Integrating Petrophysics and Machine Learning for Geomechanical Evaluation in Lawn Oilfield Niger Delta

Abstract

Due to the challenges of direct measurements and precise Geomechanical evaluation, traditional techniques and machine learning algorithms, like Deep Neural Network were used to create predictive models of vital geomechanical properties such as unconfined compressive strength (UCS) and pore pressure. The aim of this study is to integrate petrophysics and machine learning for Geomechanical evaluation in Lawn Oilfield, Niger Delta. The objectives are to, establish a petrophysical model using conventional well log data for Geomechanical evaluation in Lawn oilfield; develop an advanced machine learning algorithm for geomechanical evaluation in Lawn oilfield; and recommend actionable suggestions on enhancing Geomechanical evaluation at the oilfields in Niger Delta. The implemented methodology greatly enhanced the analysis of a wellbore stability, which allows the process of maximizing of safe mud weight windows in drilling. This straight off countermeasures the hazards of drilling, minimizes the non-productive time, and improves the safety of operations. The research findings indicate that the petrophysics-ML integration is a more efficient, less costly, and accurate model of the geomechanical analysis, and it has a high potential to be used throughout the Niger Delta basin. The study revealed a new workflow with a combination of traditional petrophysical analysis and machine learning (ML) in improving geomechanical assessment in the reservoirs of the Lawn Field, Niger Delta.