Upstream oil and gas industry services work to deliver success throughout the life cycle of the reservoir. However, conventional sources of oil and gas are declining; hence, operators are increasingly turning their attention to unexplored and underdeveloped regions, such as high-pressure/high-temperature (HP/HT) and deepwater areas, as well as working to increase recoveries in mature fields. As reservoirs become more complex and drilling operations become more expensive, there is a growing need to reduce inefficiencies and costs. Petroleum engineers are increasingly using optimized formation evaluation techniques, software with three-dimensional (3D) visualization, and multidisciplinary data interpretation techniques. In addition, data from new downhole equipment provides reliable, real-time information about downhole conditions. With these improved techniques and better data, operators can model, predict, and control their operations better in real-time, thereby reducing inefficiencies and cost. However, this massive integration of varied data moving in higher volumes and at increased speeds has created an increasing demand on the computation and use of actionable and predictive data-driven analytics. Furthermore, these analytics must work in real-time to quickly discover the important critical data attributes and features for use in forecasting. Attribute importance is a well-known statistical technique used to identify critical attributes and features within a set of attributes that could impact a specific target. The benefits of deriving attribute importance include improved data set comprehension, reduced analysis efforts, and reduced time and computational resources required for actionable predictive analytics. Furthermore, attribute importance aids in understanding the attribute space and interactions, and it enables dimensionality reduction; thereby, leading to comprehendible, accurate, "parsimonious" (i.e., simple) models that lead to actionable predictive analytics. Today, numerous standard and custom techniques—each having their own strengths and weaknesses—are available for performing attribute importance. Each technique applies a different function to evaluate the importance of an attribute and score it to produce a ranked subset of attributes. Hence, it is possible (and is fairly common) to arrive at different subsets of important attributes based on the choice and configurations of the various techniques. To solve this problem, a feasible and effective framework-based approach is proposed that uses multiple attribute importance techniques and then intelligently fuses the results of these methods to arrive at a fused subset of important attributes. This paper describes a framework/"ensemble" (i.e., combined) approach called Segmented Attribute Kerneling (SAK). In addition, it discusses the results obtained from applying this approach on a dataset incorporating real-time drilling surface data logs and drilling parameters. This approach runs multiple attribute importance algorithms simultaneously, finds the intersecting subset of important attributes across the multiple techniques, and then outputs a consolidated ranked set. In addition, the method identifies and presents a ranked subset of the attributes excluded from the union. This paper compares the results of this approach to a single-approach technique based on the output of the predictive models.
A new real-time model was developed, based on a deep recurrent neural network (DRNN), to predict response variables, such as surface pressure response, during the hydraulic fracturing process. During the stimulation process stage, fluids are inserted at the top of the wellhead, and the flow is driven by the difference between the hydrostatic pressure and reservoir pressure. The major physics and engineering aspects in this process are very complex; quite often, the measured data includes a large amount of uncertainty related to the accuracy of the measured data, as well as intrinsic noise. Consequently, the best approach uses a machine learning-based technique that can resolve both temporal and spatial non-linear variations. The approach followed in this paper provides a long short-term memory (LSTM) network-based method to predict surface pressure in a fracturing job, considering all commonly known surface variables. The surface pumping data consists of real-time data captured within each stage, including surface treating pressure, fluid pumping rate, and proppant rate. The prediction of a response variable, such as the surface pressure response, is important because it provides the basis for decisions made in several oil and gas applications to ensure success, including hydraulic fracturing and matrix acidizing. Currently available modeling methods are limited in that the estimates are not high resolution and cannot address a high level of non-linearity in the treatment pressure time series relationship with other variables, such as flow rate and proppant rate. In addition, these methods cannot predict subsurface variable responses based only on surface variable measurements. The method described in this paper is extended to accommodate the prediction of diverter pressure response. The model presented in this paper uses a deep learning neural network model to predict the surface pressure based on flow rate and proppant rate. This work represents the first attempt to predict (in real time) a response variable, such as surface pressure, during a pumping stage using a memory-preserving recurrent neural network (RNN) variant (for example, LSTM and gated recurrent unit (GRU)). The results show that the LSTM is capable of modeling the surface pressure in a hydraulic fracturing process well. The surface pressure predictions obtained were within 10% of the actual values. The current effort to model surface pressure can be used to simulate response variables in real time, providing engineers with an accurate representation of the conditions in the wellbore and in the reservoir. The current method can overcome the handling of complex physics to provide a reliable, stable, and accurate numerical solution throughout the pumping stages.
