Evaluation and optimization of different aspects related oil wells designing such as the wellbore stability evaluation, in-situ stresses prediction, drilling performance optimization, and hydraulic fracturing design are influenced by the static Young's Modulus which is determined based on the knowledge of the dynamic Young's Modulus (Edyn). Nowadays, Edyn is estimated from the shear and compressional velocities and bulk density, which in many cases may not be available. This study introduces an empirical equation developed based on an optimized artificial neural networks (ANN) model to predict the Edyn on real-time while the drilling process based on the drilling parameters of the rate of penetration, weight on bit, standpipe pressure, torque, drilling mud flowrate, and the drillpipe rotation speed. The ANN model was trained on 2054 data points from Well-A, then the empirical equation was extracted from this optimized model. This equation was then tested on 871 data points from Well-B and validated on 2912 data points from Well-C. The outcomes of this study showed that, the Edyn was predicted for the training data with an average absolute percentage error (AAPE) of 3.09%, using the optimized ANN model. The Edyn for the testing and validation data was predicted with AAPE's of 3.38% and 3.73%, respectively.
The elasticity of the rock determines its ability to recover from the deformation caused by subjecting the rock to external forces, the relationship between the applied external forces and the deformation is determined by the rock elastic properties such as the Young's modulus (Fjaer et al., 2008). Rock elastic properties are significantly influencing different aspects related oil wells designing such as the wellbore stability evaluation, in-situ stresses prediction, drilling performance optimization, and hydraulic fracturing design (Nes et al., 2005; Hammah et al., 2006). These elastic properties could be evaluated in the laboratory from compressional tests (static) or calculated from shear and compressional wave velocities (dynamic) evaluated using well logs (Barree et al., 2009).