Data Mining and Predictive Analytics Transforms Data to Barrels

Author:

Shirzadi S..1,Ziegel E..1,Bailey R..1

Affiliation:

1. BP

Abstract

Abstract BP's Field of the Future Technology Flagship is developing data-driven technologies that complement existing capabilities for reservoir management and operations. Widespread adoption of BP's proprietary wells and equipment real time surveillance and monitoring applications provides efficient workflows that deliver real time data to decision-makers. Data Analytics applications are transforming this data into information that will improve the management of operational risk, increase production and maximize both recovery and workforce efficiency. This paper describes our progress made in the areas of operational risk and increasing production and outlining our future activities. So far we have created data-driven corrosion assessment tools and have developed new technology for virtual flow meters. Corrosion assessment is now able to evaluate the efficiency of our pipeline inspection programs. Following successful proof of concept and field trials, we will package this capability and deploy it globally as a standard workflow as soon as possible. Data-driven virtual flow meters are being implemented to provide a 3-phase metering capability. Better well and zonal flow allocation will improve reservoir management. Other soft sensors are being developed to create well diagnostic and prognostic capabilities for BP's real-time surveillance toolkit. This helps assets identify well work and optimization opportunities. Future activities include new predictive technologies to increase recovery and reduce well-integrity risks. We are developing an innovative approach to diagnose and optimize waterflood performance designated as Top-Down Waterflood Diagnostics and Optimization. This novel approach combines BP's proprietary production event detection and association technology with visualization and parametric models that can quantify the subsurface connections between injectors and the producers. Successful trials have demonstrated how this can powerfully complement the conventional waterflood management and optimization workflows. In conjunction with other activities, the benefit to operations is derived from improving sweep efficiency so increasing recovery. Use of data mining and predictive analytics for business intelligence has created significant value in finance, medicine, power generation, and other industries. BP now has an established, active program to bring the business intelligence approach into its E&P operations and particularly to its reservoir management.

Publisher

SPE

Cited by 3 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

1. Machine learning regressors and their metrics to predict synthetic sonic and mechanical properties;Interpretation;2019-08-01

2. Use of Data Analytics to Optimize Hydraulic Fracture Locations Along Borehole;Petrophysics – The SPWLA Journal of Formation Evaluation and Reservoir Description;2018

3. Radial Basis Function Network to Predict Gas Flow Rate in Multiphase Flow;Proceedings of the 9th International Conference on Machine Learning and Computing;2017-02-24

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