인공신경망 해석 기법을 이용한 저류층 유체투과율 결정

임종세 한국자원공학회 2003 한국자원공학회지 Vol.40 No.4

Reservoir permeability is one of the most important parameters of oil and gas fields for reservoir characterization. Although formation permeability is often measured in a laboratory from cores or evaluated from well test data, core analysis and well test data are only available from a few wells in a field, while most wells are logged. However, permeability determination from geophysical well logs in heterogeneous formation has a difficult and complex problem. In this study, an artificial neural network(ANN) model has been developed using core analysis data and well logs to determine reservoir permeability. The ANN is trained to predict formation permeability with well log data as input. The technique is demonstrated with an application to the well data in offshore Korea. The results of this study show that permeability distribution derived from the ANN model is consistent with core analysis results. The ANN is capable of providing more accurate and reliable results compared with conventional methods and statistical regression. The ANN is a practical and powerful tool for predicting reservoir permeability of a heterogeneous formation utilizing well logs. 저류층 유체투과율은 석유 및 천연가스 저류층 특성화에 있어 가장 중요한 특성변수이다. 일반적으로 코어분석이나 유정시험을 하는 과정에서 유체투과율을 측정 할 수 있으나, 이 방법들은 선택된 유정 중 일부 구간에서만 값을 얻을 수 있어 사용에 제한적이다. 반면 물리검층 자료는 거의 모든 유정에서 얻을 수 있다. 물리검층을 이용한 유체투과율 결정도 불균질한 저류층에서 그 값을 구하는 것은 매우 복잡하며 많은 어려움이 있다. 이 연구에서는 가능한 모든 물리검층 자료를 대상으로 학습과정을 거쳐 일정한 패턴을 얻고 이를 통해 유체투과율을 예측하는 인공신경망(ANN) 해석 기법을 제시하였다. 이 연구의 해석방법을 국내대륙붕 유정 자료에 적용하여 유체투과율을 예측한 결과 ANN 방법을 통해 얻은 유체투과율 값이 가장 코어분석 결과와 일치하는 분포를 나타냈으며 회귀분석 등 기존의 통계적 방법에 비해 보다 신뢰성 있고 정확한 값을 제공함을 알 수 있었다. 인공신경망 기법은 불균질한 저류층의 물리검층 자료를 이용한 유체투과율 결정 시 유용한 해석 도구로 활용될 수 있을 것이다.

가스전 회수증진 기술 동향 및 적용사례

임종세,서광원,이선아,이정환,안승희,신창훈 한국자원공학회 2010 한국자원공학회지 Vol.47 No.5

Enhanced gas recovery means to increase gas production when a reservoir is difficult to produce commercially natural gas. There are several engineering approaches such as water shut off, CO2 injection, infill drilling, hydraulic fracturing etc. in enhanced gas recovery technique. The water shut off is a technique that control water production by applying various mechanical and chemical methods in the case that a production well has trouble with excessive water production from aquifer. The CO2 injection is method to improve natural gas production by injecting CO2 into reservoir for re-pressurization of the reservoir and displacement of natural gases. This method has been received attention recently because of its possibility for connecting with carbon capture and sequestration technique. To apply the infill drilling technique, petrophysical properties in reservoir are estimated by analysing static and dynamic data from the gas production field and a location of infill drilling are determined by various analysis techniques like ‘Moving Window’ method, reservoir simulation based method etc. The hydraulic fracturing is the way to enhance mobility and connectivity by creating artificially fractures in well vicinity. In this study, various methods for enhancing gas recovery are investigated and its applications are analysed. 가스전 회수증진은 상업적 생산이 어려운 저류층에 대해 다양한 생산기법을 적용함으로써 회수율을 증진시키는 것을 의미한다. 가스 회수증진 기법으로는 지층수 차단, 이산화탄소 주입, 추가 시추, 수압파쇄 등이 있다. 지층수 차단은 대수층으로부터 다량의 물이 유입되는 시추공에 기계적 및 화학적 기법을 적용함으로써 물의 생산을 억제하는 방법이다. 또한 최근 이산화탄소 지중처분기술과 연계하여 이산화탄소를 주입하여 저류층을 재가압하고 가스를 대체함으로써 회수율을 증진시키는 방법이 주목받고 있다. 추가 시추의 경우 기생산 가스전의 정적 및 동적 자료를 해석하여 저류층의 물성을 파악하고 ‘Moving Window’ 기법, 저류층 시뮬레이션 기반 기법 등을 통해 생산성을 향상시킬 수 있는 위치를 선정하여야 한다. 수압파쇄법은 시추공 부근에 인공적인 균열대를 형성하여 저류층의 유동도 및 연결성을 향상시키는 방법이다. 이 연구에서는 다양한 가스회수증진 기법에 대해 조사하고 각각에 대한 적용사례를 분석하였다.

