Fig. 5 | Scientific Reports

Fig. 5

From: Integrative rock physics and computer vision analysis of elastic properties and pore aspect ratios in Brazilian pre-salt carbonates

Fig. 5

The image illustrates the process of predicting S-wave velocity (\(V_S\)) using well log data from the Iracema Field, based on machine learning and multiple linear regression models. (a) input and output logs are used in the modeling process. The P-wave velocity (\(V_P\)), bulk density (RHOB), and neutron porosity (NPHI) logs were used as inputs after despiking (i.e., the removal of outliers). The \(V_S\) log is shown as the reference output, and (b) compares real \(V_S\) values with those predicted by two methods: Multilayer Perceptron (MLP) and Multiple Linear Regression (MLR).

Back to article page