Municipal bridge damage defect detection method and system

Abstract

The invention provides a municipal bridge damage defect detection method and a municipal bridge damage defect detection system, which are used for solving the problem that the quality and the accuracy of acquired data are difficult to guarantee due to the fact that vibration signals are easily interfered by environmental noise, such as vehicle load, wind load and the like. A bridge key part vibration signal data set is obtained, and a final damage area range is determined based on the vibration signal data set; whether the bridge is damaged or not is judged based on recognition of the vibration signals, so that the signal-to-noise ratio of monitoring data and the damage positioning precision are improved, a more accurate and reliable decision basis is provided for a bridge management department, and finally maintenance and reinforcement work of the bridge is effectively guided.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge damage defect detection, and in particular to a municipal bridge damage defect detection method and system. Background Art

[0002] Bridge structural health monitoring technology based on vibration signal analysis is an advanced damage and defect detection method. It deploys high-precision acceleration sensors at key bridge locations to collect real-time structural vibration response data. It then uses frequency domain analysis and modal parameter identification algorithms to monitor changes in the bridge's natural frequency and damping ratio, enabling early detection of structural stiffness degradation and localized damage.

[0003] However, in the actual bridge damage acceptance process, vibration signals are easily interfered with by environmental noise, such as vehicle loads and wind loads, making it difficult to ensure the quality and accuracy of the collected data. There is an urgent need for a new municipal bridge damage defect detection method to identify and determine whether a bridge is damaged based on vibration signals, thereby improving the signal-to-noise ratio of monitoring data and the accuracy of damage location, providing bridge management departments with more accurate and reliable decision-making basis, and ultimately effectively guiding bridge maintenance and reinforcement work. Summary of the Invention

[0004] The present invention provides a municipal bridge damage defect detection method and system to solve the problem that vibration signals are easily interfered with by environmental noise, such as vehicle load, wind load, etc., which makes it difficult to ensure the quality and accuracy of collected data.

[0005] The present invention provides a municipal bridge damage defect detection method, comprising: obtaining a vibration signal dataset of key parts of the bridge, and performing multi-scale decomposition and reconstruction on the vibration signal, filtering out environmental noise, and generating a denoised vibration signal set;

[0006] Extract the frequency components and amplitude characteristics of the denoised vibration signal set to determine the frequency distribution characteristics of the bridge;

[0007] Based on the frequency distribution characteristics, the natural frequency change and damping ratio data are analyzed to build a dynamic characteristic model, which is used to determine the distribution of modal parameters;

[0008] When the modal parameter distribution deviates from the preset threshold range, the natural frequency deviation value and the degree of damping ratio abnormality are calculated by comparing historical data to generate evidence of structural stiffness degradation;

[0009] Quantitatively evaluate the deviation value and abnormality degree of structural stiffness degradation evidence to generate damage probability indicators;

[0010] Based on the damage probability index, combined with the spatial distribution characteristics of the vibration signal and the bridge structure model, the preliminary coordinates of the damage location are determined;

[0011] The signal-to-noise ratio value of the signal in the area corresponding to the damage location coordinates is extracted, and a secondary analysis of the local vibration signal is performed to determine the final damage area range.

[0012] Preferably, vibration signals from key parts of the bridge are collected, and the timestamps of the vibration signals are calibrated to obtain original vibration records containing time series and amplitude information;

[0013] Denoising and filtering are performed on the original vibration records to decompose the original vibration records into multi-scale sub-signals to obtain pre-processed vibration data;

[0014] Extract frequency, amplitude and phase features from the preprocessed vibration data, generate feature vectors of the vibration data, and obtain a feature set for analysis;

[0015] The feature set is classified and processed, and the key parameters in the feature set are compared using a preset threshold. If the key parameters exceed the preset threshold, it is determined that the bridge vibration is abnormal, and the vibration status results of the key parts of the bridge are obtained.

[0016] Preferably, the vibration signal data set is subjected to multi-scale decomposition and reconstruction processing, and the influence of environmental noise is removed by filtering operation to obtain a denoised vibration signal set;

[0017] According to the denoised vibration signal set, the time series characteristics of the signal are extracted, the periodic changes of the signal are decomposed and the periodic distribution characteristics of the signal are obtained;

[0018] The stability of the vibration signal is analyzed by the periodic distribution characteristics of the signal. If the stability parameter deviates from the preset threshold, the signal is further segmented to determine the range of abnormal fluctuations.

[0019] Obtain the local amplitude change characteristics of the signal within the range of abnormal fluctuations, transform the local signal based on frequency domain analysis, and obtain the frequency distribution result of the local signal;

[0020] Based on the frequency distribution results of the local signal, the frequency anomaly of the vibration signal in a specific interval is determined. If the frequency distribution exceeds the preset threshold range, the interval is marked as a potential abnormal segment, and a set of marked signal segments is obtained;

[0021] Extract the phase feature information of potential abnormal segments from the marked signal segment set, perform difference analysis between the abnormal segments and normal segments, and determine the phase offset of the vibration signal;

[0022] By using the time series characteristics and frequency distribution results, a comprehensive vibration feature vector is constructed for the phase shift of the vibration signal to obtain a feature data set for subsequent analysis.

[0023] Preferably, the denoised vibration signal set is preprocessed to obtain pure vibration data;

[0024] Based on the clean vibration data, the frequency domain analysis method is implemented, the Fourier transform is used to decompose the signal and extract the frequency components and amplitude characteristics;

[0025] Analyze the distribution rules of the extracted frequency components and amplitude characteristics to determine the frequency distribution characteristics of the bridge structure;

[0026] If there are outliers in the frequency distribution characteristics, they are compared with preset thresholds to determine whether there are potential structural problems. Based on the judgment results, the specific location information of the abnormal frequency components is obtained, and the source of the abnormality is analyzed in combination with the bridge structure characteristics.

