Zhou, Xu (Louisiana State Unviersity) | Tyagi, Mayank (Louisiana State Unviersity) | Zhang, Guoyin (China University of Petroleum - Beijing) | Yu, Hao (Southwest Petroleum University) | Chen, Yangkang (Zhejiang University)
With recent developments in data acquisition and storage techniques, there exists huge amount of available data for data-driven decision making in the Oil & Gas industry. This study explores an application of using Big Data Analytics to establish the statistical relationships between seismic attribute values from a 3D seismic survey and petrophysical properties from well logs. Such relationships and models can be further used for the optimization of exploration and production operations.
3D seismic data can be used to extract various seismic attribute values at all locations within the seismic survey. Well logs provide accurate constraints on the petrophysical values along the wellbore. Big Data Analytics methods are utilized to establish the statistical relationships between seismic attributes and petrophysical data. Since seismic data are at the reservoir scale and are available at every sample cell of the seismic survey, these relationships can be used to estimate the petrophysical properties at all locations inside the seismic survey.
In this study, the Teapot dome 3D seismic survey is selected to extract seismic attribute values. A set of instantaneous seismic attributes, including curvature, instantaneous phase, and trace envelope, are extracted from the 3D seismic volume. Deep Learning Neural Network models are created to establish the relationships between the input seismic attribute values from the seismic survey and petrophysical properties from well logs. Results show that a Deep Learning Neural Network model with multi-hidden layers is capable of predicting porosity values using extracted seismic attribute values from 3D seismic volumes. Ultilization of a subset of seismic attributes improves the model performance in predicting porosity values from seismic data.
With the recent tremendous development in algorithms, computations power and availability of the enormous amount of data, the implementation of machine learning approach has spurred the interest in oil and gas industry and brings the data science and analytics into the forefront of our future energy. The idea of using automated algorithms to determine the rock facies is not new. However, the recent advancement in machine learning methods encourages to further research and revisit the supervised classification tasks, discuss the methodological limits and further improve machine learning approach and classification algorithms in rock facies classification from well-logging measurements. This paper demonstrates training different machine learning algorithms to classify and predict the geological facies using well logs data. Previous and recent research was done using supervised learning to predict the geological facies.
This paper compares the results from the supervised learning algorithms, unsupervised learning algorithms as well as a neural network machine learning algorithm. We further propose an integrated approach to dataset processing and feature selection. The well logs data used in this paper are for wells in the Anadarko Basin, Kansas. The dataset is divided into training, testing and evaluating wells used for testing the model. The objective is to evaluate the algorithms and limitations of each algorithm. We speculate that a simple supervised learning algorithm can yield score higher than neural network algorithm depending on the model parameter selected. Analysis for the parameter selection was done for all the models, and the optimum parameter was used for the corresponding classifier.
Our proposed neural network algorithm results score slightly higher than the supervised learning classifiers when evaluated with the cross-validation test data. It is concluded that it is important to calculate the accuracy within the adjacent layers as there are no definite boundaries between the layers. Our results indicate that calculating the accuracy of prediction with taking account the adjacent layers, yield higher accuracy than calculating accuracy within each point. The proposed feed-forward neural network classifier trains using backpropagation (gradient descent) provides accuracy within adjacent layers of 88%. Our integrated approach of data processing along with the neural network classifier provides more satisfactory results for the classification and prediction problem. Our finding indicates that utilizing simple supervised learning with an optimum model parameter yield comparable scores as a complex neural network classifier.
Fu, Xuebing (Goolsby Finley & Associates) | Bonifas, Paul (Goolsby Finley & Associates) | Finley, Andy (Goolsby Finley & Associates) | Lemaster, Julia (Goolsby Finley & Associates) | He, Zhiyong (ZetaWare, Inc) | Venepalli, Kiran (CMG Inc.)
Over the last decade, tight oil production has become significant with the success of horizontal drilling and hydraulic fracturing. However, the recovery factor of tight oil production remains very low and no standard secondary recovery method exists after primary depletion. We propose a new secondary recovery method: to use existing hydraulic fractures (every other fracture) in a horizontal well as gas injection and oil production sites to conduct
An ideal process of this method envisions a horizontal well centered in an enclosed reservoir, where the hydraulic fractures are evenly distributed along the well, parallel to each other. If the hydraulic fractures can be effectively isolated, and injection and production can be conducted through alternate fractures at the wellbore, then highly efficient flooding patterns can be created. Key questions include: Is there adequate injectivity and productivity in a sub-microdarcy reservoir? How far are the ideal reservoir conditions from reality? How difficult is it to isolate individual fractures within the wellbore?
