Lithofacies identification is a crucial work in reservoir characterization and modeling. The vast inter-well area can be supplemented by facies identification of seismic data. However, the relationship between lithofacies and seismic information that is affected by many factors is complicated. Machine learning has received extensive attention in recent years, among which support vector machine (SVM) is a potential method for lithofacies classification. Lithofacies classification involves identifying various types of lithofacies and is generally a nonlinear problem, which needs to be solved by means of the kernel function. Multi-kernel learning SVM is one of the main tools for solving the nonlinear problem about multi-classification. However, it is very difficult to determine the kernel function and the parameters, which is restricted by human factors. Besides, its computational efficiency is low. A lithofacies classification method based on local deep multi-kernel learning support vector machine (LDMKL-SVM) that can consider low-dimensional global features and high-dimensional local features is developed. The method can automatically learn parameters of kernel function and SVM to build a relationship between lithofacies and seismic elastic information. The calculation speed will be expedited at no cost with respect to discriminant accuracy for multi-class lithofacies identification. Both the model data test results and the field data application results certify advantages of the method. This contribution offers an effective method for lithofacies recognition and reservoir prediction by using SVM.

In VTI media, the conventional inversion methods based on the existing approximation formulas are difficult to accurately estimate the anisotropic parameters of reservoirs, even more so for unconventional reservoirs with strong seismic anisotropy. Theoretically, the above problems can be solved by utilizing the exact reflection coefficients equations. However, their complicated expression increases the difficulty in calculating the Jacobian matrix when applying them to the Bayesian deterministic inversion. Therefore, the new reduced approximation equations starting from the exact equations are derived here by linearizing the slowness expressions. The relatively simple form and satisfactory calculation accuracy make the reduced equations easy to apply for inversion while ensuring the accuracy of the inversion results. In addition, the blockiness constraint, which follows the differentiable Laplace distribution, is added to the prior model to improve contrasts between layers. Then, the concept of GLI and an iterative reweighted least-squares algorithm is combined to solve the objective function. Lastly, we obtain the iterative solution expression of the elastic parameters and anisotropy parameters and achieve nonlinear AVA inversion based on the reduced equations. The test results of synthetic data and field data show that the proposed method can accurately obtain the VTI parameters from prestack AVA seismic data.

Graphitic carbon nitride (g-C₃N₄) is used as a sustainable heterogeneous photocatalyst, which can effectively catalyze photo-atom transfer radical polymerization (photo-ATRP) without any additives.

Seismic estimation of the fluid factor and shear modulus plays an important role in reservoir fluid identification and characterization. Various amplitude variation with offset inversion methods have been used to estimate these two parameters, which are generally based on approximate formulations of the Zoeppritz equations. However, the accuracy of these methods is limited because the forward modeling ability of approximate equations is incorrect under the conditions of strong impedance contrast and large incidence angles. Therefore, to improve the estimation accuracy, we have used the Zoeppritz equations to directly invert for the fluid factor and shear modulus. Based on the poroelasticity theory, we derive the Zoeppritz equations in a new form containing the fluid factor, shear modulus, density, and dry-rock velocity ratio squared. The objective function is then constructed using these equations in a Bayesian framework with the addition of a differentiable Laplace distribution blockiness constraint term to the prior model to enhance the fluid boundaries. Finally, the nonlinear objective function is solved by combining the Taylor expansion and the iterative reweighted least-squares algorithm. Numerical experiments indicate that the inversion accuracy of our method may heavily depend on the parameter of the dry-rock velocity ratio square that is assumed to be static. However, tests on synthetic and field data show that our method can estimate the fluid factor and shear modulus with satisfactory accuracy in the case of choosing a reasonable static value of this parameter. In addition, we determine that the accuracy of this method is higher than that of the linearized formulation.

The accurate identification of lithofacies is indispensable for reservoir parameter prediction. In recent years, the application of multivariate statistical methods has gained more and more attention in petroleum geology. In terms of the identification for lithofacies, the commonly used multivariate statistical methods include discriminant analysis and cluster analysis. Fisher and Bayesian discriminant analyses are two different discriminant analysis methods, which include intrinsic advantages and disadvantages. Given the discriminant efficiency of different methods, calculation cost, difficulty in the degree of determining the parameters, and the ability to analyze statistical characteristics of data, we put forward a new method combined with seismic information to classify reservoir lithologies and pore fluids. This method integrates the advantages of Fisher discrimination, the kernel function, and Bayesian discrimination. First, we analyze training data and search a projection direction. Then, data are transformed through Fisher transformation according to this direction and different kinds of facies can be distinguished more efficiently by exploiting transformed data than by using primitive data. Subsequently, using the kernel function estimates the conditional probability density function of the transformed variable. A classifier is constructed based on Bayesian theory. Then, the pending data are input to the classifier and the solution whose posteriori probability reaches the maximum is extracted as the predicted result at each grid node. An a posteriori probability distribution of predicted lithofacies can be acquired as well, from which interpreters can evaluate the uncertainty of the results. The ultimate goal of this study is to provide a novel and efficient lithofacies discriminant method. Tests on model and field data indicate that our method can obtain more accurate identification results with less uncertainty compared with conventional Fisher approaches and Bayesian methods.