[0027] By analyzing the source of anomalies and comparing structural characteristics with frequency distribution, we can determine whether the anomalies affect the overall stability, thereby obtaining stability analysis results.

[0028] The stability analysis results are combined with the data extracted from the frequency components and amplitude features to generate the health status assessment data of the bridge structure.

[0029] Preferably, frequency distribution characteristic data of the vibration signal is collected, and the vibration signal is preprocessed by using a time domain and frequency domain conversion tool to obtain a preliminary set of dynamic characteristic parameters;

[0030] Analyze the natural frequency change and damping ratio data of the preliminary dynamic characteristic parameter set and compare them with the preset threshold range. If the parameter value exceeds the threshold range, mark the abnormal data and determine the parameter subset that needs calibration;

[0031] Abnormal data is obtained from the parameter subset that needs to be calibrated, and the abnormal values ​​are corrected. By comparing with historical bridge dynamic characteristic data, it is determined whether the corrected parameter values ​​are within the expected range, and a calibrated dynamic characteristic parameter set is obtained;

[0032] Based on the calibrated dynamic characteristic parameter set, a dynamic characteristic distribution diagram of the bridge is constructed. By comparing the distribution of different modal parameters, the correlation between the modal parameters is determined, and the evaluation results of the overall dynamic characteristics of the bridge are obtained.

[0033] Preferably, based on the bridge vibration signal, modal parameters of current data are obtained, and the modal parameters are preliminarily screened. If the modal parameters exceed a preset threshold range, they are marked as abnormal parameters, and an abnormal parameter set is obtained;

[0034] Extract the modal parameter distribution range from the historical data of the abnormal parameter set, calculate the deviation amplitude of the natural frequency between the current data and the historical data, and obtain the frequency deviation data set;

[0035] Calculate the abnormality of the damping ratio of the frequency deviation data set. If the abnormality exceeds a preset threshold, mark it as a high-risk parameter and obtain a high-risk parameter subset.

[0036] Obtain relevant data on structural stiffness from the high-risk parameter subset, and use a pre-established analytical model to quantify the basis for structural stiffness change to obtain a set of stiffness change indicators;

[0037] Generate a distribution map of structural stiffness changes using a set of stiffness change indices, determine whether there are significant abnormal areas in the distribution map, and obtain an abnormal area identification set;

[0038] Analyze the correlation between the natural frequency and damping ratio of the abnormal area identification set. If the correlation is lower than the preset threshold range, it is marked as a potential risk point and the potential risk point set is determined;

[0039] Compare historical data of the potential risk point set to further verify the matching degree between the structural stiffness change basis and the current data, determine whether the matching degree is within the expected range, and obtain the final verification result set.

[0040] Preferably, monitoring data of structural stiffness is obtained, and preliminary screening is performed for degradation evidence. By comparing the changing trends of historical data and current data, a significance judgment result of the degradation evidence is obtained;

[0041] Based on the significance judgment results of the degradation evidence, the calculation method of the deviation value and the abnormality degree is used, combined with the preset damage threshold standard, to determine the correlation assessment value between the deviation value and the abnormality degree;

[0042] By evaluating the correlation between the deviation value and the abnormality degree, the damage possibility is classified and predicted to obtain preliminary indicator data of the damage possibility;

[0043] Based on the preliminary indicator data of the possibility of injury and the distribution characteristics of potential injuries, the severity is graded to obtain a quantitative level of severity;

[0044] By combining the quantitative severity level with the standard assessment rule base, if the quantitative level exceeds the preset threshold range, the potential damage will be marked and the key monitoring area will be determined;

[0045] Based on the marking results of key monitoring areas, the structural stiffness data and updated information of degradation evidence in the relevant areas are obtained. Through data fusion processing, the dynamic changes of damage trends are judged;

[0046] Time series analysis is used to make long-term predictions of the dynamic changes in damage trends and obtain updated data on the evolution path and severity of potential damage.

[0047] Preferably, vibration signal data of the bridge structure is acquired to obtain an original vibration signal, and the original vibration signal is subjected to denoising and filtering processing to obtain a processed first vibration signal data set;

[0048] Analyzing the spatial distribution characteristics of the processed first vibration signal data set, and determining the spatial distribution regularity characteristics of the vibration signal using a pre-established distribution characteristic analysis model;

[0049] Based on the spatial distribution characteristics of the vibration signal, a finite element analysis is performed on the pre-built bridge structure model to obtain the stress distribution data corresponding to the distribution characteristics in the structural model;

[0050] If there are abnormal areas in the stress distribution data, the preset stress threshold range is compared to determine whether the abnormal area is a potential damage location, and a preliminary set of damage locations is obtained;

[0051] further calibrating the spatial coordinates of each potential damage location in the preliminary damage location set to determine a calibrated second damage location set;

[0052] By performing multi-dimensional verification on the second damage location set and combining the spatial distribution characteristics of the vibration signal with the geometric parameters of the bridge structure model, the final damage location coordinate data is obtained;

[0053] If the final damage location coordinate data does not match the preset damage possibility index range, the first vibration signal data set is traced back to adjust the signal processing parameters and re-analyze to obtain more accurate damage location coordinates.

[0054] Preferably,

[0055] By obtaining relevant information about the damage location from the preliminary coordinates, the data of the corresponding area is preliminarily extracted to obtain the initial distribution of the signal-to-noise ratio;

[0056] Based on the initial distribution of the signal-to-noise ratio, the local vibration signal is decomposed into features, and the abnormal points are screened using a preset threshold to determine the key abnormal areas in the vibration signal.

