CN-122022587-A - Integrated collaborative detection evaluation method based on steel bridge deck pavement

CN122022587ACN 122022587 ACN122022587 ACN 122022587ACN-122022587-A

Abstract

The invention relates to an integrated collaborative detection evaluation method based on steel bridge deck pavement, which comprises the steps of collecting images, temperature, rebound, ultrasound and deflection, forming a multi-source fusion data pool through space-time alignment convergence, carrying out layered analysis on the data pool, extracting characteristic parameters such as appearance, interlayer, pavement layer interior, steel plate and integral rigidity, constructing a multi-source characteristic database, carrying out fusion calculation on the multi-source characteristic based on an information entropy credibility weighting mechanism, generating a weighted comprehensive damage index, forming a comprehensive damage index spatial distribution map through a spatial interpolation technology, carrying out disease collaborative diagnosis and structural performance grading based on the distribution map, and backtracking to inquire the characteristic database, thereby overcoming the defect of fracture of traditional subitem detection data and evaluation subjective, realizing integrated fusion and intelligent collaborative analysis of multi-source data, and improving the precise identification and integral evaluation capability of hidden and composite diseases of a steel bridge deck pavement system.

Inventors

YIN YUAN
ZHANG HUI
Zhu Aizhou
LI DI
LIU WANKANG
LIN KANG
LUO CHAO
ZHU WEN

Assignees

湖北省路桥集团有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. The integrated collaborative detection evaluation method based on the steel bridge pavement is characterized by comprising the following steps of: S1, carrying out multi-sensor synchronous data acquisition on a steel bridge deck pavement system, carrying out synchronous data acquisition through a plurality of sensors integrated on a detection vehicle, carrying out space-time alignment and convergence on five types of acquired sensing data of images, temperature, rebound, ultrasound and deflection, and generating a space-time unified multi-source fusion data pool; S2, carrying out layered characteristic analysis and structural recombination on the multi-source fusion data pool, respectively extracting and constructing characteristic parameter sets of five dimensions of appearance, interlayer, paving layer interior, steel plate and overall rigidity, and generating a multi-source characteristic database facing bridge deck space position index; S3, carrying out multi-source feature fusion based on information entropy reliability weighting on the multi-source feature database, carrying out reliability evaluation and weighting calculation on multi-source feature data at each space position based on an information entropy technology, and carrying out fusion calculation on the weighted multi-source feature data to generate a weighted comprehensive damage index; S4, carrying out spatial distribution visualization processing on the weighted comprehensive damage indexes, and converting discrete index values into continuous spatial distribution representation based on a spatial interpolation technology to form a comprehensive damage index spatial distribution map covering a full bridge; S5, positioning a damage abnormal region based on the comprehensive damage index spatial distribution diagram, backtracking and inquiring detailed features in the multi-source feature database to judge the cause of the disease, and grading structural performance based on global statistical features of the comprehensive damage index spatial distribution diagram to generate a quality assessment report, wherein the quality assessment report comprises damage positions, disease types, severity and maintenance decisions.
2. The method according to claim 1, wherein S1 comprises: S11, performing multi-sensor hardware integration and synchronous control processing on a steel bridge deck pavement system, integrating a high-speed line scanning camera, a thermal infrared imager, a servo rebound instrument, an ultrasonic phased array probe and a laser Doppler vibration meter on a detection vehicle platform, and establishing a uniform time reference and space coordinate system through high-precision time service and a positioning system to generate a synchronous control signal; s12, synchronously acquiring and processing multi-source data based on the synchronous control signals, and controlling each sensor to acquire pavement layer surface images, temperature field distribution, punctiform rebound intensity, steel plate welding line ultrasonic signals and bridge deck dynamic deflection response according to preset frequency in the running process of the detection vehicle to generate an original data stream with space-time labels; And S13, carrying out space-time registration and fusion processing on the original data stream with the space-time label, mapping the sensor data to a unified bridge floor coordinate system through a coordinate transformation model, and carrying out data alignment and interpolation by utilizing a timestamp to generate a space-time unified multi-source fusion data pool.
3. The method according to claim 1, wherein S2 comprises: S21, performing apparent disease feature extraction processing on the image data in the multi-source fusion data pool, identifying cracks, ruts, pits and repair areas by adopting an edge detection and morphology segmentation algorithm, calculating geometric parameters and spatial distribution features of each disease, and generating an apparent disease feature set; S22, carrying out interlayer bonding state feature extraction processing on temperature field data in the multi-source fusion data pool, identifying a suspected region through temperature gradient analysis and abnormal region detection, extracting temperature difference, hot spot area and contour features, and generating an interlayer state feature set; s23, extracting the internal quality characteristics of the paving layer from the rebound intensity data in the multi-source fusion data pool, standardizing an actual measurement value according to a temperature-rebound correction model, constructing a rebound intensity distribution map through spatial interpolation, extracting statistical characteristics, and generating an internal quality characteristic set of the paving layer; s24, carrying out steel plate defect feature extraction processing on ultrasonic signal data in the multi-source fusion data pool, reconstructing a weld joint internal defect image by adopting a full-focus imaging algorithm, extracting defect type, size, depth and spatial position features, and generating a steel plate defect feature set; s25, carrying out overall rigidity characteristic extraction processing on deflection data in the multisource fusion data pool, calculating equivalent bending rigidity through deflection basin parametric analysis, extracting rigidity distribution characteristics and variation coefficients, and generating an overall rigidity characteristic set; and S26, carrying out structural recombination treatment on the apparent disease feature set, the interlayer state feature set, the paving layer internal quality feature set, the steel plate defect feature set and the overall rigidity feature set, and carrying out association storage on various feature parameters according to a unified space grid index to generate a multi-source feature database facing the bridge deck space position index.
4. A method according to claim 3, wherein the expression of the temperature-rebound correction model is: Wherein, the Is the rebound value at the standard temperature, For the measured rebound value, k is the material temperature correction coefficient, In order to measure the ambient temperature at the time of measurement, Is the standard reference temperature.