Gas Production Optimization of Multi-Reservoir Fields using Branch-System

임종세,오정용,박희준,Joe M. Kang 한국자원공학회 2006 Geosystem engineering Vol.9 No.3

The analysis of production systems in multi-reservoir gas fields is required to optimize production rates and enhance ultimate recovery. Conventional methods involve enormous computational efforts since production systems from reservoirs to a gathering station have been considered as one structure and production rates are optimized step-by-step in terms of objective functions. In this study, a new technique is developed to reduce the structural complexities of previous techniques and that consists of branch-systems including reservoirs and a surface branch-system without reservoirs. All branch-systems can be treated independently in this method. The decision variables including pressures and production rates in a branch-system, which are determined by the size of reservoir, the producing capacity of reservoir, and the initial pressure of reservoir, are updated simultaneously at every iteration step to accelerate convergence. The method can analyze multi-reservoir gas fields with a fixed gross production rate at the gathering station efficiently. The results for two fields have shown that the method required less computational efforts and provided relatively accurate outputs for conventional techniques applied to two fields.

Application of various deterministic decline curve analyses in resource plays

임종세,신효진,김지수 한국자원공학회 2015 Geosystem engineering Vol.18 No.1

Resource plays have recently received increasing attention. This is because of rising oil prices caused by the growing global demand for petroleum resources and improved production technologies for productivity. Conducting reservoir evaluation in target areas is a priority for oil and gas play development. Petroleum production forecasting is important for well performance evaluations, which are generally routinely performed using decline curve analysis (DCA). DCA relies on the use of empirical curve fitting or the Arps decline model. However, the period and distribution of production data, the differences in the initial production rates of well stimulation technologies, etc., that are derived by estimation have inaccurate results in resource plays. Therefore, many kinds of DCA are proposed to reflect these characteristics of production data. In this study, we applied a variety of deterministic DCA methods using production data of different feature in resource play and investigated the impact caused by the production data. The results show that production trends and estimated ultimate recovery are predicted for each field up to 360 months. In addition, this study can be used as basis process in choosing suitable DCA methods in resource plays.

리마글러벌의 과실 해외 수출사례

임종세 한국무역학회 2014 한국무역학회 학술대회 Vol.2014 No.4

북극 석유·천연가스 자원 기술개발 현황 및 전망

임종세,신효진,김지수,진영근 한국해양과학기술원 2014 Ocean and Polar Research Vol.36 No.3

The rising global demand for energy resources may lead to greater interest in the Arctic region. Since it has various resources, such as oil and gas, and large potential as a strategic location in exploration and production (E&P), there is likely to occur island sovereignty issues between the five arctic costal states and other countries. While global warming has led to the opening of the Northeast Passage and the Northwest Passage, several obstacles may impede the development of this area such as the low temperature environment, infrastructure problems in a limited area, flow assurance, environmental regulations, etc. To overcome these problems, various techniques have been applied in the exploration, development, production, transportation, and environment fields and it seems to be made technical development in extreme environment. In this study, the E&P status of representative states and development technologies in the Arctic region have been summarized with regard to carrying out E&P related to drilling, development, production, and operation in oil and gas fields. Furthermore, environmental factors have been taken into account to enhance progress with regard to E&P and ensure sustainable development in the Arctic. On that basis, it will be possible to secure oil and gas field development, production technology and R&D infrastructure in the Arctic.