[0057] If the signal strength in a critical abnormal area exceeds a preset threshold, the vibration signal in that area is deeply processed to obtain more detailed signal data and determine the potential damage range;

[0058] By analyzing the frequency domain of the fine signal data, the characteristic parameters of the local vibration are extracted and combined with the preliminary coordinates of the damage location to determine the boundary range of the final area;

[0059] If there is a significant deviation between the boundary range of the final area and the preliminary coordinates, the signal data is corrected to obtain the corrected distribution characteristics and determine the accuracy of the damage range;

[0060] According to the corrected distribution characteristics, the support vector machine algorithm is used to classify and confirm the damage range to obtain the final damage area division result;

[0061] By comparing and analyzing the final damage area division results and combining them with the historical signal data of the corresponding area, the dynamic change trend of the damage location is obtained and the key areas for long-term monitoring are determined.

[0062] The present invention provides a municipal bridge damage defect detection system, which uses the municipal bridge damage defect detection method to perform defect detection. The system includes:

[0063] The data acquisition module is used to obtain the vibration signal data set of the key parts of the bridge. The data set is collected in real time by the acceleration sensor and contains the original vibration response information;

[0064] The signal denoising module is used to perform multi-scale decomposition and reconstruction on the vibration signal dataset using a denoising method based on wavelet transform, filter out environmental noise interference, and obtain a denoised vibration signal set;

[0065] The frequency domain analysis module is used to perform Fourier transform processing based on the denoised vibration signal set using the frequency domain analysis method to extract the frequency components and amplitude characteristics and determine the frequency distribution characteristics of the bridge structure;

[0066] The modal identification module is used to analyze the natural frequency change and damping ratio data based on the frequency distribution characteristics using the modal parameter identification algorithm, build a bridge dynamic characteristics model, and determine the modal parameter distribution;

[0067] The stiffness assessment module is used to compare historical data with current data to calculate the natural frequency deviation and the degree of damping ratio anomaly if the modal parameter distribution deviates from the preset threshold range, thereby obtaining evidence of structural stiffness degradation;

[0068] The damage quantification module is used to quantify the deviation value and abnormality degree based on the evidence of structural stiffness degradation using the damage threshold standard, generate a damage probability index, and determine the potential damage severity;

[0069] The positioning analysis module is used to analyze the spatial distribution characteristics of vibration signals based on damage probability indicators using a damage location accuracy improvement algorithm, and then obtain preliminary coordinates of the damage location based on the bridge structure model.

[0070] The area determination module is used to extract the signal-to-noise ratio value of the corresponding area signal based on the preliminary coordinates of the damage location, perform secondary analysis of the local vibration signal, and determine the final damage area range.

[0071] The working principle and beneficial effects of the present invention are as follows:

[0072] The present invention provides a municipal bridge damage defect detection method, comprising: obtaining a vibration signal dataset of key parts of the bridge, and performing multi-scale decomposition and reconstruction on the vibration signal, filtering out environmental noise, and generating a denoised vibration signal set;

[0073] Extract the frequency components and amplitude characteristics of the denoised vibration signal set to determine the frequency distribution characteristics of the bridge;

[0074] Based on the frequency distribution characteristics, the natural frequency change and damping ratio data are analyzed to build a dynamic characteristic model, which is used to determine the distribution of modal parameters;

[0075] When the modal parameter distribution deviates from the preset threshold range, the natural frequency deviation value and the degree of damping ratio abnormality are calculated by comparing historical data to generate evidence of structural stiffness degradation;

[0076] Quantitatively evaluate the deviation value and abnormality degree of structural stiffness degradation evidence to generate damage probability indicators;

[0077] Based on the damage probability index, combined with the spatial distribution characteristics of the vibration signal and the bridge structure model, the preliminary coordinates of the damage location are determined;

[0078] The signal-to-noise ratio value of the signal in the area corresponding to the damage location coordinates is extracted, and a secondary analysis of the local vibration signal is performed to determine the final damage area range.

[0079] This invention provides a method for detecting damage and defects in municipal bridges, addressing the problem of vibration signals being easily affected by environmental noise, such as vehicle loads and wind loads, which can hinder the quality and accuracy of collected data. This method acquires a vibration signal dataset from key bridge locations and, based on this dataset, determines the extent of the final damage area. This method identifies damage based on the vibration signal, thereby improving the signal-to-noise ratio of the monitoring data and the accuracy of damage location. This provides bridge management departments with more accurate and reliable decision-making, ultimately effectively guiding bridge maintenance and reinforcement efforts.

[0080] The present invention provides a municipal bridge damage defect detection system, which uses the municipal bridge damage defect detection method to perform defect detection. The system includes:

[0081] The data acquisition module is used to obtain the vibration signal data set of the key parts of the bridge. The data set is collected in real time by the acceleration sensor and contains the original vibration response information;

[0082] The signal denoising module is used to perform multi-scale decomposition and reconstruction on the vibration signal dataset using a denoising method based on wavelet transform, filter out environmental noise interference, and obtain a denoised vibration signal set;

[0083] The frequency domain analysis module is used to perform Fourier transform processing based on the denoised vibration signal set using the frequency domain analysis method to extract the frequency components and amplitude characteristics and determine the frequency distribution characteristics of the bridge structure;

[0084] The modal identification module is used to analyze the natural frequency change and damping ratio data based on the frequency distribution characteristics using the modal parameter identification algorithm, build a bridge dynamic characteristics model, and determine the modal parameter distribution;

[0085] The stiffness assessment module is used to compare historical data with current data to calculate the natural frequency deviation and the degree of damping ratio anomaly if the modal parameter distribution deviates from the preset threshold range, thereby obtaining evidence of structural stiffness degradation;

[0086] The damage quantification module is used to quantify the deviation value and abnormality degree based on the evidence of structural stiffness degradation using the damage threshold standard, generate a damage probability index, and determine the potential damage severity;

[0087] The positioning analysis module is used to analyze the spatial distribution characteristics of vibration signals based on damage probability indicators using a damage location accuracy improvement algorithm, and then obtain preliminary coordinates of the damage location based on the bridge structure model.

