CN-122020077-A - Roadbed disease identification method and system based on cooperative monitoring of optical fibers and soil pressure boxes

CN122020077ACN 122020077 ACN122020077 ACN 122020077ACN-122020077-A

Abstract

The invention provides a roadbed disease identification method and a roadbed disease identification system based on cooperative monitoring of optical fibers and a soil pressure box, which belong to the technical field of road engineering structure health monitoring and intelligent diagnosis, and comprise the steps of cooperatively arranging optical fiber sensors and the soil pressure box in different soil layers of a roadbed to respectively acquire distributed strain data and local stress data of soil bodies of the roadbed; the method comprises the steps of obtaining data, establishing a correlation model among multisource monitoring data based on the obtained data, predicting the change form of the monitoring data by using distributed strain data and local stress data of a roadbed soil body, carrying out fusion analysis on measured data and predicted data, and identifying and judging roadbed diseases based on analysis results.

Inventors

DU CONG
Tian Weiyang
TIAN YUAN
WANG JIKAI
MENG FEI
CHENG ZHIHENG
WEI MINGZHAO
ZHANG ZIHAO
WU JIANQING

Assignees

山东大学

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (10)

1. The roadbed disease identification method based on the cooperative monitoring of the optical fiber and the soil pressure box is characterized by comprising the following steps: the method comprises the steps that optical fiber sensors and soil pressure boxes are cooperatively arranged in different soil layers of a roadbed, and distributed strain data and local stress data of soil bodies of the roadbed are respectively obtained; establishing a correlation model between the multi-source monitoring data based on the acquired data; the correlation model predicts the change form of the monitoring data by using the distributed strain data and the local stress data of the roadbed soil body; and carrying out fusion analysis on the measured data and the predicted data, and identifying and judging the subgrade diseases based on analysis results.
2. The method for identifying subgrade diseases based on cooperative monitoring of optical fibers and a soil pressure box, as set forth in claim 1, characterized in that optical fiber sensors and the soil pressure box are cooperatively arranged in different soil layers of the subgrade, wherein in each layer on which the optical fiber sensors are arranged, the optical fiber sensors are arranged in parallel along the direction of the subgrade for obtaining distributed strain data of the subgrade soil body, the soil pressure box is arranged at the soil layer position corresponding to the optical fiber sensors, and the soil pressure box is arranged at intervals with the optical fiber sensors for obtaining stress data of the subgrade soil body; And forming a multi-source roadbed monitoring data set based on the strain data and the stress data of each acquired soil layer at different times.
3. The roadbed disease identification method based on the collaborative monitoring of the optical fiber and the soil pressure box according to claim 1 is characterized by further comprising the steps of preprocessing the collected optical fiber strain data and stress data, including outlier rejection, noise filtering and data normalization; And then, according to sampling frequencies of the strain data and the stress data, performing time synchronization processing on the strain data and the stress data to form a multi-source monitoring data sequence with consistent space and time.
4. The method for identifying subgrade diseases based on collaborative monitoring of optical fibers and a soil pressure box according to claim 1, wherein a correlation model between multisource monitoring data is established based on the acquired data, the correlation model is specifically a stress-strain space-time mapping model based on machine learning, and the establishment process comprises the following steps: step 1, constructing a training sample set, namely taking soil pressure data and optical fiber strain data which are aligned in time and space in the same soil layer as input and output pairs; Selecting a modeling method, namely selecting one or more machine learning methods to model according to the characteristics of data; step 3, model training and optimization, namely dividing a sample set into a training set and a verification set; Optimizing model super-parameters by taking mean square error or average absolute error as a loss function; after training, the model parameters are saved to form a correlation model which can be used for real-time prediction.
5. The method for identifying the subgrade diseases based on the collaborative monitoring of the optical fiber and the soil pressure box, which is disclosed in claim 1, is characterized in that when the measured data and the predicted data are subjected to fusion analysis, a subgrade state characteristic parameter set is constructed, wherein the subgrade state characteristic parameter set at least comprises measured data characteristics, predicted data characteristics and deviation characteristics between the measured data and the predicted data and is used for representing abnormal change conditions of the mechanical state of a subgrade soil body.
6. The method for identifying subgrade diseases based on cooperative monitoring of optical fibers and a soil pressure box according to claim 5, which is characterized by further comprising the steps of inputting the subgrade state characteristic parameter set into a disease identification model, identifying and classifying the subgrade diseases, and outputting the types of the subgrade diseases and the corresponding soil layer positions thereof; the types of diseases include sedimentation, uneven deformation, shear failure, pipe leakage, and void subgrade diseases.
7. Roadbed disease recognition system based on optical fiber and soil pressure cell collaborative monitoring, characterized by includes: the data acquisition module is configured to acquire distributed strain data and local stress data of the roadbed soil body; the association model construction module is configured to establish an association model among the multi-source monitoring data based on the acquired data; the roadbed disease identification module is configured to predict the change form of the monitoring data by using the distributed strain data and the local stress data of the roadbed soil body by the association model; and carrying out fusion analysis on the measured data and the predicted data, and identifying and judging the subgrade diseases based on analysis results.
8. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 6.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any of the preceding claims 1-6 when the program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, performs the steps of the method of any of the preceding claims 1-6.

