북해 Volve 유전 현장자료의 공공데이터화와 저류층 모델에 대한 분석

권서윤,지민수,박가영,민배현,정훈영 한국자원공학회 2021 한국자원공학회지 Vol.58 No.4

Equinor, a Norwegian multinational energy company, disclosed approximately 5 TB reservoir big data of the Volve oilfield in the North Sea for academic purposes in June 2018. This disclosure is the first for oilfield data worldwide acquired during the whole life cycle of an oilfield. This data disclosure has been highlighted in areas with limited field data for educational and research purposes. This review introduces the big data of the Volve oilfield and analyze the reservoir model based on reservoir simulation. In addition, we discuss the significance of reservoir data opening that can contribute to the E&P business in the Republic of Korea.

다중비선형회귀분석을 이용한 탄산염암 저류층의 CO2-EOR 효율 평가에 관한 연구

권서윤,박가영,민배현,김기홍,이태엽,한정민 한국자원공학회 2020 한국자원공학회지 Vol.57 No.2

This study aims to derive an equation for a preliminary economic evaluation that quickly assesses the efficiency of CO2-EOR at carbonate oil reservoirs. Previous works have qualitatively analyzed the CO2-EOR efficiency by, for example, using a lookup table based on reservoir properties. This study selects a series of influential parameters on the CO2-EOR efficiency: five static (e.g., reservoir pressure and temperature, API gravity, porosity, and R35) and two dynamic properties (e.g., CO2 breakthrough time and recovery factor). A database for the derivation of the equation is built by running reservoir simulation with varying values of the influential parameters. By conducting a multiple nonlinear regression analysis, the equation is designed as a combination of oil initially in place and recovery factor. The five static properties were utilized as inputs of the equation, while the dependent variables representing the outputs were obtained by merging the two dynamic properties. The proposed economic evaluation system can be utilized as an efficient CO2-EOR screening tool for carbonate reservoirs under field exploration and development. 이 연구는 탄산염암 저류층에서 석유증진회수법인 CO2-EOR(carbon dioxide enhanced oil recovery)의 효율을 신속히 평가하는 사전 경제성 평가식을 도출한다. 선행 CO2-EOR 스크리닝 연구들은 저류층 물성에 기반한 순람표등을 활용하는 정성 평가에 집중하였다. 이 연구는 CO2-EOR 효율의 정량 평가를 위해 5개의 정적 물성(저류층 압력및 온도, API 비중, 공극률, R35)과 2개의 동적 물성(CO2 돌파 시간 및 회수율)을 영향인자로 선택한 후, 영향인자 값을변화시키며 저류층 시뮬레이션을 수행하여 평가식 설계를 위한 데이터베이스를 구축하였다. 평가식은 원시석유부존량과 회수율 지표의 조합으로 구성하였다. 회수율 지표는 정적인자들을 독립변수, 동적인자들을 종속변수로 설정한 후 다중비선형회귀분석을 수행하여 계산하였다. 제안한 사전 평가 시스템은 탐사 및 개발 중인 탄산염암 저류층을위한 저렴하고 효율적인 CO2-EOR 스크리닝 도구로 활용할 수 있다.

연구논문 : 전차병복 착용실태에 관한 연구

권서윤 ( Seo Yoon Kwon ),임채근 ( Chae Guen Lim ),신동우 ( Dong Woo Shin ),정현미 ( Hyun Mi Jung ) 한국의류산업학회 2011 한국의류산업학회지 Vol.13 No.4

The purpose of this study was to investigate problems of design, fitness, suitability for movement, and comfort in current Korean military tank driver`s clothing through analysis of actual wearing condition by questionnaire and field evaluation and. to provide basic data for developing a improved design of Korean military tank driver`s clothing. The survey was done for 477 military tank driver and the field evaluation was also done for evaluation. The overall satisfaction for design of military tank driver`s clothing(3.25) was higher than that for the easiness in wearing and taking off(2.76). The military tank drivers evaluated that current coverall type of clothing is more suitable than two-piece type of clothing. The overall satisfaction for fitness of clothing was as a whole low(2.82~3.09), Especially, the satisfaction for fitness of from front and back rise length was the lowest one. In the satisfaction for clothing materials, the satisfaction for the breathability of material was the lowest, followed by clothing insulation and air permeability. The satisfaction for movement was low in bending waist and raising forward and aside. The part which surveyors think most dissatisfactory was also front and back rise length. The frequency in use of pocket was the highest in chest pocket, followed by waist and pants pockets. The satisfaction for opening easiness of hips opening part was very low(2.64).

