shahoo maleki

Having information about Young’s modulus is extremely essential for characterization of the hydrocarbon reservoirs. This property can be conventionally determined by core sample data analysis in laboratory that is time-consuming, critically expensive and discontinuous. Therefore, many researchers have always been looking for suitable methods to estimate Young’s modulus with acceptable accuracy. The current research aims to create an advanced, precise model for estimating Young’s modulus by utilizing backpropagation neural network (BPNN), support vector regression (SVR), and gene expression programming (GEP) methods based on the conventional well logs data. Thus, after determination of dynamic Young’s modulus, some empirical correlations are proposed for estimation of static Young’s modulus. The results demonstrate that the Jambunathan equation is more appropriate than other empirical models. Finally, artificial intelligence (AI) techniques were run, and their results indicated that all techniques (BPNN (R=0.999), SVR (R=0.997) and GEP (R=0.996)) deliver highly accurate values of static Young’s modulus. Comparing these results shows that the BPNN technique is relatively more precise than other ones. Although, in this research, the GEP technique was not more accurate than BPNN and SVM techniques, it provides a new nonlinear equation that can be used for estimating Young’s modulus in other similar fields. As a new finding, it was found that a simultaneous combination of the Jambunathan equation and BPNN technique delivers highly accurate results. Hence, it can be applied to slim down the cost of exploratory operations for determination of the Young’s modulus of limestone rocks.

This paper uses support vector machine (SVM), back propagation neural network (BPNN), and Multivariate Regression Analysis (MLA) methods to predict the gold in the Dalli deposit situated in the central province of Iran. After analyzing the data, the dataset was prepared. Subsequently, through comprehensive statistical analyses, Au was chosen as the output element for modelling, while Cu, Al, Ca, Fe, Ti, and Zn were considered input parameters. Then, the dataset was divided into two groups: training and testing datasets. For this purpose, 70% of the datasets were randomly entered into the data process, while the remaining data were assigned for the testing stage. The correlation coefficients for SVM, BPNN, and MLA were 94%, 75%, and 68%, respectively. A comparison of these coefficients revealed that all used methods successfully predicted the actual grade of Au. However, the SVM was more reliable and accurate than other methods. Considering the sensitivity of the gold data and the small number in the exploratory database, the results of this research are used to prepare the main layer in the mineral prospectivity mapping (MPM) of gold in 2 and 3D.

Sand production during extraction and exploitation from oil wells is one of the most important problems in delaying the exploiting operations. Different methods are presented to predict sand production in the field the reservoir rocks of which are sandstone or carbonates. This methods are typically comprised of rock mechanics tests (such as uniaxial compressive strength and triaxial tests), and mineralogical analyses or using of well logs. Rock mechanics tests and mineralogical analyses, as well as other tests used to analyze oil and gas wells, are very time consuming and costly in comparison with conventional well logs. Therefore, in this paper, well logs are used for the prediction of sand production in the southwest of Iran. For this purpose, by having traditional petrophysical logs such as sonic logs, the uniaxial compressive strength and elastic modulus in rocks are first calculated, and then the values of pore pressure are computed; these values are also calibrated with the real data. Since in situ stresses at deep depths are very necessary for the prediction of sand production, by using values obtained for elastic modulus, uniaxial compressive strength, and pore pressure, the vertical and horizontal stresses were obtained. Finally, Mohr-Columb criteria and caliper log data were used to obtain the optimum safe mud window for the prediction of local zones that have the potential of sand productions.

Estimation of the metal value in the metallic deposits is one of the important factors to evaluate the deposits in exploration studies and mineral processing. Therefore، one accurate estimator is essential to obtain a fine insight into the accumulation of the ore body. There are geostatistical methods for the estimation of the concentration of iron which performance of these models is complexity of analysis. The support vector machine (SVM) is by far one of the most robust artificial intelligence techniques used successfully for predictions and estimations of deposits because of its ability to generalize. Keeping this is view، the aim of this article is to use the SVM and back propagation neural networks (BPNN) to estimate the concentration of the iron element in the Choghartdeposit، in Iran. Comparing the obtained results with those of the validation process demonstrates that the SVM method is faster than the BPNN method and is more precise for the estimation of the iron concentration in the Choghartmine. The results of this study show that artificial intelligence– based models can evaluate the iron concentration with an acceptable accuracy.