5. The method according to claim 1, wherein S3 comprises: s31, performing space gridding treatment on the multi-source characteristic database, dividing a bridge deck into regular geographic grid cells, distributing corresponding multi-source characteristic data for each grid cell, and generating a gridding characteristic data set; S32, carrying out normalization processing on multi-source characteristic data in each grid unit in the grid characteristic data set, converting actual measurement values of various characteristics into normalized damage indexes in the range of 0 to 1 based on physical meanings and damage threshold values of the actual measurement values, and generating a normalized damage index matrix; S33, carrying out information entropy calculation processing on the multi-class feature data in each grid unit in the normalized damage index matrix, calculating the information entropy of each class of feature based on the distribution discrete degree of the feature values, and evaluating the credibility of the corresponding feature in the current unit according to the information entropy to generate a feature credibility evaluation result; s34, carrying out dynamic weight distribution processing based on the feature credibility evaluation result, endowing features with lower information entropy, namely higher consistency, with higher weight, ensuring the sum of various feature weights to be 1 through normalization processing, and generating a dynamic feature weight matrix; and S35, carrying out weighted fusion calculation processing on the normalized damage index matrix and the dynamic characteristic weight matrix, multiplying the normalized damage indexes of various characteristics with the corresponding dynamic weights in each grid unit, and then summing the multiplied normalized damage indexes to generate the weighted comprehensive damage index of each grid unit.
6. The method of claim 1, wherein the comprehensive damage index spatial distribution map is calculated by the formula: Wherein, the For any point to be solved in comprehensive damage index space distribution diagram The continuous damage index after interpolation at the position, For a known total number of grid cells for the weighted composite impairment index, For the weighted composite impairment index of the jth known grid cell, For the point to be solved To the first The euclidean distance of the center points of the grid cells is known, The local coefficient of variation for the jth known grid cell, In order to adjust the coefficient of variation, Presetting standard deviation of weighted comprehensive damage indexes of all grid cells in a neighborhood window by taking the jth grid cell as the center, The weighted composite impairment index is the average of all grid cells within the neighborhood window.
7. The method according to any one of claims 1-6, wherein S5 comprises: S51, carrying out abnormal region positioning processing on the comprehensive damage index spatial distribution map, identifying a high damage index region through a preset multi-level alarm threshold value, combining adjacent abnormal units into a continuous damage region by combining a spatial clustering algorithm, and generating a damage abnormal region set; S52, for each region in the damage abnormal region set, the detailed multi-source characteristic data corresponding to the multi-source characteristic data are recalled from the multi-source characteristic database to carry out multi-evidence collaborative analysis processing, and specific disease types and causes are judged according to the matching of the combination modes of various characteristic damage indexes and a preset disease mode library, so that a disease diagnosis result is generated; S53, carrying out overall statistical feature extraction processing on the comprehensive damage index spatial distribution map, calculating statistics such as an average value, a standard deviation, a high damage area occupation ratio and the like of the full bridge damage index, and generating overall performance statistical features of the bridge deck system; S54, carrying out structural performance grading treatment based on the overall performance statistical characteristics of the bridge deck system, and dividing the performance of the bridge deck system into a plurality of grades according to a preset grading standard to generate a performance grade assessment result; And S55, integrating the damage abnormal region set, the disease diagnosis result and the performance grade assessment result to generate a comprehensive report, integrating the damage cloud picture, the disease distribution, the performance grade and the targeted maintenance suggestion into a structural document, and generating a quality evaluation report comprising the damage position, the disease type, the severity and the maintenance decision.
8. Based on steel bridge deck integration collaborative detection evaluation system of mating formation, its characterized in that, the system includes: The multi-sensor data synchronous acquisition and space-time convergence module is used for carrying out multi-sensor synchronous data acquisition on the steel bridge deck pavement system, carrying out synchronous data acquisition through a plurality of sensors integrated on the detection vehicle, carrying out space-time alignment and convergence on five types of acquired sensing data of images, temperature, rebound, ultrasound and deflection, and generating a space-time unified multi-source fusion data pool; the multi-dimensional characteristic hierarchical analysis and structuring module is used for conducting hierarchical characteristic analysis and structuring recombination on the multi-source fusion data pool, respectively extracting and constructing characteristic parameter sets of five dimensions of appearance, interlayer, paving layer interior, steel plate and overall rigidity, and generating a multi-source characteristic database facing bridge deck space position index; The information entropy weighting multi-source feature fusion module is used for carrying out multi-source feature fusion based on information entropy reliability weighting on the multi-source feature database, carrying out reliability evaluation and weighting calculation on multi-source feature data at each space position based on an information entropy technology, and carrying out fusion calculation on the weighted multi-source feature data to generate a weighted comprehensive damage index; The damage index spatial distribution visualization module is used for carrying out spatial distribution visualization processing on the weighted comprehensive damage index, converting discrete index values into continuous spatial distribution representation based on a spatial interpolation technology, and forming a comprehensive damage index spatial distribution map covering a full bridge; The disease cause judgment and performance grading evaluation module is used for positioning a damage abnormal region based on the comprehensive damage index spatial distribution diagram, backtracking and inquiring detailed features in the multi-source feature database to judge the disease cause, and meanwhile, carrying out structural performance grading based on global statistical features of the comprehensive damage index spatial distribution diagram to generate a quality evaluation report, wherein the quality evaluation report comprises damage positions, damage types, severity and maintenance decisions.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the method of any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any of claims 1 to 7.