전차병 점퍼의 착용만족도 및 보온성에 관한 연구

권서윤 ( Seo Yoon Kwon ),최은미 ( Eun Mi Choi ),임채근 ( Chae Guen Lim ),신동우 ( Dong Woo Shin ),김경필 ( Kyung Pil Kim ),권오경 ( Oh Kyung Kwon ),정현미 ( Hyun Mi Jeong ) 한국의류산업학회 2012 한국의류산업학회지 Vol.14 No.2

The purpose of this study is to investigate problems of design, fitness, suitability for movement, and wearing comfort of jumper for Korean military tank drivers through analysis of actual wearing condition by questionnaire and field evaluation and to provide basic data for developing its improved design. The survey was done for 477 military tank drivers and evaluation was performed using thermal manikin to measure insulation. The overall satisfaction for design of jumper for military tank driver was over 3.5 (likert scale). The overall satisfaction for fitness of jumper for military tank driver was also over 3.5. The satisfactions for material was between 2.39 and 3.13 and the satisfaction for pilling property was the lowest, followed by static property and shape stability after laundering. The satisfactions for movement suitability were standing (3.81), sitting (3, 38), raising hand (forward: 2.90, sideward: 3.01), respectively. In insulation evaluation of jumper for military tank drivers and outwears (jacket, jumper), the insulation of jumper for military tank drivers was lower than outwear (jumper) and same with outwear (jacket). The insulation in dynamic and still condition (without wind) of jumper for military tank driver was 0.37clo and 0.31clo, respectively. Its decreation rate in dynamic condition comparing to still condition was 59% which was lower than jacket (0.73clo) and jumper (1.15clo).

물리검층자료를 활용한 딥러닝 기반 수포화도 예측

지민수,권서윤,박가영,민배현,Nguyen Xuan Huy 한국자원공학회 2021 한국자원공학회지 Vol.58 No.3

This study develops a surrogate model to predict water saturation from well log data using neural-network-based deep learning algorithms. The model performance is evaluated by comparing the water saturation estimates obtained using deep learning algorithms and Archie’s law. The surrogate model evaluates the water saturation of a target reservoir using four well-log data types (density, porosity, resistivity, and gamma ray). Long Short-Term Memory (LSTM) is employed as the deep neural network algorithm, and its performance is compared with that of a multi-layer artificial neural network. Prediction via the LSTM based model showed outstanding results with the coefficient of determination above 0.7. Sensitivity analysis is conducted through sequence tuning, switching of well type, and k-fold cross-validation. The applicability of the model has been validated for the Volve oilfield in the North Sea and an offshore oilfield in Vietnam. 이 연구는 다양한 물리검층자료를 학습한 딥러닝 알고리듬을 이용하여 저류층의 수포화도를 예측하는 대리 모델을 구축한다. 딥러닝 알고리듬으로 계산한 수포화도 추정치를 Archie 방정식 결과와비교함으로써 개발 모델의 성능 평가를 수행하였다. 이 연구는 4가지 물리검층자료(밀도, 공극률, 비저항, 감마선)를 심층신경망의 입력인자로 활용하여 수포화도를 평가하였다, 심층신경망 기법으로는 장단기메모리학습법을 사용하였으며 전형적인 다층 인공신경망과의 비교를 통해 성능을확인하였다. 장단기메모리학습법을 기반으로 수포화도를 예측한 결과, 결정계수의 값이 0.7 이상으로 우수한 성능을 보이는 것으로 확인하였다. 모델의 민감도 분석으로는 시퀀스 조정, 유정 역할전환, k-폴드 교차검증을 시행하였다. 모델의 적용 가능성은 북해 Volve 유전과 베트남 해상 유전에적용하여 성능을 검증하였다.

석유가스 개발사업의 인공지능기술 활용 현황 및 전망

민배현,권서윤,박가영,정대인,정대인 한국자원공학회 2020 한국자원공학회지 Vol.57 No.3

This study reviewed the current status and prospects of using artificial intelligence (AI) technology in oil and gas exploration and production (E&P) in facing the era of digital transformation. Beginning with a brief introduction to artificial intelligence, machine learning (ML), and deep learning (DL), this manuscript discusses the following: the use artificial intelligence in E&P projects, an introduction to the state-of-the-art deep learning techniques highlighted in recent E&P projects, an analysis of the trends in global and domestic E&P business, relevant considerations, and concluding remarks. Thus, how the oil and gas E&P business is encouraged to cope with and move forward in the era of digital transformation is examined in detail. 이 연구는 디지털 전환의 시대를 맞이하여 석유가스 개발사업의 인공지능기술 활용 현황과 전망을 살펴본다. 인공지능, 머신러닝, 딥러닝에 대한 간략한 소개를 시작으로 석유가스 개발사업에서 인공지능이 어떻게 활용되는지, 다양한 인공지능기술 중 최근 석유가스 개발사업에서 각광받는 딥러닝 기술의 종류에 대한 소개, 국내외 석유가스개발사업의 딥러닝 기술 활용 현황과 전망, 고찰, 그리고 맺음말로 구성하였다. 이를 토대로 디지털 전환의 시대에서석유가스 개발사업이 대처하고 나아가야 할 방안에 대하여 논하고자 한다.