[0088] The area determination module is used to extract the signal-to-noise ratio value of the corresponding area signal based on the preliminary coordinates of the damage location, perform secondary analysis of the local vibration signal, and determine the final damage area range.

[0089] This invention provides a municipal bridge damage and defect detection system to address the problem of vibration signals being easily affected by environmental noise, such as vehicle loads and wind loads, which can hinder the quality and accuracy of collected data. The system acquires a vibration signal dataset from key bridge locations and uses this dataset to determine the final damage area. This system identifies damage based on vibration signals, thereby improving the signal-to-noise ratio of monitoring data and the accuracy of damage location. This provides bridge management departments with more accurate and reliable decision-making, ultimately effectively guiding bridge maintenance and reinforcement efforts.

[0090] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0091] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0093] Figure 1 It is a schematic diagram of the process of the present invention;

[0094] Figure 2 This is a flow chart of the present invention when the modal parameters are abnormal. DETAILED DESCRIPTION

[0095] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0096] according to Figure 1-Figure 2 As shown, an embodiment of the present invention provides a method for detecting damage defects in municipal bridges, comprising:

[0097] Step S1: Obtain a vibration signal dataset of key parts of the bridge, perform multi-scale decomposition and reconstruction on the vibration signal, filter out environmental noise, and generate a denoised vibration signal set;

[0098] Specifically, vibration signals from key parts of the bridge are collected, and the timestamps of the vibration signals are calibrated to obtain the original vibration records containing time series and amplitude information;

[0099] Denoising and filtering are performed on the original vibration records to decompose the original vibration records into multi-scale sub-signals to obtain pre-processed vibration data;

[0100] Extract frequency, amplitude and phase features from the preprocessed vibration data, generate feature vectors of the vibration data, and obtain a feature set for analysis;

[0101] The feature set is classified and processed, and the key parameters in the feature set are compared using a preset threshold. If the key parameters exceed the preset threshold, it is determined that the bridge vibration is abnormal, and the vibration status results of the key parts of the bridge are obtained.

[0102] The vibration signal dataset is subjected to multi-scale decomposition and reconstruction, and the influence of environmental noise is removed by filtering to obtain a denoised vibration signal set.

[0103] According to the denoised vibration signal set, the time series characteristics of the signal are extracted, the periodic changes of the signal are decomposed and the periodic distribution characteristics of the signal are obtained;

[0104] The stability of the vibration signal is analyzed by the periodic distribution characteristics of the signal. If the stability parameter deviates from the preset threshold, the signal is further segmented to determine the range of abnormal fluctuations.

[0105] Obtain the local amplitude change characteristics of the signal within the range of abnormal fluctuations, transform the local signal based on frequency domain analysis, and obtain the frequency distribution result of the local signal;

[0106] Based on the frequency distribution results of the local signal, the frequency anomaly of the vibration signal in a specific interval is determined. If the frequency distribution exceeds the preset threshold range, the interval is marked as a potential abnormal segment, and a set of marked signal segments is obtained;

[0107] Extract the phase feature information of the potential abnormal segments in the marked signal segment set, perform difference analysis between the abnormal segments and the normal segments, and determine the phase offset of the vibration signal;

[0108] By using the time series characteristics and frequency distribution results, a comprehensive vibration feature vector is constructed for the phase shift of the vibration signal to obtain a feature data set for subsequent analysis.

[0109] In this scheme, a vibration signal dataset of key parts of the bridge is obtained. The dataset is collected in real time by acceleration sensors and contains original vibration response information. For the vibration signal dataset, a denoising method based on wavelet transform is used to perform multi-scale decomposition and reconstruction to filter out environmental noise interference and obtain a denoised vibration signal set.

[0110] Specifically, vibration signals from key bridge locations are collected using accelerometers. A time synchronization protocol is used to calibrate the timestamps of the vibration signals, resulting in raw vibration records containing time series and amplitude information. Denoising and filtering are performed on the raw vibration records, and wavelet transforms are used to decompose the raw vibration records into multi-scale sub-signals to obtain preprocessed vibration data. Frequency, amplitude, and phase features are extracted from the preprocessed vibration data, and a Fourier transform is used to generate feature vectors for the vibration data, resulting in a feature set for analysis. Classification processing is then performed on the feature set, and key parameters within the feature set are compared using preset thresholds. If a key parameter exceeds the preset threshold, the bridge vibration pattern is judged to be abnormal, resulting in the vibration status of the key bridge locations.

[0111] A wavelet transform method is used to perform multi-scale decomposition and reconstruction on the vibration signal dataset. The effects of ambient noise are removed through filtering to obtain a denoised vibration signal set. Based on this denoised vibration signal set, the time series characteristics of the signals are extracted. Time-domain analysis methods are used to decompose the periodic variations in the signals and determine their periodic distribution characteristics. This periodic distribution is used to analyze the stability of the vibration signals. If the stability parameters deviate from a preset threshold, the signal is further segmented to determine the range of abnormal fluctuations.

