Geophysical potential field data gridding and noise removing method

CN118941732AActive Publication Date: 2024-11-12NORTHEASTERN UNIV CHINA

Patent Information

Application Number
CN202411069155.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-12
Estimated Expiration
2044-08-06

AI Technical Summary

Technical Problem

The prior art has problems with low accuracy in the gridding and noise removal of geophysical position field data, especially under the influence of irregular distribution of data and noise, making it difficult to achieve high-precision data processing and interpretation.

Method used

The random model forward method is used to generate geophysical position field data, add noise to generate noise-containing data sets, and select networks through diffusion model and noise level for data gridding and noise removal. The scattered data is redefined using the Euclidean distance, combined with diffusion Model and noise level networks enable accurate reconstruction and denoising of data.

Benefits of technology

High-precision gridding and noise removal of geophysical position field data is achieved, the MSE value is reduced to 0.1457, the accuracy and robustness of data processing are improved, and it is suitable for gridding and denoising of measured data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941732A_ABST
    Figure CN118941732A_ABST
Patent Text Reader

Abstract

The invention provides a geophysical potential field data gridding and noise removing method, and relates to the technical field of gravity and magnetic exploration data processing. The method comprises the following steps: firstly, establishing a set of original data set comprising geophysical potential field data and a noisy geophysical potential field data set; inputting the specified original data set into the diffusion model for training to obtain the diffusion model; secondly, obtaining a denoising step number corresponding to the noise data set, and training a noise level selection network by the denoising step number and the noise-containing data set; then, according to the coordinates and the abnormal values of the geophysical potential field data to be processed, determining the positions of the scatter data, and assigning the abnormities at the corresponding positions to generate the scatter data; and finally, gridding of scatter data is realized by using a diffusion model, and gridding data denoising is realized by using a noise selection network model in combination with the diffusion model. The method provided by the invention can improve the gridding precision of the potential field data, more accurately restore the distribution characteristics of the potential field, and remove noise interference.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Compensation direction filtering method of two-dimensional grid data

    CN103513288A

  • Satellite magnetic field data time-varying background field establishing method based on variational mode decomposition

    CN112327371A

  • Logging data three-dimensional space expansion modeling method, device and equipment and storage medium

    CN114742964A

  • Method and system for processing geophysical survey data

    US20060036367A1

Cited By

  • Boundary identification method based on intelligent gravitational magnetic field data denoising

    CN120929728A

  • A boundary identification method based on intelligent gravity and magnetic field data denoising

    CN120929728B