Enhancing overpressure origin analysis in sedimentary basins using machine learning

This $381,428 federal Project Grant awarded by the National Science Foundation (NSF) under the STEM Education (CFDA 47.076) program supports research at Southern University Shreveport, a Historically Black College and University (HBCU), to enhance the understanding of overpressure in sedimentary basins using machine learning.

The project aims to minimize risks of uncontrolled fluid flow (blowouts) in oil and gas wells by improving the characterization of geologic reservoir conditions through advanced data analytics. It will investigate signatures of overpressure in well log and drilling data to develop predictive machine learning models. This interdisciplinary research provides opportunities for undergraduate students at Southern University Shreveport to engage in geoscience and data science activities. The award is intended to establish new research capabilities, improve teaching, and increase student participation in STEM fields at the institution. No sub-awards are planned under this grant, which has a performance period from August 2024 to July 2026.

Abstract

Non-technical abstract: the hbcu-up research initiation awards offer support to junior and mid-career faculty at historically black colleges and universities. These awards aim to assist faculty in establishing new research programs or revitalizing existing ones. The award is designed to enhance the research capabilities and effectiveness of the faculty member, improve research and teaching at their institution, and engage undergraduate students in research activities. This award to southern university at shreveport provides opportunity to undergraduate students to engage in an interdisciplinary scientific research related to GEO and data science. This project aims to minimize the risks of uncontrolled fluid flow (blowout) in oil and gas wells by improving the understanding of geologic aspects of the reservoir using machine learning (ML). Overpressure occurs in specific conditions that can be traced or predicted using various forms of data either before or during the drilling operation. Technical abstract: this research aims to utilize recent advancements in the field of data sciences and machine learning to enhance the understanding of overpressure in sedimentary basins. The objective of this research is to investigate the signatures of overpressure on the well log and drilling parameters that are relevant to overpressure on a broad scale that includes various geologic settings. This study employs a holistic approach, initially searching for overpressure signatures within sedimentary rocks and subsequently scrutinizing evidence to validate or refute the initial findings. Overpressure indicators in well log and drilling data within both the overpressured stratum and the surrounding layers will be used in training the model. The novelty of this research is incorporating data science and machine learning to identify patterns within a wide range of well logs to unlock some of the reservoir characterization complexities which were not solved by conventional methods. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.