一种基于多源异构监测数据的加筋土挡墙风险预警方法
By discretizing and segmenting reinforced soil retaining walls and assessing data quality, freezing-thaw and infiltration events are extracted, event characteristics are determined, degradation increments are calculated, and the state is recursively predicted. This solves the problems of unified recording index and quality quantification of heterogeneous monitoring data, and realizes traceable hierarchical early warning and adaptive state prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JILIN JIANZHU UNIVERSITY
- Filing Date
- 2026-01-08
- Publication Date
- 2026-07-17
AI Technical Summary
In existing monitoring methods for reinforced soil retaining walls, the sampling rhythm and dimensions of heterogeneous monitoring data vary greatly, and there is a lack of unified record index and quality quantification. It is difficult to define the event window and express the stage/intensity characteristics of the freeze-thaw and infiltration processes, which makes it difficult to carry out traceable graded early warning.
By discretizing the reinforced soil retaining wall into segments, collecting raw data streams and calculating quality scores, extracting freeze-thaw and infiltration events, determining the start and end times of events, generating stage labels and strength features, extracting the response increments of displacement, reinforcement strain, and pore pressure, and recursively combining freeze-thaw strength features to obtain the cumulative degradation state of the wall segment, calculating the risk value based on the evidence strength and residual consistency index, and realizing the early warning level determination.
It achieves same-scale degradation characterization, adaptive state prediction, and outputs interpretable low false alarm warnings. By standardizing response increments and freezing intensity features, it accumulates the degradation state of wall segments using fractional recursion, obtains the predicted state vector and covariance matrix, accumulates evidence strength to map risk values, and realizes traceable hierarchical warnings.
Smart Images
Figure CN121884556B_ABST
Abstract
Citation Information
Patent Citations
Geopolymer pervious concrete permeability attenuation determination method
CN118425005A
Reinforced retaining wall risk early warning method and system based on multi-source heterogeneous monitoring data
CN120975542A