Augmented Learning Parameter Advisor for Wellbore Domain Interpretations

Author:

Rekik Karim1,Bouyghf Abdelkabir1,Zened Olfa2,Kontsedal Tanya3

Affiliation:

1. SLB, Montpellier, FRA

2. SLB, Dubai, UAE

3. AkerBP ASA, Oslo, NO

Abstract

Abstract The Parameter Advisor introduces an AI-powered solution for automating the selection of optimal parameter values in wellbore data interpretation. The aim is to reduce effort and time required for accurate interpretations. The software leverages machine learning algorithms, a comprehensive knowledge base, and collaboration among experts to enhance the interpretation process. The overall approach includes data gathering, quality control, and validation. Relevant data is collected and stored in a cloud storage system. The software applies statistical techniques and unsupervised learning algorithms to ensure accuracy and identify patterns in the data. Once the database is established, the software provides recommendations for future analyses based on past interpretations and expert knowledge. The results of tests conducted in the GRONINGEN and CASABE fields showed 92% accuracy compared to manual interpretation. The execution time for a Shale Volume interpretation was reduced by 64%. Collaborative studies with AkerBP in the Valhall field demonstrated an interpretation time reduction of approximately 70%. This study presents a novel approach in the petroleum industry by automating parameter initiation using machine learning and cloud computing. It improves the speed, accuracy, and efficiency of wellbore data interpretation. The software's ability to recommend optimal parameter values based on previous interpretations and expert knowledge contributes to better decision-making. The findings emphasize the effectiveness of machine learning in automating interpretation tasks and enabling non-experts to interpret data accurately. In summary, the proposed software streamlines the wellbore data interpretation process, reduces errors, and saves time. It enhances collaboration among experts, captures expert knowledge, and improves decision-making. The solution adds valuable insights to the petroleum industry by showcasing the power of machine learning in interpretation tasks and demonstrating its potential for transforming the field.

Publisher

SPE

Reference11 articles.

1. Adeiza James Onumanyi . "AutoElbow : An Automatic Elbow Detection Method for Estimating the Number of Clusters in a Dataset". In : Applied Sciences12.15 (2022). issn : 2076-3417. doi : 10.3390/app12157515. url : https://www.mdpi.com/2076-3417/12/15/7515.

2. All of Linear Regression;Kuchibhotla,2019

3. GeoStat-Framework/PyKrige : v1.7.0. Version v1.7.0;Murphy,2022

4. Support Vector Method for Novelty Detection;Schölkopf,1999

5. Isolation forest;Liu,2008

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