Data-driven method for inferring the safety status of tunnel structures

By establishing a crack description feature set and a data-driven prediction model for the remaining bearing capacity of the lining, the problems of long detection cycles and low accuracy in tunnel structure safety assessment were solved, and a more comprehensive and accurate tunnel structure safety assessment was achieved.

CN119849254BActive Publication Date: 2026-05-26SHIJIAZHUANG TIEDAO UNIV +1
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHIJIAZHUANG TIEDAO UNIV
Filing Date
2025-01-08
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing technologies for tunnel structural safety assessment are characterized by long detection cycles, high costs, and limited accuracy, making it difficult to accurately reflect the impact of cracks on the overall safety of the lining.

Method used

A crack description feature set was established, the remaining bearing capacity of the lining was calculated using an extended finite element model, and the safety level of the tunnel structure was calculated by training and evaluating a data-driven prediction model of the remaining bearing capacity of the lining and combining it with the surrounding rock pressure.

Benefits of technology

This improves the comprehensiveness and accuracy of tunnel structure safety assessment, reduces the influence of human factors, and makes the evaluation results more objective and realistic.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119849254B_ABST
    Figure CN119849254B_ABST
Patent Text Reader

Abstract

This disclosure provides a data-driven method for inferring the safety status of tunnel structures, applicable to the field of tunnel engineering safety assessment. The method includes: establishing a crack description feature set; calculating the remaining bearing capacity of the lining corresponding to different crack types in the crack description feature set using an extended finite element model, obtaining a set of remaining bearing capacity results for the lining; using the crack description feature set as training samples and the set of remaining bearing capacity results as annotations; training a prediction model for the remaining bearing capacity of the lining based on the training samples; inputting the crack description feature set corresponding to the remaining bearing capacity of the lining to be predicted into the trained prediction model to obtain the predicted remaining bearing capacity of the lining; obtaining the surrounding rock pressure by testing the tunnel support force, calculating the ratio of the surrounding rock pressure to the predicted remaining bearing capacity of the lining, and obtaining the lining safety level and the tunnel structure safety level. This improves the comprehensiveness and accuracy of the safety assessment and makes the evaluation results more objective and realistic.
Need to check novelty before this filing date? Find Prior Art