Determination of an infill well placement using a data-driven multi-modal convolutional neural network
- 주제(키워드) Infill well , Convolutional neural network , Multi-modal learning , Productivity
- 주제(기타) Energy & Fuels
- 주제(기타) Engineering, Petroleum
- 설명문(일반) [Chu, Min-gon] Korea Natl Oil Corp, E&P Technol Ctr, 305 Jongga Ro, Ulsan 44538, South Korea; [Min, Baehyun; Kwon, Seoyoon; Park, Gayoung] Ewha Womans Univ, Dept Climate & Energy Syst Engn, 52 Ewhayeodae Gil, Seoul 03760, South Korea; [Min, Baehyun] Ewha Womans Univ, Ctr Climate Environm Change Predict Res, 52 Ewhayeodae Gil, Seoul 03760, South Korea; [Kim, Sungil] Korea Inst Geosci & Mineral Resources, Petr & Marine Res Div, 124 Gwahak Ro, Daejeon 34132, South Korea; [Nguyen Xuan Huy] Ho Chi Minh Univ, VNU HCM, Fac Geol & Petr Engn, 268 Ly Thuong Kiet,Dist 10, Ho Chi Minh City, Vietnam
- 관리정보기술 faculty
- 등재 SCIE, SCOPUS
- 발행기관 ELSEVIER
- 발행년도 2020
- URI http://www.dcollection.net/handler/ewha/000000175098
- 본문언어 영어
- Published As http://dx.doi.org/10.1016/j.petrol.2019.106805
초록/요약
This study determines the optimal placement for a vertical infill well using a multi-modal convolutional neural network (CNN). 3D arrays composed of static and dynamic reservoir properties near a candidate infill well are inputted to the convolution stage of CNN. Multi-modal learning is applied to CNN for feature extraction of inputs. The features are compressed via fully connected layers for evaluating the productivity of every candidate infill scenario. The proposed CNN is applied to a channelized oil reservoir, and its performance is compared to that of a feedforward neural network. Dataset for the neural networks is obtained by running full-physics simulations for selected scenarios. CNN outperforms the feedforward neural network for the test scenarios of single- and dualmodal cases. Both neural networks yield comparable predictability for a quad-modal case. Results of the quad-modal CNN are in agreement with reservoir simulation results at cheaper computational costs. The results highlight the potential of data-driven machine learning in expediting the optimal well placement by partially replacing expensive simulation runs.
more