Relevance. Young modulus prediction from drilling data is highly relevant to enhancing the efficiency and safety of oil and gas operations, particularly in complex geological settings. This study addresses the pressing need for cost-effective, real-time geomechanical characterization methods, reducing dependence on costly traditional techniques. Aim. To develop a methodology employing machine learning and artificial intelligence to accurately predict the Young modulus using drilling parameter data from oil and gas wells at the CT oil field, offshore Vietnam. Objects. Encompass rock and soil formations encountered in oil and gas wells, focusing on drilling parameters such as weight on bit, torque, rotary speed, and rate of penetration. Methods. Advanced machine learning techniques, including deep neural networks and ensemble methods, to analyze and map the non-linear relationships between drilling data and Young modulus. Models are trained and validated using reference data from core testing and well logs, with data preprocessing applied to mitigate noise and enhance predictive accuracy. Results. By applying machine learning algorithms, the research team successfully developed models to directly predict Young modulus from drilling parameters measured in real-time during the drilling of wells at the CT oil field, offshore Vietnam. The model employing a backpropagation neural network demonstrated superior performance, achieving a correlation coefficient of up to 0.94 and an RMSE error of only 0.483 when subjected to a blind test on a new well within the study area.

For citation: Vu Hong Duong, Nguyen Tien Hung, Vu Thiet Thach, Nguyen Minh Hoa, Nguyen Xuan Duy. Young modulus prediction using drilling data at CT oil field, offshore Vietnam. Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering, 2026, vol. 337, no. 4, pp. 64-74.