[0112] For the range of abnormal fluctuations, the local amplitude change characteristics of the signal are obtained, and the frequency domain analysis tool is used to transform the local signal to obtain the frequency distribution result of the local signal. According to the frequency distribution result of the local signal, the frequency anomaly of the vibration signal in a specific interval is judged. If the frequency distribution exceeds the preset threshold range, the interval is marked as a potential abnormal segment, and the marked signal segment set is obtained. Through the marked signal segment set, the phase feature information of the potential abnormal segment is extracted, and the phase contrast method is used to perform difference analysis between the abnormal segment and the normal segment to determine the phase offset of the vibration signal. For the phase offset of the vibration signal, the time series characteristics and the frequency distribution results are combined to construct a comprehensive vibration feature vector to obtain a feature data set for subsequent analysis.

[0113] Step S2, extracting the frequency components and amplitude characteristics of the denoised vibration signal set to determine the frequency distribution characteristics of the bridge;

[0114] Specifically, the denoised vibration signal set is preprocessed to obtain pure vibration data;

[0115] Based on the clean vibration data, the frequency domain analysis method is implemented, the Fourier transform is used to decompose the signal and extract the frequency components and amplitude characteristics;

[0116] Analyze the distribution rules of the extracted frequency components and amplitude characteristics to determine the frequency distribution characteristics of the bridge structure;

[0117] If there are outliers in the frequency distribution characteristics, they are compared with preset thresholds to determine whether there are potential structural problems. Based on the judgment results, the specific location information of the abnormal frequency components is obtained, and the source of the abnormality is analyzed in combination with the bridge structure characteristics.

[0118] By analyzing the source of anomalies and comparing structural characteristics with frequency distribution, we can determine whether the anomalies affect the overall stability, thereby obtaining stability analysis results.

[0119] The stability analysis results are combined with the data extracted from the frequency components and amplitude features to generate the health status assessment data of the bridge structure.

[0120] In this scheme, the frequency domain analysis method is used to perform Fourier transform processing based on the denoised vibration signal set to extract the frequency components and amplitude characteristics and determine the frequency distribution characteristics of the bridge structure.

[0121] More specifically, by processing the vibration signal, the signal set is pre-processed using denoising technology to obtain pure vibration data. Based on the pure vibration data, the frequency domain analysis method is implemented, and the signal is decomposed using Fourier transform to extract the frequency components and amplitude characteristics. The distribution pattern of the extracted frequency components and amplitude characteristics is analyzed to determine the frequency distribution characteristics of the bridge structure. If there are outliers in the frequency distribution characteristics, a comparison is performed using a preset threshold to determine whether there are potential structural problems. Based on the judgment results, the specific location information of the abnormal frequency components is obtained, and the source of the abnormality is analyzed in combination with the structural characteristics of the bridge. By analyzing the source of the abnormality, a comparison method between the structural characteristics and the frequency distribution is used to determine whether the abnormality affects the overall stability. The stability analysis results are obtained, and combined with the feature extraction data, the health status assessment data of the bridge structure is generated.

[0122] Step S3, analyzing the natural frequency change and damping ratio data based on the frequency distribution characteristics, constructing a dynamic characteristic model, and using it to determine the distribution of modal parameters;

[0123] Specifically, the frequency distribution characteristic data of the vibration signal is collected, and the vibration signal is preprocessed through time domain and frequency domain conversion tools to obtain a preliminary set of dynamic characteristic parameters;

[0124] Analyze the natural frequency change and damping ratio data of the preliminary dynamic characteristic parameter set and compare them with the preset threshold range. If the parameter value exceeds the threshold range, mark the abnormal data and determine the parameter subset that needs calibration;

[0125] Abnormal data is obtained from the parameter subset that needs to be calibrated, and the abnormal values ​​are corrected. By comparing with historical bridge dynamic characteristic data, it is determined whether the corrected parameter values ​​are within the expected range, and a calibrated dynamic characteristic parameter set is obtained;

[0126] Based on the calibrated dynamic characteristic parameter set, a dynamic characteristic distribution diagram of the bridge is constructed. By comparing the distribution of different modal parameters, the correlation between the modal parameters is determined, and the evaluation results of the overall dynamic characteristics of the bridge are obtained.

[0127] In this scheme, based on the frequency distribution characteristics, the modal parameter identification algorithm is used to analyze the natural frequency changes and damping ratio data, build a bridge dynamic characteristics model, and determine the modal parameter distribution.

[0128] Specifically, based on bridge vibration signal data, the frequency distribution characteristic data collected by sensors is preprocessed using time-domain and frequency-domain conversion tools to obtain a preliminary set of dynamic characteristic parameters. For this preliminary set of dynamic characteristic parameters, a modal parameter identification tool is used to analyze the natural frequency variation and damping ratio data. Comparisons are performed within a preset threshold range. If the parameter value exceeds the threshold range, the abnormal data is marked, and the parameter subset requiring calibration is determined. Abnormal data is obtained from the parameter subset requiring calibration, and the abnormal values ​​are corrected using a data smoothing tool. By comparing historical bridge dynamic characteristic data, it is determined whether the corrected parameter values ​​fall within the expected range, resulting in a calibrated dynamic characteristic parameter set. Based on the calibrated dynamic characteristic parameter set, a visualization tool is used to construct a bridge dynamic characteristic distribution diagram. By comparing the distributions of different modal parameters, the correlation between the modal parameters is determined, and an assessment of the overall dynamic characteristics of the bridge is obtained.

[0129] Step S4: When the modal parameter distribution deviates from the preset threshold range, the natural frequency deviation value and the degree of damping ratio abnormality are calculated by comparing with historical data to generate evidence of structural stiffness degradation;

[0130] Specifically, based on the bridge vibration signal, the modal parameters of the current data are obtained, and the modal parameters are preliminarily screened. If the modal parameters exceed the preset threshold range, they are marked as abnormal parameters, and an abnormal parameter set is obtained;

[0131] Extract the modal parameter distribution range from the historical data of the abnormal parameter set, calculate the deviation amplitude of the natural frequency between the current data and the historical data, and obtain the frequency deviation data set;

[0132] Calculate the abnormality of the damping ratio of the frequency deviation data set. If the abnormality exceeds a preset threshold, mark it as a high-risk parameter and obtain a high-risk parameter subset.