Is there adequate injectivity and productivity in a sub-microdarcy reservoir?
How far are the ideal reservoir conditions from reality?
How difficult is it to isolate individual fractures within the wellbore?
In addressing these key questions, first, the success of a flooding process relies on reasonable injectivity and response time between the injector and producer – in this case the injector being one hydraulic fracture and the producer being an adjacent hydraulic fracture; both economical rates and reasonable communication time between adjacent fractures are demonstrated through analytical calculations and reservoir simulations in a typical well setting; nearly 100% recovery is achieved in the reservoir units between the fractures in a miscible flooding process. Second, actual reservoir conditions are incorporated in our study, focusing on direct fracture communications; the effects are demonstrated, and comparisons among different methods are made. Finally, potential challenges in operations are summarized and current technologies are reviewed; the gaps between the current settings and the required settings are demonstrated. Economic discussions are made, indicating positive scenarios with large tolerances.
With the rapid development of tight oil reservoirs, Enhanced Oil Recovery (EOR) technologies are urgently required to improve the recovery factor beyond primary depletion. An effective flooding process may be conducted if the hydraulic fractures can be used as injection and production ports. As a first attempt to envision an inter-fracture flooding process, key aspects are defined and examined, showing promising results. Inter-fracture gas flooding may become a standard secondary recovery technique for tight oil reservoirs and add significant reserves.
Electromagnetic images can show where water flows during a hydraulic fracture. A test in the Anadarko Basin showed a fault there was a bigger hazard than expected. Electromagnetic (EM) reservoir imaging is likely to get more attention from operators thanks to a collaboration between Halliburton and a leader in this emerging technology, GroundMetrics.
Electromagnetic images can show where water flows during a hydraulic fracture. A test in the Anadarko Basin showed a fault there was a bigger hazard than expected. Good diagnostic testing is often painstaking, time-consuming, and costly, but recent studies suggest that a lack of knowledge can be even costlier. The future of unconventional exploration will require a break from the status quo. With well productivity stalling, it is time to look for a new plan of attack.
Producers in Oklahoma’s newly opened Merge play are sitting atop a resource that rivals some major world gas fields and discoveries, Citizen Energy’s Geology CEO Greg Augsburger told the SPE Gulf Coast Section Business Development Group recently. Linn Energy recently sold its Williston Basin properties for $285 million. This deal brings Linn’s year-to-date total sales agreements to more than $1.5 billion as it financially restructures after bankruptcy. At the recent Leaders in Industry luncheon in Houston, Jonny Jones told the interesting story of how Jones Energy grew from a small private entity to a company whose shares trade on the New York Stock Exchange.
Considering most of the rigs deal with human-machine interface systems, the role of human factors is at the heart of any successful operation. Eye-tracking technology can be useful in real-time operation centers where ocular movement data can improve the professionals’ performance. Environmentalists have sued a US agency to try to stop it from allowing oil and gas drilling on a vast stretch of federal land in Nevada, where the government is reversing protections put in place 9 months ago under the Obama administration. Two months after a Colorado home exploded near an Anadarko well, the reverberations are still rattling the oil industry, driving down driller shares and raising fears of a regulatory backlash. More than 100 members of Congress are urging the Trump administration not to open up the Atlantic or Pacific oceans for oil and gas drilling as part of the Interior Department’s review of federal offshore policies.
Anadarko Petroleum wants a fleet of at least six vehicles with armor heavy enough to stop AK-47 bullets at its natural-gas project in Mozambique. And it needs them soon. Anadarko Petroleum said late on 30 June that it has tested more than 4,000 active oil and gas lines and plugged another 2,400 inactive ones per a state order issued after a fatal home explosion in Firestone, Colorado, in April. Two months after a Colorado home exploded near an Anadarko well, the reverberations are still rattling the oil industry, driving down driller shares and raising fears of a regulatory backlash. The company that owns a gas well linked to a fatal home explosion in Colorado says it will permanently shut down that well and two others in the neighborhood.