Well logging productivity prediction system and method based on improved attention mechanism

By improving the attention mechanism of the well logging productivity prediction system, and utilizing a hybrid model of multi-scale residual network and Transformer-CNN, the automation and accuracy problems of traditional well logging productivity prediction are solved, and efficient and accurate productivity assessment is achieved.

CN122114263APending Publication Date: 2026-05-29YANGTZE UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
YANGTZE UNIVERSITY
Filing Date
2026-01-22
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Traditional well logging productivity prediction methods rely on manual operation, which makes it difficult to handle multi-source data, inconsistent depths and missing data, insufficient feature extraction, and unstable prediction accuracy, especially under complex geological conditions.

Method used

The well logging productivity prediction system employs an improved attention mechanism. By integrating a multi-scale residual network and a Transformer-CNN hybrid coding model, combined with an automated data processing workflow, it achieves deep feature extraction and high-precision prediction, including modules for semi-structured data transformation, depth alignment and resampling, training data construction, and unknown well production prediction.

Benefits of technology

It enables automated parsing and structured transformation of well logging data, provides a unified data foundation, improves preprocessing efficiency, integrates local and long-range feature extraction capabilities, enhances prediction accuracy and stability, and supports productivity assessment under complex geological conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122114263A_ABST
    Figure CN122114263A_ABST
Patent Text Reader

Abstract

The application provides a well logging productivity prediction system and method based on an improved attention mechanism, relates to the field of oil and gas exploration and development, and comprises a semi-structured data conversion module, a depth alignment and resampling module, a training data construction module, a hybrid model construction and training module and an unknown well yield prediction module. The system automatically analyzes WIS logging files, extracts multiple logging curves, realizes depth unification, missing value filling and interactive feature construction; based on a parallel coding structure of a one-dimensional convolution network and a multi-head attention mechanism, the local features and long-range dependencies of the logging curves are jointly modeled by combining a multi-scale residual network, so that high-precision prediction of the oil test yield is realized. The system can generate a depth-yield curve according to the prediction result, divide the yield into five grades, and realize automatic evaluation of the productivity of a new well.
Need to check novelty before this filing date? Find Prior Art