A new real-time machine learning model has been developed based on the deep recurrent neural network (DRNN) model for performing step-down analysis during the hydraulic fracturing process. During a stage of the stimulation process, fluids are inserted at the top of the wellhead, while the flow is primarily driven by the difference between the bottomhole pressure (BHP) and reservoir pressure. The major physics and engineering aspects involved are complex and, quite often, there is a high level of uncertainty related to the accuracy of the measured data, as well as intrinsic noise. Consequently, using a machine learning-based method that can resolve both the temporal and spatial non-linear variations has advantages over a pure engineering model. The approach followed provides a long short-term memory (LSTM) network-based methodology to predict BHP and temperature in a fracturing job, considering all commonly known surface variables. The surface pumping data consists of real-time data captured within each stage, such as surface treating pressure, fluid pumping rate, and proppant rate. The accurate prediction of a response variable, such as BHP, is important because it provides the basis for decisions made in several well treatment applications, such as hydraulic fracturing and matrix acidizing, to ensure success. Limitations of the currently available modeling methods include low resolution BHP predictions and an inability to properly capture non-linear effects in the BHP/temperature time series relationship with other variables, including surface pressure, flow rate, and proppant rate. In addition, current methods are further limited by lack of accuracy in the models for fluid properties; the response of the important sub-surface variables strongly depends on the modeled fluid properties. The novel model presented in this paper uses a deep learning neural network model to predict the BHP and temperature, based on surface pressure, flow rate, and proppant rate. This is the first attempt to predict response variables, such as BHP and temperature, in real time during a pumping stage, using a memory-preserving recurrent neural network (RNN) variant, such as LSTM. The results show that the LSTM can successfully model the BHP and temperature in a hydraulic fracturing process. The BHP and temperature predictions obtained were within 5% relative error. The current effort to model BHP can be used for step-down analysis in real time, thereby providing an accurate representation of the subsurface conditions in the wellbore and in the reservoir. The new method described in this paper avoids the need to manage the complex physics of the present methods; it provides a robust, stable, and accurate numerical solution throughout the pumping stages. The method described in this paper is extended to manage step-down analysis using surface-measured variables to predict perforation and tortuosity friction.
Reservoir simulation results currently provide the basis for important reservoir engineering decisions; grid complexity and non-linearity of these models demand high computational time and memory. The physics-based simulation process must be repeated to increase model prediction accuracy or to perform history matching; consequently, the simulation process is often time-consuming. This paper describes a new methodology based on a deep neural network (DNN) technique, the graph convolutional neural network (G-CNN). G-CNN increases the modeling prediction speed and efficiency by compressing the computational time and memory usage of the reservoir simulation. A G-CNN model was used to perform the reservoir simulations described. This new methodology combines physics-based and data-driven models in reservoir simulation. The workflow generates training datasets, enabling intelligent sampling of the reservoir production data in the G-CNN training process. Bottomhole pressure constraints were set for all simulations. The production data generated by the reservoir model, with the mesh connectivity information, is used to generate the G-CNN model. This approach can be viewed as hybrid data-driven, retaining the underlying physics of the reservoir simulator. The resulting G-CNN model can perform reservoir simulations for any computational grid and production time horizon. The method uses convolutional neural network and mesh connections in a fully differentiable scheme to compress the simulation state size and learns the reservoir dynamics on this compressed form. G-CNN analysis was performed on an Eagle Ford-type reservoir model. The transmissibility, pore volume, pressure, and saturation at an initial time state were used as input features to predict final pressure and saturations. Prediction accuracy of 95% was obtained by hyperparameter tuning off the G-CNN architecture. By compressing the simulation state size and learning the time-dependent reservoir dynamics on this compressed representation, reservoir simulations can be emulated by a graph neural network which uses significantly less computation and memory. The G-CNN model can be used on any computational grid because it preserves the structure of the physics. The G-CNN model is trained by mapping an initial time state to a future prediction time state; consequently, the model can be used for generalizing predictions in any grid sizes and time steps while maintaining accuracy. The result is a computationally and memory efficient neural network that can be iterated and queried to reproduce a reservoir simulation. This novel methodology combines field-scale physics-based reservoir modeling and DNN for reservoir simulations. The new reservoir simulation workflow, based on the G-CNN model, maps the time state predictions in a resampled grid, reducing computational time and memory. The new methodology presents a general method for compressing reservoir simulations, assisting in fast and accurate production forecasting.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.