A method and system for generating a subsurface rock surface based on a radial basis function interpolator

By selecting representative radial basis function center points and using GPU parallel acceleration technology, combined with geological constraints and gradient descent smoothing optimization, the problem of low efficiency in existing technologies is solved, generating an efficient and reliable underground rock layer model that meets the modeling needs under complex geological conditions.

CN122452146APending Publication Date: 2026-07-24BEIJING HKRSOFT TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING HKRSOFT TECH CO LTD
Filing Date
2026-05-06
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing radial basis implicit interpolation methods are inefficient when processing massive amounts of geological data, making it difficult to meet the needs of rapid modeling and to effectively construct underground rock strata under complex geological conditions.

Method used

By selecting representative radial basis function center points and combining GPU parallel acceleration technology, a radial basis implicit interpolation model is constructed. Geological constraints and gradient descent smoothing optimization are introduced to generate an efficient and reliable underground rock strata model.

Benefits of technology

It achieves efficient processing of large-scale geological data, and the generated geological model achieves a good balance between efficiency and accuracy, ensuring the geological rationality and spatial continuity of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122452146A_ABST
    Figure CN122452146A_ABST
Patent Text Reader

Abstract

The application provides an underground rock stratum surface generation method and system based on a radial basis implicit interpolator, and belongs to the technical field of geological modeling; the method comprises the following steps: generating radial basis function center points based on obtained discrete geological data; constructing a radial basis implicit interpolation model through GPU parallel acceleration; constructing boundary conditions and optimization targets of the radial basis implicit interpolation model according to preset geological constraints; adjusting parameters of the radial basis implicit interpolation model; extracting an isosurface representing a target rock stratum interface to obtain an initial rock stratum surface; and performing gradient descent smoothing optimization on the initial rock stratum surface to obtain an underground rock stratum surface. The application fundamentally avoids the kernel matrix expansion problem caused by using all data points in the traditional method by selecting representative radial basis center points from a large amount of discrete data. In combination with the GPU parallel acceleration technology, the calculation complexity and memory consumption are reduced, and efficient processing of large-scale geological data is realized.
Need to check novelty before this filing date? Find Prior Art