Frac-Driven Facies: Integrating Mechanical Stratigraphy and Unsupervised Machine Learning to Unlock Geomechanical Zonation in Complex Reservoirs

Read full text Summarise with VIRAT
Authors: IJBST Faculty of Science
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The depositional signatures that are employed to interpret lithofacies using traditional well logs are inadequate in the reflection of the mechanical behavior of the reservoir rocks, which is essential in the contemporary application of subsurface engineering. This gap is discussed in this review, which suggests the Frac-Driven Facies classification scheme that is based on geomechanical stratigraphy calculated with the help of unsupervised machine learning methods based on the analysis of well-log data. Recent advancements in the last 10 years in cluster analysis, well-log interpretation and elastic property inversion are integrated to suggest a more efficient approach in zonation of reservoirs. The suggested framework would be especially relevant to heterogeneous unconventional systems and carbonates where mechanical variability is a decisive factor of unconventional reservoir performance. The discussion shows that using mechanical characteristics of the rock in the form of Youngs modulus, ratio of poison and brittle index in the classification of facies enhances predictative ability of a number of subsurface engineering tasks. They involve the optimization of hydraulic fracturing, the enhancement of wellbore stability during the drilling tasks, and determination of caprock integrity in CO 2 sequestration projects. The existing literature is critically reviewed to point out ongoing challenges especially the calibration of log derivatives of geomechanically proxies in relation to laboratory measurements of rock-mechanics. Other issues are connected to transparency and reproducibility where proprietary algorithms happen in machine-learning processes. A combination of sedimentology, rock mechanics and data analytics can offer a powerful geomechanical model that can contribute to enhancing the development of forecasts and accuracy of underground characterization. In addition to hydrocarbon production, other uses of the approach include carbon storage, geothermal energy generation, and sustainable underground resources based on the global climate and energy transition objectives.

Fields of Study & Subjects

Hydraulic Fracturing and Reservoir Analysis Hydrocarbon exploration and reservoir analysis Rock Mechanics and Modeling

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
22 March 2026
Language
EN