An Integrated Approach of Numerical Well Test for Well Intersecting Fractures Based on FMI Image

Fluid flow is strongly affected by fractures in unconventional reservoirs. It is essential to deeply understand the flow characteristics with fractures for improving the production and efficiency of unconventional reservoir exploitation. The purpose of this work is to develop an accurate numerical model to evaluate the transient-pressure response for well intersecting fractures. The meshes generated from Fullbore Formation Micro-Imager (FMI) images ensure an efficient numerical description of the geometries for fractures and interlayers. The numerical simulation is implemented by an inhouse finite element method-based code and benchmarked with drill stem test (DST) data. The results show that three flow regimes appear in the reservoir with fractures within the test period: wellbore afterflow, pseudolinear flow, and radial flow. In contrast, only the wellbore afterflow and radial flow appear for the wells without fractures. The results also reveal that fractures dominate the flow near the wellbore. Verification and application of the model show the practicability of the integrated approach for investigating the transient-pressure behaviors in the unconventional reservoir.

A Leap into Automation for Advanced Fracture Characterization

Fractures are the prime conduits of flow for hydrocarbons in reservoir rocks. Identification and characterization of the fracture network yields valuable information for accurate reservoir evaluation. This study aims to portray the benefits and limitations for various existing fracture characterization methods and define strategic workflows for automated fracture characterization targeting both conventional and unconventional reservoirs separately. While traditional seismic provides qualitative information of fractures and faults on a macro scale, acoustics and other petrophysical logs provide a more comprehensive picture on a meso and micro level. High resolution image logs, with shallow depth of investigation are considered the industry standard for analysis of fractures. However, it is imperative to understand the framework of fracture in both near and far field. Various reservoir-specific collaborative workflows have been elucidated for a consistent evaluation of fracture network, results of which are further segregated using class-based machine learning techniques. This study embarks on understanding the critical requirements for fracture characterization in different lithological settings. Conventional reservoirs have good intrinsic porosity and permeability, yet presence of fractures further enhances the flow capacity. In clastic reservoirs, fractures provide an additional permeability assist to an already producible reservoir. In carbonate reservoirs, overall reservoir and production quality exclusively depends on presence of extensive fracture network as it quantitatively controls the fluid flow interactions among otherwise isolated vugs. Devoid of intrinsic porosity and permeability, the presence of open-extensive fractures is even more critical in unconventional reservoirs such as basement, shale-gas/oil and coal-bed methane, since it demarcates the reservoir zone and defines the economic viability for hydrocarbon exploration in reservoirs. Different forward modeling approaches using the best of conventional logs, borehole images, acoustic data (anisotropy analysis, borehole reflection survey and stoneley waveforms) and magnetic resonance logs have been presented to provide reservoir-specific fracture characterization. Linking the resolution and depth of investigation of different available techniques is vital for the determination of openness and extent of the fractures into the formation. The key innovative aspect of this project is the emphasis on an end-to-end suitable quantitative analysis of flow contributing fractures in different conventional and unconventional reservoirs. Successful establishment of this approach capturing critical information will be the stepping-stone for developing machine learning techniques for field level assessment.

Investigating Transfer Learning for Characterization and Performance Prediction in Unconventional Reservoirs

Transfer learning is a machine learning concept whereby the knowledge gained (e.g., a model developed) in one task can be transferred (applied) to solve a different but related task. In the context of unconventional reservoirs, the concept can be used to transfer a machine learning model that is learned from data in one field (or shale play) to another, thereby significantly reducing the data needs and efforts to build a new model from scratch. In this work, we study the feasibility of developing deep learning models that can capture and transfer common features in a rich dataset pertaining to a mature unconventional play to enable production prediction in a new unconventional play with limited available data. The focus in this work is on method development using simulated data that correspond to the Bakken and Eagle Ford Shale Plays as two different unconventional plays in the US. We use formation and completion parameter ranges that correspond to the Bakken play with their simulated production responses to explore different approaches for training neural network models that enable transfer learning to predict production responses of input parameters corresponding to the Eagle Ford play (previously unseen input parameters). We explore different schemes by accessing the internal components of the model to extrapolate and categorize salient features that are represented in the trained neural network. Ultimately, our goal is to use these new mechanisms to enable effective sharing and reuse of discovered features from one unconventional well to another. To extract salient trends from formation and completion input parameters and their corresponding simulated production responses, we use deep learning architectures that consist of convolutional encoder-decoder networks. The architecture is then trained with rich simulated data from one field to generate a robust mapping between the input and the output feature spaces. The "learned" parameters from this network can then be "transferred" to develop a different predictive model for another field that may lack sufficient historical data. The results show that using standard training approaches, a neural network model that is trained with sufficiently large data samples from Bakken could produce reliable prediction models for typical wells that may be found in that field. The same neural network, however, could not produce reliable predictions for a typical Eagle Ford well. Furthermore, we observe that a neural network trained with insufficient data samples from Eagle Ford produces a poor prediction model for typical wells that may be found in Eagle Ford. However, when extrapolated feature components of the Bakken neural network were integrated into the training process of the Eagle Ford neural network, the resulting predictions for typical Eagle Ford wells improved significantly. Moreover, we observe that the ability to transfer learning can improve when specialized training strategies are adopted to enable transfer learning. Using several numerical experiments, the paper presents and assesses various transfer learning strategies to predict the production performance of unconventional wells in a new area with limited information by integrating knowledge from more mature plays.

Integrating Deep Learning and Physics-Based Models for Improved Production Prediction in Unconventional Reservoirs

The physics of fluid flow and transport processes in hydraulically fractured unconventional reservoirs are not well understood. As a result, the predicted production behavior using conventional simulation often does not agree with the observed field performance data. The discrepancy is caused by potential errors in the simulation model and the physical processes that take place in complex fractured rocks subjected to hydraulic fracturing. Additionally, other field data such as well logs and drilling parameters containing important information about reservoir condition and reservoir characteristics are not conveniently integrated into existing simulation models. In this paper, we discuss the development of a deep learning model to learn the errors in simulation-based performance prediction in unconventional reservoirs. Once trained, the model is expected to forecast the performance response of a well by augmenting physics-based predictions with the learned prediction errors from the deep learning model. To learn the discrepancy between simulated and observed production data, a simulation dataset is generated by using formation, completion, and fluid properties as input to an imperfect physics-based simulation model. The difference between the resulting simulated responses and observed field data, together with collected field data (i.e. well logs, drilling parameters), is then used to train a deep learning model to learn the prediction errors of the imperfect physical model. Deep convolutional autoencoder architectures are used to map the simulated and observed production responses into a low-dimensional manifold, where a regression model is trained to learn the mapping between collected field data and the simulated data in the latent space. The proposed method leverages deep learning models to account for prediction errors originating from potentially missing physical phenomena, simulation inputs, and reservoir description. We illustrate our approach using a case study from the Bakken Play in North Dakota.