عنوان مقاله [English]
Estimation of compressional and shear wave velocities are very important in the oil and gas industry. Unlike compressional wave velocity, shear wave velocity is not measured in all wells in a field due to its higher costs. Therefore, in the oil and gas industry, the use of a method that estimates the shear wave velocity at a lower cost and with unacceptable accuracy is inevitable. In this study, to estimate the shear wave velocity in a well, the correlation of other logs in that well (i.e. acoustic logs, density, neutron porosity, resistivity, gamma ray, dolomite volume, quartz volume, and water saturation) with shear wave velocity investigated. And it was found that the logs of compressional wave velocity, density, dolomite volume and quartz volume are more correlated with shear wave velocity and these logs were selected as inputs for estimating shear wave velocity using different methods. In the next step, among the various methods, the method that best matches the actual shear wave velocity data is selected as the optimal method. This method is then used to estimate the shear wave velocity in other wells that do not have a shear wave velocity log. In this paper, multiple regression method and machine learning algorithms (support vector regression, adaptive Nero-fuzzy inference system and deep artificial neural network) were used to estimate 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, so this field data limitation has caused the data of well #7 to be divided into training, testing and validation data. In multiple variable regression methods (linear and interaction models), support vector regression and adaptive Nero-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 mean square error and correlation between actual and estimated data are calculated. Linear model, interaction model, support vector regression, adaptive Nero-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 in the training data and 139.77, 133.44, 166.03, 126.15 and 98.04 for the testing data, respectively. Our results show that deep artificial neural network has provided a better answer than other methods. Therefore, the method proposed in this study (deep artificial neural network) was used to estimate the shear wave velocity in other wells. To validate the results obtained from the deep artificial neural network in wells without shear wave velocity, the Castagna experimental model was used, which shows a good fit between the two models.