Geomechanical Properties of Unconventional Shale Reservoirs

Production from unconventional reservoirs has gained an increased attention among operators in North America during past years and is believed to secure the energy demand for next decades. Economic production from unconventional reservoirs is mainly attributed to realizing the complexities and key fundamentals of reservoir formation properties. Geomechanical well logs (including well logs such as total minimum horizontal stress, Poisson’s ratio, and Young, shear, and bulk modulus) are secured source to obtain these substantial shale rock properties. However, running these geomechanical well logs for the entire asset is not a common practice that is associated with the cost of obtaining these well logs. In this study, synthetic geomechanical well logs for a Marcellus shale asset located in southern Pennsylvania are generated using data-driven modeling. Full-field geomechanical distributions (map and volumes) of this asset for five geomechanical properties are also created using general geostatistical methods coupled with data-driven modeling. The results showed that synthetic geomechanical well logs and real field logs fall into each other when the input dataset has not seen the real field well logs. Geomechanical distributions of the Marcellus shale improved significantly when full-field data is incorporated in the geostatistical calculations.

Download Full-text

Evaluating proxies for the drivers of natural gas productivity using machine learning models.

Interpretation ◽

10.1190/int-2020-0200.1 ◽

2021 ◽

pp. 1-81

Author(s):

William Harbert ◽

Richard Hammack ◽

Erich Zorn ◽

Alexander Bear ◽

Timothy Carr ◽

...

Keyword(s):

Machine Learning ◽

Natural Gas ◽

Marcellus Shale ◽

Gas Production ◽

Unconventional Reservoirs ◽

Data Driven ◽

Machine Learning Techniques ◽

Well Completion ◽

Stimulation Parameters ◽

Learning Techniques

The extensive development of unconventional reservoirs using horizontal drilling and multistage hydraulic fracturing has generated large volumes of reservoir characterization and production data. The analysis of this abundant data using statistical methods and advanced machine learning techniques can provide data-driven insights into well performance. Most predictive modelling studies have focused on the impact that different well completion and stimulation strategies have on well production but have not fully exploited the available in-situ rock property data to determine its role in reservoir productivity. In this study, we utilized machine learning techniques to rank rock mechanical properties, microseismic attributes, and stimulation parameters in the order of their significance for predicting natural gas production from an unconventional reservoir. The data for this study came from a hydraulically fractured well in the Marcellus Shale in Monongalia County, West Virginia. The data classes included measurements aggregated by well completion stage that included: 1) gas production; 2) well-log-derived measurements including bulk density, elastic moduli, shear impedance, compressional impedance, brittleness, and gamma measurements; 3) microseismic attributes; 4) Long Period Long Duration (LPLD) event counts; 5) fracture counts; and 6) stimulation parameters that included fluid injection volume and average pumping pressure. In this study to identify observable proxies for the drivers of gas production we evaluated five commonly used machine learning approaches including Multivariate adaptive regression spline (MARS), Gaussian mixture model (GMM), Random forest (RF), Gradient boosting (GB), and Neural network (NN). We selected five variables including LPLD event count, seismogenic b-value, hydraulic diffusivity, cumulative moment, and fluid volume as the features most likely to impact gas productivity at the stage level in the study area. The data-driven selection of these parameters for their importance in determining gas production can help reservoir engineers design more effective hydraulic fracture treatments in the Marcellus Shale and other similar unconventional reservoirs.

Download Full-text

Petroelastic and geomechanical classification of lithologic facies in the Marcellus Shale

Interpretation ◽

10.1190/int-2014-0047.1 ◽

2015 ◽

Vol 3 (1) ◽

pp. SA51-SA63 ◽

Cited By ~ 12

Author(s):

Dario Grana ◽

Kristen Schlanser ◽

Erin Campbell-Stone

Keyword(s):

Marcellus Shale ◽

Expectation Maximization Algorithm ◽

Rock Physics ◽

Rock Properties ◽

Geomechanical Properties ◽

Facies Classification ◽

Well Location ◽

Input Variables ◽

Shale Oil Reservoirs ◽

The Impact

Log-facies classification at the well location allows determination of the number of facies, the facies definition, and the correlation between facies and rock properties along the well profile. In unconventional reservoirs, because of the necessity for hydraulic fracturing in shale gas and shale oil reservoirs, facies classification should account for petroelastic and geomechanical properties. We developed a facies classification methodology based on the expectation-maximization algorithm, a statistical method that allows finding the most likely facies classification and the associated probability distribution, given the set of geophysical measurements in the borehole. We applied the proposed workflow to a complete set of well logs from the Marcellus shale and developed the corresponding facies classification from log properties measured and computed in three different domains: petrophysics, rock physics, and geomechanics. In thne preliminary well-log and rock-physics analysis, we identify three main lithofacies: limestone, shale, and sandstone. The application of the classification method provided the vertical sequence of the three lithofacies and their pointwise probability of occurrence. A sensitivity analysis was finally evaluated to investigate the impact of the number of input variables on the classification and the effects of cementation and kerogen.

Download Full-text

Benchmark Datasets, Data Driven Modeling, and Predicting Full Field Quantities of Interest for Soft Heterogeneous Materials

10.26226/morressier.612f6735bc9810372410078c ◽

2021 ◽

Author(s):

Emma Lejeune

Keyword(s):

Heterogeneous Materials ◽

Data Driven ◽

Full Field ◽

Benchmark Datasets ◽

Data Driven Modeling ◽

Quantities Of Interest

Download Full-text

Geomechanical Properties Estimation Utilizing Artificial Intelligence Prediction Tool

10.2118/204672-ms ◽

2021 ◽

Author(s):

Mohammed Alabbad ◽

Mohammad Alqam ◽

Hussain Aljeshi

Keyword(s):

Mechanical Properties ◽

Oil And Gas ◽

Oil And Gas Industry ◽

Significant Loss ◽

Wellbore Stability ◽

Rock Properties ◽

Ann Model ◽

Full Field ◽

Geomechanical Properties ◽

Rock Mechanical Properties

Abstract Drilling and fracturing are considered to be one of the major costs in the oil and gas industry. Cost may reach tens of millions of dollars and improper design may lead to significant loss of money and time. Reliable fracturing and drilling designs are governed with decent and representative rock mechanical properties. Such properties are measured mainly by analyzing multiple previously cored wells in the same formation. The nature of the conducted tests on the collected plugs are destructive and samples cannot be restored after performing the rock mechanical testing. This may disable further evaluation on the same plugs. This study aims to build an artificial neural network (ANN) model that is capable of predicting the main rock mechanical properties, such as Poisson's ratio and compressive strength from already available lab and field measurements. The log data will be combined together with preliminary lab rock properties to build a smart model capable of predicting advance rock mechanical properties. Hence, the model will provide initial rock mechanical properties that are estimated almost immediately and without undergoing costly and timely rock mechanical laboratory tests. The study will also give an advantage to performing preliminary estimates of such parameters without the need for destructive mechanical core testing. The ultimate goal is to draw a full field geomechanical mapping with this tool rather than having localized scattered data. The AI tool will be trained utilizing representative sets of rock mechanical data with multiple feed-forward backpropagation learning techniques. The study will help in localizing future well location and optimizing multi-stage fracturing designs. These produced data are needed for upstream applications such as wellbore stability, sanding tendency, hydraulic fracturing, and horizontal/multi-lateral drilling.

Download Full-text