DeepSim: cluster level behavioural simulation model for deep learning

2019 ◽  
Vol 6 (3/4) ◽  
pp. 224
Author(s):  
Kaushik Balasubramanian ◽  
Adam Procter ◽  
Ramesh Illikkal ◽  
Zhaojuan Bian ◽  
Kevin J. Long ◽  
...  
Keyword(s):  
Deep Learning ◽  
Simulation Model ◽  
Behavioural Simulation ◽  
Cluster Level

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

2021 ◽  
Author(s):  
Syamil Mohd Razak ◽  
Jodel Cornelio ◽  
Atefeh Jahandideh ◽  
Behnam Jafarpour ◽  
Young Cho ◽  
...  
Keyword(s):  
Deep Learning ◽  
Simulation Model ◽  
Field Data ◽  
Learning Model ◽  
Transport Processes ◽  
Well Logs ◽  
Unconventional Reservoirs ◽  
Prediction Errors ◽  
Drilling Parameters ◽  
Deep Learning Model

Abstract 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.

scholarly journals Physics-Aware Deep-Learning-Based Proxy Reservoir Simulation Model Equipped With State and Well Output Prediction

2021 ◽  
Vol 7 ◽  
Author(s):  
Emilio Jose Rocha Coutinho ◽  
Marcelo Dall’Aqua ◽  
Eduardo Gildin
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Simulation Model ◽  
Reservoir Simulation ◽  
The State ◽  
Data Driven ◽  
Training Procedure ◽  
State Variables ◽  
Input Output ◽  
Modeling Framework

Data-driven methods have been revolutionizing the way physicists and engineers handle complex and challenging problems even when the physics is not fully understood. However, these models very often lack interpretability. Physics-aware machine learning (ML) techniques have been used to endow proxy models with features closely related to the ones encountered in nature; examples span from material balance to conservation laws. In this study, we proposed a hybrid-based approach that incorporates physical constraints (physics-based) and yet is driven by input/output data (data-driven), leading to fast, reliable, and interpretable reservoir simulation models. To this end, we built on a recently developed deep learning–based reduced-order modeling framework by adding a new step related to information on the input–output behavior (e.g., well rates) of the reservoir and not only the states (e.g., pressure and saturation) matching. A deep-neural network (DNN) architecture is used to predict the state variables evolution after training an autoencoder coupled with a control system approach (Embed to Control—E2C) along with the addition of some physical components (loss functions) to the neural network training procedure. Here, we extend this idea by adding the simulation model output, for example, well bottom-hole pressure and well flow rates, as data to be used in the training procedure. Additionally, we introduce a new architecture to the E2C transition model by adding a new neural network component to handle the connections between state variables and model outputs. By doing this, it is possible to estimate the evolution in time of both the state and output variables simultaneously. Such a non-intrusive data-driven method does not need to have access to the reservoir simulation internal structure, so it can be easily applied to commercial reservoir simulators. The proposed method is applied to an oil–water model with heterogeneous permeability, including four injectors and five producer wells. We used 300 sampled well control sets to train the autoencoder and another set to validate the obtained autoencoder parameters. We show our proxy’s accuracy and robustness by running two different neural network architectures (propositions 2 and 3), and we compare our results with the original E2C framework developed for reservoir simulation.

Sign in / Sign up

Export Citation Format

Share Document