가스하이드레이트 저류층 모델링을 위한 심층학습 기반 물리검층 해석의 최신 기술동향 분석

박가영,권서윤,지민수,민배현,Nguyen Xuan Huy,김광현,김성일,이경북 한국자원공학회 2021 한국자원공학회지 Vol.58 No.2

This study aims to provide the fundamentals of deep learning-based modeling for gas hydrate-bearing sediments. For this purpose, we analyze state-of-the-art neural network applications in geophysical logs for conventional oil and gas reservoirs and unconventional gas-hydrate-bearing sediments. According to the type of neural network outputs, the previous studies are categorized into four application types: estimation of reservoir lithofacies, estimation of fluid flow parameters, estimation of geomechanical parameters, and generation of synthetic logs. For each type, we describe the key points of previous studies and provide summary tables. 본 연구는 국내 가스하이드레이트 저류층에 대한 심층학습 모델링 연구의 기반을 제공하고자 한다. 이를 위해 유·가스 저류층 및 가스하이드레이트 저류층에서 물리검층 자료에 신경망 기법을 적용한 최신 기술동향을 분석한다. 물리검층 분야에서 신경망 기법을 적용한 국내외 연구사례들을신경망이 예측하고자 하는 출력자료 유형에 따라 저류층 암상 추정, 유체유동인자 추정, 지질역학인자 추정, 합성로그 생성 등 네 가지 적용 유형으로 분류하였다. 각 사례들의 주요 특징을 기술하는한편 특징들을 요약한 종합표를 제공한다.

딥러닝 알고리듬을 활용한 고해상도 물리검층 자료의 생성 연구

박가영,민배현,권서윤,김민,지민수,이수진,최수인 한국자원공학회 2022 한국자원공학회지 Vol.59 No.5

This study proposed a deep-learning-based approach that generates synthetic high-resolution log data from original-resolution log data for accurate reservoir characterization, where the resolution of the synthetic data is comparable to that of core data. The reliability of the proposed approach was tested with application to the Volve oil field in Norway using three deep-learning algorithms (i.e., deep neural network, convolutional neural network, and long short-term memory). These deep-learning algorithms were employed to generate high-resolution sonic log data from other log-type data. The overall performance of each algorithm was acceptable. In particular, the long short-term memory algorithm yields a coefficient of determination greater than 0.9 when the high-to-original-resolution ratios are two, five, and ten. We anticipate that the proposed model can be used to derive logging-based reservoir parameters with a resolution that is comparable to that of core-based reservoir parameters.

비전통 오일의 생산 공법

민배현,김민,권서윤 대한기계학회 2022 기계저널 Vol.62 No.7

한국 남성용 단일의복의 앙상블 조합시의 온열특성 변화에 관한 연구 -무풍, 풍속환경하에서-

송민규 ( Min Kyu Song ),권서윤 ( Seo Yoon Kwon ),정현미 ( Hyun Mi Jung ) 한국의류산업학회 2012 한국의류산업학회지 Vol.14 No.4

The purpose of this study is to analyze the thermal characteristics of garments marketed for Korean males and to investigate the influence of each garment on ensemble, by measuring their insulation values (clo) using thermal manikins. The results are as follows. The total insulations (clo) of ensembles for S/S seasons are between 1.46 and 2.6 clo, with the mean of 2.12 clo. The insulation in the still air condition is 1.23 clo, which means a decrease of 42% compared to the total insulation of all the component garments. The insulation of ensembles for S/S seasons in the dynamic air condition decreased by 46.8%, compared to the still air condition. The total insulation (clo) of ensembles for F/W seasons is between 3.84 and 7.36 clo with the mean of 4.74 clo. The insulation in the still air condition is 2.26 clo, which means a decrease of 53.6% compared to the total insulation of all the component garments. The insulation of ensembles for F/W seasons in the dynamic air condition decreased by 36.2%, compared to the still air condition. As the clo value of each component garment gets higher, the insulation of ensembles gets higher. Especially, the insulation of ensembles was more influenced by outer wear than inner wear. The insulation of ensembles could be predicted by the insulation of outerwear better.