E. Pratama, D.R. Febriansanu, A.J. Widiatama
In a conventional way, evaluating the impact of depositional environment on the reservoir performance requires a comprehensive subsurface study from geological modelling to numerical simulation which is a very expensive and time-consuming process. This study aims to establish a link between depositional environments and the performance of non-associated gas reservoirs using an integrated data mining workflow. Primary data for this study is from the public domain data. To mine the data, a customized R script was developed using optical character recognition, regular expression, and rule-based logic to extract subsurface data attributes from thousands of documents in unstructured formats. Having identified subsurface data types and attributes, the data structure was then created in order to develop the subsurface database in a structured query language (SQL) database relational format. All extracted contents were transformed, cleansed, and quality checked before loaded into the database. A business intelligent dashboard was then developed to visualize the knowledge discovery from exploratory data analysis (EDA) and machine learning (ML) analysis. By applying the data mining workflow proposed in this study would provide meaningful insights from raw data in a quick way and can be a prudent complementary tool for the conventional subsurface study. © EAGE Asia Pacific Virtual Geoscience Week 2021. All rights reserved.
AEM Energy Solutions Sdn Bhd; Institut Teknologi Sumatera
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock