geometry complex, demanding more computational capacity and time to run numerical simulations. The neural network is a tool that can help real time daily operations when a quick decision is necessary and there is no time to run numerous simulations. This paper focuses on the development of an artificial neural network to predict the interaction between natural and hydraulic fractures. The artificial neural network predicts if hydraulic fractures cross or open natural fractures. In-situ stresses, fracture energy, friction angle of the natural fracture, and angle of approach between fractures are the required input for the artificial neural network, while mode of interaction such as opening, or crossing is the output. The database to train the neural network includes experimental data available in literature and results from numerical simulations. A considerable number of numerical simulations using interface elements to represent fractures provide a suitable database to build an accurate neural network. The neural network was trained for predicting fracture interaction with more than 90% accuracy for opening or crossing behavior. Results were compared with experimental data available in literature and numerical simulations. The results show that for higher angles of approach and higher stress differential, there is a strong tendency to fracture crossing. This is in agreement with the experimental conclusions presented by previous publications in the literature. Additionally, intermediary stress differentials, low values of cohesion and friction angle, and approach angle below 60° define conditions with a strong tendency to open the natural fracture. The scenarios predicted by the neural networks is in agreement with the rock mechanics concepts and with the expected tendencies, which gives more reliability to the neural network.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0382
... adequately dealt with to still achieve the economic benefits. Reservoir Characterization drilling operation wellbore integrity annular pressure drilling loss event machine learning Artificial Intelligence Upstream Oil & Gas mechanism ECD monitoring failure mechanism well control...
Abstract
ABSTRACT: Batch drilling is common practice these days in order to efficiently execute and complete wells within the budgeted time and cost. While the approach may offer substantial economic benefits, in the event of any subsurface challenges, a steep and rapid learning curve needs to be adequately dealt with to still achieve the economic benefits.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0455
... simplified analytical model is presented to simulate the natural fractures' stress-sensitive effects on production. machine learning complex reservoir flow in porous media Fluid Dynamics Simulation modeling natural fracture hydraulic fracturing element method calculation Artificial...
Abstract
ABSTRACT: Defining the SRV is a complex task involving inference and interpreting indirect and multiple sources of information. Although the newly created fracture network, enhanced permeability zone, and induced stresses and stress shadow fields are explicitly related to the SRV creation, the key factors controlling unconventional reservoir production remain to be defined quantitatively. u stress ratio, natural fracture reservoir network properties, interconnectivity stimulation mechanisms and fracture propagation are discussed and quantified as the important factors for production enhancement.For a given SVR the fracture density, aperture, and flow characteristics of each fracture can be different for different u conditions and thus different production enhancements may be expected. Herein a DDM method simulating the interaction between HF and natural fractures is discussed and a simplified analytical model is presented to simulate the natural fractures' stress-sensitive effects on production.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0438
... ABSTRACT: Parameter calibration is an indirect problem in which responses of a system are known, but the properties of the said system are not. Often, trial and error strategies are used for parameter calibration. However, they can be labor intensive and time consuming. Machine learning...
Abstract
ABSTRACT: Parameter calibration is an indirect problem in which responses of a system are known, but the properties of the said system are not. Often, trial and error strategies are used for parameter calibration. However, they can be labor intensive and time consuming. Machine learning methods have been employed to automatize the process. The present work proposes a combination of Genetic Programming (GP) with an optimization method to estimate the elastic parameters of a finite element model of an oil and gas field. The estimated parameters were the Young's modulus and the Poisson's ratio of each rock layer and the target responses were the field measurements of horizontal and vertical total stress profiles and the maximum surface subsidence. GP provided functions relating each response to the elastic properties. An optimization procedure using the interior-point algorithm was applied in order to estimate the values for the properties that would lead to the target outputs. Results show that the calibration procedure provided properties that, in most cases, differed in less than 10% from the expected values. Thus, the proposed method has potential to be a relatively straightforward method to estimate parameters of a geomechanical model.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-0433
... control production logging production monitoring hydraulic fracturing Upstream Oil & Gas viscosity sensor machine learning Artificial Intelligence deep learning subsurface virtual fiberoptic sensor fracture height hydraulic fracture physics-informed deep neural network analysis dataset...
Abstract
ABSTRACT: Fiber optic based distributed acoustic sensors (DAS) provide a new approach for monitoring signals of interest in the subsurface with unprecedented spatiotemporal resolution. These sensors produce measurements that are fundamentally different from their traditional counterparts, such as geophones, and produce significantly larger volumes of data. To interpret these data, we begin by using the physics-based thermal-hydraulic-mechanical (THM) model in the GEOS code to simulate synthetic DAS measurements for a range of subsurface conditions (fracture propagation, fault slip, etc.) and sensor configurations (e.g.: horizontal or vertical well deployments). These synthetic DAS measurements are then algorithmically labeled based upon features of interest within their parent models, such as the extent of any generated hydraulic fractures or the distribution of proppant particles, and are compiled into a database. The synthetic database can be used to train and test an initial deep neural network (DNN) representation of the subsurface, which can then be optimized by incorporating any available field measurements through transfer learning. This hybrid, physics informed DNN model is capable of interpreting DAS measurements in near-real time, making it a useful tool for decision making by field engineers, and works under both data-rich and data-poor conditions. To demonstrate this approach, we consider the problem of imaging hydraulic fracture propagation in an unconventional oil and gas reservoir. Our results indicate that we are can use a trained DNN model to accurately estimate the extents of a hydraulic fracture using the location information and DAS measurements for a single fiber-optic sensor as inputs.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-1570
... data machine learning Upstream Oil & Gas seismic hazard assessment comprehensive assessment Velenje fractal dimension-based index information dimension dimension-based index application fractal dimension 1. INTRODUCTION Seismic hazard in mining is the probability of hazard events...
