Convolutional neural networks (CNN) for feature-based model calibration under uncertain geologic scenarios
- PDF / 19,419,848 Bytes
- 25 Pages / 595.224 x 790.955 pts Page_size
- 56 Downloads / 153 Views
ORIGINAL PAPER
Convolutional neural networks (CNN) for feature-based model calibration under uncertain geologic scenarios Syamil Mohd Razak1 · Behnam Jafarpour1 Received: 5 August 2019 / Accepted: 28 April 2020 © Springer Nature Switzerland AG 2020
Abstract This paper presents convolutional neural network architectures for integration of dynamic flow response data to reduce the uncertainty in geologic scenarios and calibrate subsurface flow models. The workflow consists of two steps, where in the first step the solution search space is reduced by eliminating unlikely geologic scenarios using distinguishing salient flow data trends. The first step serves as a pre-screening to remove unsupported scenarios from the full model calibration process in the second step. For this purpose, a convolutional neural network (CNN) with a cross-entropy loss function is designed to act as a classifier in predicting the likelihood of each scenario based on the observed flow responses. In the second step, the selected geologic scenarios are used in another CNN with an 2 -loss function (as a regression model) to perform model calibration. The regression CNN model (step 2) learns the inverse mapping from the production data space to the lowrank representation of the model realizations within the feasible set. Once the model is trained off-line, a fast feed-forward operation on the observed historical production data (input) is used to reconstruct a calibrated model. The presented approach offers an opportunity to utilize flow data in identifying plausible geologic scenarios, results in an off-line implementation that is conveniently parallellizable, and can generate calibrated models in real time, i.e., upon availability of data and without in-depth technical expertise about model calibration. Several synthetic Gaussian and non-Gaussian examples are used to evaluate the performance of the method. Keywords Inverse problems · Machine learning · Uncertainty · Geologic scenarios · Convolutional neural networks · Model calibration
1 Introduction Developing a computer model that can be used to simulate and predict the performance of subsurface flow systems often constitutes a complex multi-stage process that involves acquisition, processing, integration, and interpretation of various types of data. The model development process is composed of hierarchical sequential steps, each with its own sources of uncertainty and subjectivity that must be captured and propagated along the chain. Some of the major sources of uncertainty tend to be present at early Behnam Jafarpour
[email protected] Syamil Mohd Razak [email protected] 1
Mork Family Department of Chemical Engineering and Material Science, Viterbi School of Engineering, University of Southern California, Los Angeles, CA 90089, USA
stages, where very limited data is used to develop the reservoir structure and the conceptual geologic model for the reservoir. Faced with limited data, geoscientists often have to resort to subjective assumptions in developing a conceptual model tha
Data Loading...