[0133] Obtain relevant data on structural stiffness from the high-risk parameter subset, and use a pre-established analytical model to quantify the basis for structural stiffness change to obtain a set of stiffness change indicators;

[0134] Generate a distribution map of structural stiffness changes using a set of stiffness change indices, determine whether there are significant abnormal areas in the distribution map, and obtain an abnormal area identification set;

[0135] Analyze the correlation between the natural frequency and damping ratio of the abnormal area identification set. If the correlation is lower than the preset threshold range, it is marked as a potential risk point and the potential risk point set is determined;

[0136] Compare historical data of the potential risk point set to further verify the matching degree between the structural stiffness change basis and the current data, determine whether the matching degree is within the expected range, and obtain the final verification result set.

[0137] In this approach, if the modal parameter distribution deviates from a preset threshold, historical data is compared with current data to calculate the natural frequency deviation and damping ratio anomaly, providing evidence of structural stiffness degradation. This evidence serves as the final verification result set.

[0138] More specifically, it includes:

[0139] Step 1: Obtain the modal parameters of the current data from the collected bridge vibration signal, and use data processing tools to perform preliminary screening of the modal parameters. If the modal parameters exceed the preset threshold range, they are marked as abnormal parameters, and an abnormal parameter set is obtained.

[0140] Step 2: For the abnormal parameter set, use the data comparison tool to extract the modal parameter distribution range in the historical data, calculate the deviation amplitude of the natural frequency between the current data and the historical data, and obtain the frequency deviation data set.

[0141] Step 3: Based on the frequency deviation dataset, an analysis tool is used to calculate the abnormality of the damping ratio. If the abnormality exceeds the preset threshold range, it is marked as a high-risk parameter, and a high-risk parameter subset is obtained.

[0142] Step 4: Obtain relevant data on structural stiffness from the high-risk parameter subset, use the pre-established analysis model to quantify the basis for structural stiffness change, and obtain a set of stiffness change indicators.

[0143] Step 5: Using the stiffness change index set, a visualization tool is used to generate a distribution map of the structural stiffness change, and whether there are significant abnormal areas in the distribution map is determined to obtain an abnormal area identification set.

[0144] Step 6: Based on the abnormal area identification set, use the data fusion tool to analyze the correlation between the natural frequency and the damping ratio. If the correlation is lower than the preset threshold range, it is marked as a potential risk point and the potential risk point set is determined.

[0145] Step 7: For the set of potential risk points, further verify the matching degree between the structural stiffness change basis and the current data through the historical data comparison tool, determine whether the matching degree is within the expected range, and obtain the final verification result set.

[0146] The natural frequency deviation value is the deviation amplitude of the natural frequency, and the frequency deviation data set is obtained based on the deviation amplitude;

[0147] Step S5, quantitatively evaluating the deviation value and abnormality degree of the structural stiffness degradation evidence to generate a damage possibility index;

[0148] Specifically, the monitoring data of structural stiffness is obtained, and preliminary screening is performed for degradation evidence. By comparing the changing trends of historical data and current data, the significance judgment results of the degradation evidence are obtained;

[0149] Based on the significance judgment results of the degradation evidence, the calculation method of the deviation value and the abnormality degree is used, combined with the preset damage threshold standard, to determine the correlation assessment value between the deviation value and the abnormality degree;

[0150] By evaluating the correlation between the deviation value and the abnormality degree, the damage possibility is classified and predicted to obtain preliminary indicator data of the damage possibility;

[0151] Based on the preliminary indicator data of the possibility of injury and the distribution characteristics of potential injuries, the severity is graded to obtain a quantitative level of severity;

[0152] By combining the quantitative severity level with the standard assessment rule base, if the quantitative level exceeds the preset threshold range, the potential damage will be marked and the key monitoring area will be determined;

[0153] Based on the marking results of key monitoring areas, the structural stiffness data and updated information of degradation evidence in the relevant areas are obtained. Through data fusion processing, the dynamic changes of damage trends are judged;

[0154] Time series analysis is used to predict the dynamic changes in damage trends over the long term, providing updated data on the evolution path and severity of potential damage. Quantitative assessments of deviations and anomalies involve comparing historical data to calculate natural frequency deviations and damping ratio anomalies when modal parameter distributions deviate from preset thresholds.

[0155] In this scheme, based on the evidence of structural stiffness degradation, the damage threshold standard is used to quantitatively evaluate the deviation value and abnormality degree, generate a damage possibility index, and determine the potential damage severity.

[0156] Specifically, structural stiffness monitoring data is obtained and initially screened for degradation evidence. By comparing the changing trends of historical and current data, the significance of the degradation evidence is determined. Based on the significance of the degradation evidence, a method for calculating deviation values ​​and anomaly levels is used, combined with a preset damage threshold standard, to determine the correlation between the deviation values ​​and the anomaly levels. Using this correlation, a support vector machine model is used to classify and predict damage potential, generating preliminary indicators of damage potential. Based on these preliminary indicators of damage potential and the distribution characteristics of potential damage, severity is graded to obtain a quantitative severity level. Based on this quantitative severity level, combined with a standard assessment rule base, if the quantitative level exceeds the preset threshold, the potential damage is marked and key monitoring areas are identified. Based on the marking results for key monitoring areas, updated structural stiffness data and degradation evidence are obtained for the relevant areas. Through data fusion, the dynamic changes in damage trends are determined. Based on these dynamic changes in damage trends, a time series analysis method is used to perform long-term predictions of the changing trends, generating updated potential damage evolution paths and severity data.

