Application of Support Vector Machine (SVM) Method in Reservoir Lithology Classification in Sichuan Basin

Reservoir evaluation is essential for the scientific and rational design of geological exploration plans, and reservoir evaluation relies on obtaining information about reservoir lithology. With the deep integration of geophysical exploration technology and artificial intelligence, the use of artificial intelligence technology provides great convenience for complex and diverse geophysical exploration sites. Based on the support vector machine (SVM) method in machine learning, this paper uses actual well logging data from a gas well in the Sichuan Basin to train the algorithm and classify reservoir lithology in oil and gas development. The application of the support vector machine method in reservoir lithology is less explored. This paper effectively trains the support vector machine model with a large amount of well logging data and focuses on the influence of the C value and gamma value in support vector machines on classification accuracy. By optimizing the C and gamma values in support vector machines, the classification accuracy was improved from 0.63 to 0.75, effectively improving the convergence speed and optimization accuracy of the Sichuan Basin reservoir lithology identification algorithm. The successful application of the support vector machine in the classification of reservoir lithology in the Sichuan Basin demonstrates the significant cost reduction and efficiency improvement significance of artificial intelligence technology in geophysical exploration.