Brittleness Index Prediction From Conventional Well Logs in Unconventional Reservoirs Using Artificial Intelligence

Author(s):  
Xian Shi ◽  
Gang Liu ◽  
Shu Jiang ◽  
Lei Chen ◽  
Liu Yang
2021 ◽  
Author(s):  
Syamil Mohd Razak ◽  
Jodel Cornelio ◽  
Atefeh Jahandideh ◽  
Behnam Jafarpour ◽  
Young Cho ◽  
...  

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.


2017 ◽  
Author(s):  
R. M. Alloush ◽  
S. M. Elkatatny ◽  
M. A. Mahmoud ◽  
T. M. Moussa ◽  
A. Z. Ali ◽  
...  

2021 ◽  
Author(s):  
◽  
Raul Correa Rechden Filho

<p>Within New Zealand the East Coast Basin encompasses the primary shale oil and gás (unconventional) play areas in which both the Waipawa and Whangai formations are widespread. These formations are oil and gas prone and prevalent throughout a large area of the East Coast Basin. To characterise these two formations and evaluate their shale oil and gas potential, existing analytical results were supplemented by a set of new sample analyses of organic and inorganic geochemistry, and rock properties. Thus, some 242 samples from the Whangai Formation have organic geochemical analyses and 40 have inorganic geochemical analyses; for the Waipawa Formation there are 149 organic and 9 inorganic geochemical analyses. In addition, downhole logs from three exploration wells have been used to calculate the brittleness index of the Whangai Formation. All these data have been grouped by structural block and used to determine where the sweet spots are in each formation. Both basic and more robust statistical analysis (machine-learning) is applied to identify the best prospective area. The Rakauroa Member (Whangai Formation) and the Waipawa Formation have the best rock characteristics as unconventional reservoirs, based on quantity and quality. Maturation appears to be an issue for these formations, although there are some localised areas where the Whangai Formation has better maturity. The brittleness index is calculated only for the Rakauroa Member, given the lack of data available for other members of the Whangai Formation and the Waipawa Formation, and yielded promising results. The Motu block appears to be the best area in which to explore for unconventional oil and gas. The prospective resource volumes for the best case scenario for the Whangai (Rakauroa Member) and Waipawa formations combined in the Motu Block are 17% higher (713MMbbl) than the 2P (proved + probable) reserves of New Zealand for oil and condensate (588MMbbl) and 26% (2.1TCF) of the 2P (proved + probable) reserves of natural gas (7.8 TCF). Economic analysis shows feasibility to explore these unconventional reservoirs for both shale oil or shale gas with an oil price of US$60 for both methodologies tested. However, the methodology applied using standard shale oil and gas assessments shows feasibility only for shale oil. Shale gas would not be economic, unless a higher oil prices, lower costs or a technology was developed to improve the recovery factor of these reservoirs. These results indicate a minimum economic field size of 4.5 km² for this area.</p>


Sign in / Sign up

Export Citation Format

Share Document