support vector regression

Estimation of compressional and shear wave velocities is very important in the oil and gas industry. Unlike compressional wave velocity, shear wave velocity is not measured in all wells of a field due to its higher costs. Therefore, using an alternative method that estimates the shear wave velocity at a lower cost and with acceptable accuracy is inevitable. In this study, to estimate the response variable in a well, the correlation of several logs in that well (i.e., acoustic logs, density, neutron porosity, resistivity, gamma ray, dolomite volume, quartz volume, and water saturation) with target log investigated. It was found that the compressional wave velocity, density, dolomite volume, and quartz volume logs are more correlated with shear wave velocity. Therefore, these logs were selected as input features for estimating shear wave velocity using different approaches. In the next step, among the various methods, The estimated values obtained from a method that has the best match with the actual shear wave velocity is introduced as the optimal model. Afterward, it is performed to estimate the shear wave velocity in other wells that do not have a shear wave velocity log. In this paper, multiple regression methods and machine learning algorithms (support vector regression, adaptive Neuro-fuzzy inference system, and deep artificial neural network) were applied to predict the shear wave velocity. In this study, data from seven wells were used. Due to the fact that only in well #7 shear wave velocity has been measured, and in six other wells this feature has not been recorded, this field data limitation has caused the data of well #7 to be divided into training, testing, and validation data. In multiple regression methods (linear and interaction models), support vector regression, and adaptive Neuro-fuzzy inference system, Randomly, 70% of the data has been used for training and 30% for testing, but in the artificial neural network method, Randomly, 70% of the data has been used for training, 15% for validation and 15% for network testing. For all methods, the root means square error and correlation between actual and estimated data are calculated. Linear model, interaction model, support vector regression, adaptive Neuro-fuzzy inference system, and deep artificial neural network have provided 91, 92, 89, 94, and 98% correlation in training data, and 88, 89, 86, 90 and 92% in testing data, respectively. Also, the RMSE for each of the mentioned methods is 125.59, 115.86, 148.23, 84.36, and 80.49 (m/s) in the training data and 139.77, 133.44, 166.03, 126.15, and 98.04 (m/s) in the testing data, respectively. Our results show that deep artificial neural network has provided a better solution than other methods. Hence, in this study deep artificial neural network has been proposed to estimate the shear wave velocity in other blind wells. Moreover, the Castagna empirical model was used to validate the obtained results from the deep artificial neural network in these wells, which show a good fit between the two models.

The background of 3D modeling of fluid inclusion data goes back to use of inverse distance weighting (IDW) method in the Caixiashan Pb and Zn deposit (Sun et al., 2011). This method in spite of having some advantages such as simplicity in basis is associated with disadvantages such as uncertainty in selection of weighting function and ignoring data distribution. Today, new methods have been proposed for estimation including the support vector machine method (Dutta et al., 2010). One of this method’s capabilities is in dealing with small data sets (Dutta, 2006; Zhang et al., 1998). In this study, fluid inclusion thermodynamic parameters have been estimated using support vector regression method. Predictive model of mineralization has been provided acording to 3D models resulted for fluid inclusion data and also assumption of proper thermodynamic conditions for chalcopyrite deposition in the Sungun porphyry copper deposit.

In this study, a total of 173 data sets of fluid inclusions were obtained from 59 locations. This dataset using genetic algorithm method divided into training and testing sets (80% and 20%, respectively). Modeling of fluid inclusion thermodynamic parameters has been done by support vector regression method. The SVR is based on the statistical learning theory and the structural risk minimization.

After preparing and determination of training and test datasets, radial basis kernel function (RBF) was selected in order to estimate and model the fluid inclusion thermodynamic parameters using the support vector regression method. Better functionality was the main reason of using this kernel. In the next step, parameters were needed to be carefully determined to obtain a model with high generalization ability. In this regard, the grid search method with cross validation was used to determine optimal values for the model parameters. Model was then trained using the training dataset and finally evaluated on the test dataset. Then fluid inclusion thermodynamic parameters for each block of deposit were estimated using support vector regression method. According to mineralogical and fluid inclusion studies in the Sungun porphyry copper deposit, it has been determined that chalcopyrite deposition is related to fluids with moderate to high salinity and temperatures of 300-400 °C. The predictive model was prepared based on these conditions and estimated thermodynamic Parameters in block model. In this model, each arbitrary block has been labeled on a scale of 1 to 4 (based on the favorable conditions for chalcopyrite deposition). These labels are possibility index for copper deposition. According to possibility index, proper zones have been determined in 3D model. In order to performance evaluation of support vector regression method, the predictive model was compared with 3D model of copper grade. The results of this comparison showed that prepared predictive 3D model has high consistent with copper grade block model.

In this study, 3D modeling of fluid inclusion data was performed to estimate the thermodynamic parameters affecting mineralization (homogenization and eutectic temperatures and salinity) using support vector regression method to determine potential mineralization points in the area. Using the 3D models, we found the homogenization and eutectic temperatures and fluids salinity (in different ranges of these factors) in the Sungun porphyry copper deposit. To evaluate the 3D modeling efficiency in advancing the exploration process of the porphyry deposits, the conformity between mineralization and thermodynamic variations of the fluid inclusions was investigated and, based on it; a tool called “Predictive Model” was presented for the evaluation of the occurrence of mineralization in different parts of the region. A comparison of the SVR-based predictive model and the copper grade block model shows acceptable conformity in low, medium, and high-grade regions.