Description

Integrated collaborative detection evaluation method based on steel bridge deck pavement Technical Field The invention relates to the technical field of nondestructive testing and engineering structure health monitoring, in particular to a steel bridge deck pavement-based integrated collaborative detection evaluation method. Background The steel bridge deck pavement system is used as a key bearing and driving interface of a large-span bridge, and the structural health condition of the steel bridge deck pavement system is directly related to the operation safety and the service life of the bridge. The system is composed of an orthotropic steel bridge deck, a waterproof bonding layer, an asphalt or epoxy asphalt pavement layer and the like, and is easy to induce a series of interrelated diseases such as pavement layer cracking, rutting, interlayer void, steel plate fatigue cracking, integral rigidity attenuation and the like under the long-term coupling effect of complex vehicle load and environmental factors. Therefore, the system, the comprehensive detection and the scientific evaluation are performed on the steel bridge deck pavement system, and the system is a foundation for implementing accurate maintenance and guaranteeing the long-acting safety of the bridge. The traditional detection and evaluation thought often depends on single or separate detection of various diseases, and then comprehensive judgment of the overall situation is attempted. However, existing solutions to the term detection technique have inherent limitations. All types of detection are usually carried out independently, the acquired data are mutually split in time, space and format, and effective association is difficult to establish. The evaluation process is seriously dependent on subjective experience of engineers to splice and read the isolated information, so that the efficiency is low, the inherent causal relationship among apparent diseases, internal damage and structural performance degradation is difficult to accurately reveal, and effective early warning and accurate positioning of early and hidden composite damage are not possible. This "data islanding" phenomenon makes the final assessment result lacking in integrity and objectivity, and difficult to support efficient preventive maintenance decisions. Disclosure of Invention Based on the information, the invention aims to provide the steel bridge pavement integrated collaborative detection evaluation method based on the steel bridge pavement integrated collaborative detection, which can integrate multidimensional information and realize data depth fusion and intelligent collaborative analysis so as to comprehensively, accurately and efficiently evaluate the health condition of the steel bridge pavement system. The invention adopts the following scheme: In a first aspect, the invention provides a steel bridge deck pavement integrated collaborative detection evaluation method, which comprises the following steps: S1, carrying out multi-sensor synchronous data acquisition on a steel bridge deck pavement system, carrying out synchronous data acquisition through a plurality of sensors integrated on a detection vehicle, carrying out space-time alignment and convergence on five types of acquired sensing data of images, temperature, rebound, ultrasound and deflection, and generating a space-time unified multi-source fusion data pool; S2, carrying out layered characteristic analysis and structural recombination on the multi-source fusion data pool, respectively extracting and constructing characteristic parameter sets of five dimensions of appearance, interlayer, paving layer interior, steel plate and overall rigidity, and generating a multi-source characteristic database facing bridge deck spatial position index; S3, carrying out multi-source feature fusion based on information entropy reliability weighting on the multi-source feature database, carrying out reliability evaluation and weighting calculation on multi-source feature data at each space position based on an information entropy technology, and carrying out fusion calculation on the weighted multi-source feature data to generate a weighted comprehensive damage index; s4, carrying out spatial distribution visualization processing on the weighted comprehensive damage indexes, and converting discrete index values into continuous spatial distribution representation based on a spatial interpolation technology to form a comprehensive damage index spatial distribution map covering a full bridge; s5, positioning a damage abnormal region based on the comprehensive damage index spatial distribution diagram, backtracking and inquiring detailed features in the multi-source feature database to judge the disease cause, and grading structural performance based on global statistical features of the comprehensive damage index spatial distribution diagram to generate a quality assessment report, wherein the quality assessment report compr