Abstract
ABSTRACT: Seismic hazards have become one of the common risks in underground coal mining and their assessment is an important component of the safety management. In this study, a methodology, involving nine fractal dimension-based indices and a fuzzy comprehensive evaluation model, has been developed based on the processed real time microseismic data from an underground coal mine, which allows for a better and quantitative evaluation of the likelihood for the seismic hazards. In the fuzzy model, the membership function was built using a Gaussian shape and the weight of each index was determined using the performance metric F score derived from the confusion matrix. The assessment results were initially characterised as a probability belonging to each of four risk levels (none, weak, moderate and strong). The comprehensive result was then evaluated by integrating the maximum membership degree principle (MMDP) and the variable fuzzy pattern recognition (VFPR). The model parameters of this methodology were first calibrated using historical microseismic data over a period of seven months at Coal Mine Velenje in Slovenia, and then applied to analyse more recent microseismic monitoring data. The results indicate that the calibrated model was able to assess seismic hazards in the mine.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-1791
... class machine learning classification accuracy ratio Upstream Oil & Gas accuracy variation identification characteristic input parameter topology classification accuracy characterization limestone lithological classification p-wave velocity Visualization dataset information self...
Abstract
ABSTRACT: Lithological classification of geological formations usually depends on the geologist’ expertise through a laborious and time consuming process. Automating the rock classification procedure is a step towards fast and accurate characterization of geological formations where large data is available. In this study, we assessed the use of a clustering algorithm, namely self-organizing maps, for classification of different lithological classes of limestone. The input parameters are geo-mechanical and geological properties of limestones such as density, uniaxial compressive strength, p-wave velocity and geological origin. The results indicate that the self-organizing maps have a promising potential in lithological clustering. The algorithm accurately clusters limestone classes that have consistent high or low values for at least one of the input parameters. However, the accuracy decreases if the limestone class have inherent high variance. The results provided in this paper provides a preliminary confirmation of the applicability of the self-organizing maps. Future work will focus on the replication of the results with a larger dataset.
Proceedings Papers
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-1863
... and cavities in the near-well formation can be analyzed through drilling data and fracturing data. machine learning reservoir reconstruction coefficient formation fracture Upstream Oil & Gas drilling data instability neuron Artificial Intelligence neural network displacement...
Abstract
ABSTRACT: The fractures in the carbonate reservoirs have a high degree of development, and the natural fractures are susceptible to unstable cracking due to construction such as drilling and reservoir reconstruction. Based on data feedback during drilling and fracturing construction, combined with the advantages of neural network algorithm in data fault tolerance and prediction, and the ability to extrapolate more data based on existing data, the basic principles of neural network are established. The multi-layer back-propagating neural network uses the existing drilling and fracturing data as training camp samples to train the neural network, and then uses the trained network to predict the samples. Through the analysis of neural network, the integration of drilling and fracturing data and the instability of formation fractures are correlated, so that the cracks of natural fractures and cavities in the near-well formation can be analyzed through drilling data and fracturing data.
Proceedings Papers
G. A. P. Batista, J. H. Da Silva, J. E. F. Ramires, A. S. Carvalho, M. P. Campos, A. Gontijo, A.R. Matos
Publisher: American Rock Mechanics Association
Paper presented at the 53rd U.S. Rock Mechanics/Geomechanics Symposium, June 23–26, 2019
Paper Number: ARMA-2019-1917
... propagation anglogold ashanti minimum ppv serra grande accident vibration machine learning Upstream Oil & Gas cme excavation peak particle velocity maximum load unstable block wave attenuation 1. INTRODUCTION The vibrations caused by the drill and blats process are a matter of concern due...
Abstract
ABSTRACT: The rockfalls are recognized as the main cause of accidents at underground mining activities and are denominated as Fall of Ground. Aiming to reach a high level of risk management, real rockfalls examples will be expose as the developed methodology for risk prevention at the Mineração Serra Grande, located in the city of Crixás, in the state of Goiás, in central Brazil, and operated by AngloGold Ashanti. The result has shown the possibility to identify instability potential targets and by applying abacuses and modeling, is possible to manage these locals aiming to minimize the risks caused by rockfalls. The attenuation rule model proposed for this mine presents a good equivalence of 93% with the correlation coefficient. Defining the minimum vibration level for forming instable blocks is possible to determine the blast influence, turning possible to manage the project from an efficient and safe perspective. Therefore, this article aims to contribute for mineral operations management on reducing rockfalls hazards.