육상 석유 심부시추 기술 동향 및 적용사례

임종세,강판상,김현태 한국자원공학회 2010 한국자원공학회지 Vol.47 No.5

Unlike the shallow drilling, deep oil well drilling may be operated in the high formation pressure and high temperature conditions. In these drilling environments, kick, differential sticking, damage of drilling equipments or lost circulation may occur more frequently. Drilling cost and non-productive time could be increased accordingly. More improved equipments and technologies are required to overcome these difficulties and improve productivity while drilling deep formation. The goals of this study are to investigate current technologies for deep oil well drilling such as underbalanced drilling, managed pressure drilling, casing drilling, expandable tubular, mud hammers, mud-pulse drilling, and their applications at Tizon Field in Mexico and El Furrial Field in Venezuela. 석유 심부시추는 천부시추와는 달리 시추 깊이가 깊어질수록 온도 및 압력의 증가로 작업이 어려운 환경에서 시추가 진행된다. 이로 인해 킥, 편향점착, 장비손상, 이수손실 등이 더 자주 발생할 가능성이 높아지므로 시추비용이 증가하고 작업중단기간이 매우 길어질 수 있다. 이러한 시추 환경을 극복하고 생산성을 향상시키기 위해서는 보다 향상된 장비와 기술이 필요하다. 이 연구에서는 석유 심부시추 기술 중 공내감압 시추법 및 압력제어 시추법과 적용사례인 멕시코의 Tizon Field, 베네수엘라의 El Furrial Field에 대해 알아보고자 한다. 또한 최신 석유 심부시추 기술인 케이싱 시추법, 확장성 파이프, Mud hammers, 이수 파동 시추법에 대해 소개하고자 한다.

지구통계학적 기법을 이용한 터널 주변 암반물성의 3차원 공간정보 모델링

임종세,최용근,김태혁 한국암반공학회 2004 한국암반공학회 학술대회 및 세미나 자료집 Vol.- No.-

검층자료로부터 저류층 유체투과율 추정을 위한 다항식 신경망 기법의 적용

임종세,박희준 한국자원공학회 2005 한국자원공학회지 Vol.42 No.3

Reservoir permeability is one of the most important parameters of oil and gas fields for reservoir characterization. In un-cored intervals, reservoir description and evaluation methods using well log data represent a significantly technical and economic advantage because well logs can provide a continuous record over the entire well. Permeability estimation from well logs in heterogeneous formation has a difficult and complex problem to solve by conventional statistical methods. Recently artificial neural networks(ANN) have been successfully used to solve the complex problems in reservoir permeability determination. However, the applications of the neural network to mapping complex nonlinear relationship have revealed a number of unsolved technical limitations despite of the high versatility. This paper proposes a group method of data handling(GMDH) based on an advanced polynomial neural network(PNN) for permeability prediction from well logs to alleviate limitations of the conventional neural network approach. 저류층 유체투과율은 석유 및 천연가스 저류층 특성화에 가장 중요한 특성변수이다. 코어가 채취되지 않은 구간에서 유체투과율을 추정하는 데 있어 거의 모든 유정에서 전 구간에 제공되는 물리검층자료의 활용은 상당한 이점이 있다. 물리검층을 이용한 유체투과율 추정에 일반적인 통계적 방법을 이용하여 불균질한 저류층에서 그 값을 구하는 것은 매우 복잡하며 많은 어려움이 있다. 최근 이에 대한 대안으로 인공신경망 기법을 이용한 해석방법이 활용되고 있으나 기존의 인공신경망 기법은 사용에 많은 한계점이 지적되었다. 이 연구에서는 인공신경망 기법의 문제점을 해결하기 위해 GMDH 알고리즘에 기반한 개선된 다항식 신경망(PNN) 해석법을 제시하고 이를 검층자료로부터 저류층 유체투과율을 추정하는 데 활용한 것이다.

석유/천연가스 매장량의 분류 및 평가방법

임종세,김지영 한국해양대학교 해양과학기술연구소 2001 硏究論文集 Vol.10 No.1

It is essential for future planning that governments and industry have a clear assessment of the quantities of petroleum available for production and quantities which are anticipated to become available within a practical time frame through additional field development, technological advances, or exploration. This paper describes the petroleum reserves classification and estimating methods based on the SPE/WPC (1997) definition. Reserves are those quantities of petroleum which are anticipated to be commercially recovered from known accumulations from a given date forward. Reserve estimating methods usually are categorized into analogy, volumetric and performance techniques. The performance techniques are subdivided into simulation studies, material balance calculations, and decline trend analysis. Reserve can be estimated using either deterministic or probabilistic methods. This paper can be helpful to understand petroleum reserve estimation and evaluate the reserves in oil & gas development projects properly.