[0157] Step S6, based on the damage possibility index, combined with the spatial distribution characteristics of the vibration signal and the bridge structure model, determining the preliminary coordinates of the damage location;

[0158] Specifically, vibration signal data of the bridge structure is acquired to obtain an original vibration signal, and the original vibration signal is subjected to denoising and filtering to obtain a processed first vibration signal data set;

[0159] Analyzing the spatial distribution characteristics of the processed first vibration signal data set, and determining the spatial distribution regularity characteristics of the vibration signal using a pre-established distribution characteristic analysis model;

[0160] Based on the spatial distribution characteristics of the vibration signal, a finite element analysis is performed on the pre-built bridge structure model to obtain the stress distribution data corresponding to the distribution characteristics in the structural model;

[0161] If there are abnormal areas in the stress distribution data, the preset stress threshold range is compared to determine whether the abnormal area is a potential damage location, and a preliminary set of damage locations is obtained;

[0162] further calibrating the spatial coordinates of each potential damage location in the preliminary damage location set to determine a calibrated second damage location set;

[0163] By performing multi-dimensional verification on the second damage location set and combining the spatial distribution characteristics of the vibration signal with the geometric parameters of the bridge structure model, the final damage location coordinate data is obtained;

[0164] If the final damage location coordinate data does not match the preset damage possibility index range, the first vibration signal data set is traced back to adjust the signal processing parameters and re-analyze to obtain more accurate damage location coordinates.

[0165] In this scheme, based on the damage possibility index, a damage location accuracy improvement algorithm is used to analyze the spatial distribution characteristics of the vibration signal, and combined with the bridge structure model, the preliminary coordinates of the damage location are obtained.

[0166] Specifically, vibration signal data from the bridge structure is collected via a sensor network. Signal preprocessing techniques are used to denoise and filter the raw vibration signals, generating a processed first vibration signal dataset. Based on the processed first vibration signal dataset, its spatial distribution characteristics are analyzed, and a pre-established distribution characteristic analysis model is used to determine the spatial distribution characteristics of the vibration signals. Based on the spatial distribution characteristics of the vibration signals, a finite element analysis method is used, combined with a pre-constructed bridge structure model, to obtain stress distribution data corresponding to the distribution characteristics within the structural model. If abnormal areas are present in the stress distribution data, the abnormal areas are compared against a preset stress threshold to determine whether they are potential damage locations, thereby generating a preliminary set of damage locations. Based on this preliminary set of damage locations, a damage localization accuracy enhancement algorithm is used to further calibrate the spatial coordinates of each potential damage location, resulting in a second set of calibrated damage locations. Multi-dimensional validation of the second set of damage locations is performed, combining the spatial distribution characteristics of the vibration signals with the geometric parameters of the bridge structure model to obtain the final damage location coordinate data. If the final damage location coordinate data does not match the preset damage possibility index range, the first vibration signal data set is traced back to adjust the signal processing parameters and re-analyze to obtain more accurate damage location coordinates.

[0167] Step S7: extract the signal-to-noise ratio value of the signal in the area corresponding to the damage location coordinates, perform secondary analysis of the local vibration signal, and determine the final damage area range.

[0168] Specifically, by obtaining relevant information about the damage location from the preliminary coordinates, the data of the corresponding area is preliminarily extracted to obtain the initial distribution of the signal-to-noise ratio;

[0169] Based on the initial distribution of the signal-to-noise ratio, the local vibration signal is decomposed into features, and the abnormal points are screened using a preset threshold to determine the key abnormal areas in the vibration signal.

[0170] If the signal strength in a critical abnormal area exceeds a preset threshold, the vibration signal in that area is deeply processed to obtain more detailed signal data and determine the potential damage range;

[0171] By analyzing the frequency domain of the fine signal data, the characteristic parameters of the local vibration are extracted and combined with the preliminary coordinates of the damage location to determine the boundary range of the final area;

[0172] If there is a significant deviation between the boundary range of the final area and the preliminary coordinates, the signal data is corrected to obtain the corrected distribution characteristics and determine the accuracy of the damage range;

[0173] According to the corrected distribution characteristics, the support vector machine algorithm is used to classify and confirm the damage range to obtain the final damage area division result;

[0174] By comparing and analyzing the final damage area division results and combining them with the historical signal data of the corresponding area, the dynamic change trend of the damage location is obtained and the key areas for long-term monitoring are determined.

[0175] In this scheme, based on the preliminary coordinates of the damage location, the signal-to-noise ratio value of the corresponding area is extracted, and a secondary analysis of the local vibration signal is performed to determine the final range of the damage area.

[0176] Specifically, by obtaining information about the damage location from the preliminary coordinates, signal processing tools are used to perform preliminary data extraction on the corresponding area to obtain an initial signal-to-noise ratio (SNR) distribution. Based on this initial SNR distribution, the local vibration signal is subjected to feature decomposition, and outliers are screened using a preset threshold to identify key abnormal regions within the vibration signal. If the signal strength in a key abnormal region exceeds the preset threshold, the vibration signal in that region is further processed to obtain more detailed signal data to determine the potential damage extent. Frequency domain analysis of the refined signal data is used to extract characteristic parameters of the local vibration. Combined with the preliminary coordinates of the damage location, the boundaries of the final region are determined. If the boundaries of the final region deviate significantly from the preliminary coordinates, the signal data is corrected to obtain corrected distribution characteristics and determine the accuracy of the damage extent. Based on the corrected distribution characteristics, a support vector machine algorithm is used to classify and confirm the damage extent, resulting in the final damage region demarcation. By comparing and analyzing the final damage region demarcation results with historical signal data from the corresponding region, the dynamic trend of the damage location is determined, and key areas for long-term monitoring are identified.

