Automated Well Log-Based Prediction of Formation Tops: Case Study of Norway Offshore Data

Author:

Shakirov A. B.1,Lipko A.1,Mezghani M.2

Affiliation:

1. Aramco Innovations LLC, Moscow, Russia

2. Saudi Aramco, Dhahran, Saudi Arabia

Abstract

Abstract A geological formation represents a combination of certain genetic types of rocks associated with the proximity of the sedimentation conditions occurring at certain stages of development of the main structural elements of the Earth's crust. In petroleum engineering, the knowledge on spatial distribution of formations is necessary for many applied tasks including construction of hydrocarbon field geological model, petrophysical modelling, basin and petroleum system modelling, calculation of hydrocarbon reserves, planning field development strategy, and many others [Ahmed, 2018; Garb, 1985]. There are different approaches in upstream industry (mainly in geophysics) to determine spatial distribution of formations and depth intervals of wells where the formations are altered. One of the main geophysical methods that enables reliable prediction of spatial distribution of the formation is seismic survey. The main benefit of the seismic survey is that it covers large areas and has solid geomechanical background to detect formation boundaries [Vail, 1987]. However, the ground seismic data has two advantages: (1) relatively low vertical resolution (~ ±20 m) and (2) requires well logging data to convert recorded signal (arrival time of sonic waves) to the depth values. During drilling of boreholes, the information on formation tops can be inferred from visual analysis of drill cuttings that are recovered from wells. Although that approach is cost-effective and has a good potential to be automated in real-time mode [Ismailova et al., 2022; Tolstaya et al., 2023], there are several disadvantages to be considered. The process of recovering drill cuttings from in situ conditions to the surface is complicated by many aspects, including well inclination, distribution of cuttings sizes, properties of drilling fluids, etc. [Zakerian et al., 2018]. Thus, this approach exhibits high uncertainties in absolute values of depth intervals. Moreover, in many cases the alteration of formations cannot be detected visually and requires laboratory support. To overcome the problems mentioned above, petroleum engineers conduct well logging. The main advantages of well logging are (1) high vertical resolution (~ ± 0.5 m) and (2) coverage of large depth intervals during investigations.

Publisher

SPE

Reference19 articles.

1. A survey of methods for time series change point detection;Aminikhanghahi;Knowledge and information systems,2017

2. Bormann P. , AursandP., DilibF., DischingtonP., ManralS. 2020. 2020 FORCE Machine Learning Contest. https://github.com/bolgebrygg/Force-2020-Machine-Learning-competition

3. A machine-learning-based approach to assistive well-log correlation;Brazell;Petrophysics,2019

4. Selective review of offline change point detection methods;Truong;Signal Processing,2020

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3