Synthetic Geomechanical Logs for Marcellus Shale

Today, I read the paper ‘Synthetic Geomechanical Logs for Marcellus Shale’ and find about five more papers on ANN.

Summary
Conventional logs such as Gamma Ray and Density Porosity can generate synthetic geomechanical logs using Artificial Intelligence and Data Mining (AI&DM), Data-driven Models.
AI&DM is capable of developing data-driven models for generating rock geomechanical properties.
Methodology
a. Data Preparation
The database contains the well name, the depth, the well coordinates, the values for gamma ray, bulk density, sonic porosity, bulk modulus, shear modulus, young’s modulus, poisson’s ratio and total minimum horizontal stress for each well. Some includes geomechanical well logs.
b. Data-driven Model Development
Conventional Models

Geomechanical Models

c. Validation of Data-driven Models
Blind wells have been chosen from different location of Marcellus Shale asset.
d. Geomechanical Property Distribution (Maps and Volumes)
Sequential Gaussian Simulation (SGS) creates distribution.
Training technique of ANN is back-propagation technique.

Tomorrow, I will go on to read this paper and more paper about ANN.