Description

Roadbed disease identification method and system based on cooperative monitoring of optical fibers and soil pressure boxes Technical Field The invention belongs to the technical field of road engineering structure health monitoring and intelligent diagnosis, and particularly relates to a roadbed disease identification method and system based on cooperative monitoring of optical fibers and a soil pressure box. Background The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. Along with the continuous increase of road traffic load and the extension of service life, the roadbed is easy to generate diseases such as sedimentation, uneven deformation, shearing damage, partial void and the like under the action of long-term load and environmental factors. The disease has the characteristics of strong concealment, complex development process and the like, and if the disease can not be identified in time, the damage of a road surface structure is easily caused, and the road safety and the service performance are affected. The existing roadbed disease monitoring method mainly comprises manual inspection, single-point sensor monitoring and an analysis method based on an empirical model. The manual inspection has the defects of strong subjectivity, poor real-time performance and the like, and the single type sensor is difficult to comprehensively reflect the complex mechanical state inside the roadbed. For example, the distributed optical fiber sensing technology can acquire continuous strain information along the road foundation layout direction, but is difficult to directly reflect the real stress state of the soil body, while the soil pressure box can measure local soil body stress change, but has discrete measuring points, and is difficult to characterize the integral deformation characteristics. The analysis method based on the experience model is characterized in that the statistical relationship or semi-physical semi-empirical relationship between disease development and key influence factors is established by utilizing historical monitoring data, environmental factors and expert experience, and the model is distorted when the data is missing, the monitoring error or the record is incomplete because the accuracy of the model is completely established on the accuracy, consistency and long-term of the historical data. In addition, in the prior art, the utilization mode of the multi-source monitoring data mostly adopts simple superposition or independent analysis, and the inherent association relation between different types of monitoring data cannot be effectively established. When partial sensing data is missing, abnormal or noise is large, the reliability of a monitoring result and the accuracy of identifying the diseases are obviously reduced, and the intelligent identifying requirement of the roadbed diseases under the complex working condition is difficult to meet. Disclosure of Invention In order to overcome the defects of the prior art, the invention provides the roadbed disease identification method and the roadbed disease identification system based on the cooperative monitoring of the optical fiber and the soil pressure box, which can cooperatively utilize the optical fiber sensing and the soil pressure monitoring data to establish a correlation model between multi-source data, and realize the technical scheme of roadbed disease identification on the basis of the correlation model so as to improve the accuracy of roadbed disease identification and the reliability of system operation. To achieve the above object, one or more embodiments of the present invention provide the following technical solutions: in a first aspect, a method for identifying subgrade damage based on collaborative monitoring of optical fibers and a soil pressure box is disclosed, comprising the following steps: the method comprises the steps that optical fiber sensors and soil pressure boxes are cooperatively arranged in different soil layers of a roadbed, and distributed strain data and local stress data of soil bodies of the roadbed are respectively obtained; establishing a correlation model between the multi-source monitoring data based on the acquired data; the correlation model predicts the change form of the monitoring data by using the distributed strain data and the local stress data of the roadbed soil body; and carrying out fusion analysis on the measured data and the predicted data, and identifying and judging the subgrade diseases based on analysis results. As a further technical scheme, optical fiber sensors and soil pressure boxes are cooperatively arranged in different soil layers of a roadbed, wherein in each layer of the optical fiber sensors, the optical fiber sensors are arranged in parallel along the direction of the roadbed and used for acquiring distributed strain data of the soil body of the roadbed, the soil pressure b