|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
The objective of this paper is to develop predictive models to optimize the (1) characterization of the stimulated reservoir volume (SRV), (2) discretization of the fracture network, and (3) hydraulic fracturing modeling, by combining machine learning (ML) algorithms and reservoir engineering in low permeability reservoirs.
An unsupervised learning algorithm is implemented to characterize the fracture network developed by micro-seismic observations during hydraulic fracturing. A Self Organizing Map (SOM) and Multi-Attribute Analysis are performed on the available seismic data to map the extension of the hydraulic fracturing stages and the fracture network complexity in a low permeability reservoir.
To correlate the mapped fracture network and discretized SRV, a 3D Finite Element Model (FEM) is developed to estimate fracture behavior, stress response, and hydraulic fracture propagation, on the predicted and forecasted multi-attribute map of the reservoir.
A 3D hydraulic fracture propagation model (HFPM) is introduced, to delimit the fracture geometry and remove data outliers in the SOM algorithm. Unsupervised algorithms rely on data quality. The efficiency of hydraulic fracturing modeling is improved with a machine learning approach by refining the certainty and quality of the data. An Artificial Neural Network (ANN) model helps to select the most significant parameters related to fracture modeling and simulation in the field. This approach allows us to recreate and forecast complex fracture networks in low permeability reservoirs, based on the learned geostatistical maps and hydraulic fracturing parameters, particularly where the microseismicity is limited or unavailable.
To validate the implementation of the 3D-HFPM in the field, an earthquake model is compared with statistically significant microseismic events obtained by the unsupervised iso-cluster algorithm. The relationship showed a good agreement, which suggests the HFPM agrees with seismic observations in the field.
The machine learning application to fracture network modeling provides the capability to identify susceptible areas to well interference and possible frac hits with higher certainty. This is so because the approach improves the selection of seismic data and hydraulic fracturing parameters, employed to develop the complex fracture network in numerical commercial reservoir simulators. This helps to determinate the reservoir interconnectivity and flow patterns in the fracture network.
This approach presents a robust manner for characterizing the SRV using a relative fast methodology, based on the combination of geostatistical and unsupervised learning modeling. The seismicity and hydraulic fracturing are connected using a multi-attribute and multi-disciplinary interpretation. It is a powerful tool for characterizing problematic fracture networks in unconventional reservoirs.
Summary A hybrid-hydraulic-fracture (HHF) model composed of (1) complex discrete fracture networks (DFNs) and (2) planar fractures is proposed for modeling the stimulated reservoir volume (SRV). Modeling the SRV is complex and requires a synergetic approach between geophysics, petrophysics, and reservoir engineering. The objective of this paper is to characterize and evaluate the SRV in nine horizontal multilaterals covering the Muskwa, Otter Park, and Evie Formations in the Horn River Shale in Canada, with a view to match their production histories and to evaluate the effectiveness and potential problems of the multistage hydraulic-fracturing jobs performed in the nine laterals. To accomplish this goal, the HHF model is run in a numerical-simulation model to evaluate the SRV performance in planar and complex fracture networks using good-quality microseismicity data collected during 75 stages of hydraulic fracturing (out of 145 stages performed in nine laterals). The fracture-network geometry for each hydraulic-fracture (HF) stage is developed on the basis of microseismicity observations and the limits obtained in the fracture-propagation modeling. Post-fracturing production is appraised with ratetransient analysis (RTA) for determining effective permeability under flowing conditions. Results are compared with the HHF simulation and the hydraulic-fracturing design. The HHF modeling of the SRV leads to a good match of the post-fracturing production history. The HHF simulation indicates interference between stages. The vertical connectivity in the reservoir is larger than the horizontal connectivity. This is interpreted to be the result of the large height achieved by HFs, and the absence of barriers between the formations. It is concluded that the HHF model is a valuable tool for evaluating hydraulic-fracturing jobs and the SRV in shales of the Horn River Basin in Canada. Because of the generality of the Horn River application, the same approach might have application in other shale gas reservoirs around the world.
This observation leads to the objective of this paper: to examine geoscience and engineering data of tight and shale reservoirs in Mexico with a view to estimating the oil and gas endowment, and to determine the economics of developing these plays under current and forecast possible oil and gas prices. Plays considered in this study include the Burgos, Sabinas, Tampico, Tuxpan (Platform), Veracruz, and Chihuahua Basins. Endowment is defined by the US Geological Survey (USGS) (USGS 2000) as the sum of known volumes of oil and gas (cumulative production plus remaining reserves) and undiscovered volumes. The economics of these plays is examined with the use of cumulative long-run supply (or availability) curves. These are presented as crossplots of production costs per barrel of oil or per Mcf of gas vs. endowments for the aggregate of basins, and are very useful to demonstrate how endowment volumes vary at different price levels. It is concluded that the potential of unconventional resources in Mexico is quite significant and will help to change the slope of production rates in the country from negative to positive. As a result, it is anticipated that Mexico will become an important part of the shale-petroleum revolution started in the US.
The objective of this paper is to improve the evaluation and characterization of the fracture network as well as the production matching in the Horn River Shale of Canada. The task is carried out by extending the hybrid hydraulic fracture (HHF) model introduced by
In this paper, the fracture network is discretized using microseismic observations, when available. However, microseismic data may be limited in some of the fractured stages, or like in the case of most hydraulically fractured wells it might be non-existent. The fully coupled HHF model is developed to (1) improve the shale characterization and the simulation history matching, (2) study the fracture closure and permeability change in the fracture network due to gas production, and (3) alleviate microseismic data scarcity by generating a representative fracture network of those stages where microseismic data are unavailable.
The stress change from the initial hydraulic fracturing is evaluated in nine paths multi-level horizontal wells that penetrated the Horn River Shale. The stress shadow is corroborated with microseismic observations and exhibited areas with high fracture density and productivity.
The HHF model further evaluates the reservoir response to pore pressure depletion stemming from production, which leads to stress and permeability changes, fracture closure, and fracture reorientation. The procedure improves the simulation history matching by improving reservoir characterization, especially in stages closer to the toe where an understanding of fracture network geometry is problematic due to the cloud dispersion and scarcity of the microseismicity. The model also evaluates interference between well-paths and helps to determinate the optimum well, fracture and stage spacing.
The HHF model was used to observe changes in volume, permeability and fracture connectivity in undepleted areas close to the fracture network. These areas reveal possible candidates for refracturing. A refracturing scenario that restores fracture conductivity and increases the drainage area of the fracture network is analyzed economically for evaluating the viability of that type of operation in the Horn River Shale.
The HHF simulation model improves the shale reservoir understanding and simplifies the use of a highly complex fracture network for evaluating history matching, fracture closure and permeability changes during gas production. Furthermore, it provides a viable methodology to optimize well and stage spacing, and to evaluate potential refracturing candidates, where microseismic data is unavailable and a fracture network needs to be developed.