[0177] This invention provides a method for detecting damage and defects in municipal bridges, addressing the problem of vibration signals being easily affected by environmental noise, such as vehicle loads and wind loads, which can hinder the quality and accuracy of collected data. The method can identify damage to bridges based on vibration signals, thereby improving the signal-to-noise ratio of monitoring data and the accuracy of damage location. This provides bridge management departments with more accurate and reliable decision-making, ultimately effectively guiding bridge maintenance and reinforcement efforts.

[0178] More specifically, in step S1, the technical content of obtaining a vibration signal dataset of key parts of a bridge and collecting raw vibration response information in real time through an acceleration sensor can be integrated into the following detailed process:

[0179] First, assuming that the critical parts of the bridge are the piers and main beams, high-precision accelerometers (such as model XYZ-100, with a sampling frequency of 200 Hz and a sensitivity of 100 mV / g) are selected and placed on the tops of the piers and in the middle of the main beams. Three sensors are installed at each location to cover the three axes (X, Y, and Z) to ensure that comprehensive vibration data is collected. The data collection time is set to be continuous for 24 hours, and the data volume generated is expected to be approximately 432,000 data points per sensor per hour.

[0180] Secondly, the acquisition system uploads data to the cloud server in real time through a wireless transmission module, using a 5G network to ensure a transmission rate of 50Mbps to avoid data loss. At the same time, the Hadoop distributed storage system is used in the cloud to store and manage data to ensure data integrity.

[0181] Next, the original vibration signal was preprocessed. The time domain signal was converted into a frequency domain signal using the Fast Fourier Transform (FFT) algorithm, and the dominant frequency component was extracted. Hypothesis analysis found that the dominant frequency was concentrated in the range of 2.5Hz to 5.0Hz, indicating a possible risk of structural resonance. Subsequently, further denoising was performed using wavelet transform (using the Daubechies wavelet basis) to filter out noise signals below 0.1Hz and above 50Hz, resulting in relatively pure vibration response data.

[0182] Finally, the bridge model was simulated using finite element analysis software (such as ANSYS). The processed vibration data was input and the stress distribution in key areas was calculated. The maximum stress at the top of the pier was 15.2 MPa, which did not exceed the design limit of 20 MPa. The structure was deemed safe, but continued monitoring of the main frequency was required to prevent fatigue damage. This process forms a closed-loop logic from data collection to analysis, ensuring the validity of the vibration signal and the reliability of the analysis results. Furthermore, the analysis results can be linked to the bridge health monitoring system, combining them with environmental data such as temperature and wind speed to further optimize the monitoring model.

[0183] In step S4, assume that the modal parameters of a bridge are collected via vibration sensors, including the natural frequency and damping ratio. The preset threshold ranges are 8.0-8.5 Hz for natural frequency and 0.02-0.05 for damping ratio. Signal processing software then performs a fast Fourier transform (FFT) on the acceleration data collected by the sensors, obtaining the current natural frequency of 7.8 Hz and a damping ratio of 0.06. Upon detecting that the parameters deviate from the threshold range, the system automatically extracts the modal parameters recorded monthly for the past year from a historical database, containing 12 sets of data, such as 8.3 Hz and 0.03 for June 2024, and 8.2 Hz and 0.035 for December 2024.

[0184] Calculate the natural frequency deviation value, the algorithm is the difference between the current value and the historical mean: , deviation = 7.8-8.25 = -0.45 Hz, which means that the frequency is reduced.

[0185] The degree of abnormality of the damping ratio is calculated using the relative rate of change: the historical average is 0.032, , an increase of 87.5%.

[0186] Combining the deviation value and the degree of abnormality, the stiffness change is calculated using the stiffness degradation model K: (m is mass, assuming 1000 tons);

[0187] , current stiffness ,

[0188] , or 10.4%. Analysis shows that

[0189] Stiffness degradation is positively correlated with decreasing frequency, while an increase in damping ratio may be due to material fatigue or loose connections. This generates a report recommending further inspection of the bridge beam connections. This process is automated through scripts. Frequency deviations logically trigger stiffness calculations, while damping anomalies provide additional verification, ensuring reliable conclusions.

[0190] In step S5, to assess evidence of structural stiffness degradation, sensors are first used to collect displacement data of the structure under different loads. For example, if the displacement of a bridge beam under a standard load of 100 kN is 5.2 mm, while the displacement under the historical healthy state is 4.8 mm, the calculated deviation is 0.4 mm.

[0191] Then, the damage threshold standard is used for quantitative evaluation. The displacement threshold obtained by statistical analysis of historical data is set to 0.5mm. The deviation value of 0.4mm is less than the threshold, but close to the critical point. Therefore, the abnormality degree formula is used to The calculated abnormality level was 80%, indicating that some abnormality existed but did not reach the standard of severe injury.

[0192] Then, the damage possibility index is generated and the index calculation model is set as , where the weight coefficient is 0.9 according to the structural importance, and the calculated probability index is 72%, reflecting a high potential damage risk.

[0193] Ultimately, the severity of the potential damage was determined. Combining the likelihood index with structural material aging data (assuming a 30% reduction in material fatigue life), a comprehensive assessment algorithm (severity = likelihood index × (1 + aging rate)) yielded a severity of 93.6%, placing it at a medium-to-high risk level and requiring priority attention. To establish a rigorous logical chain, these results were linked to routine bridge monitoring operations, automatically triggering the system to generate maintenance recommendation reports. Monitoring frequency was automatically adjusted based on the severity value. For example, if the severity exceeded 90%, monitoring frequency would be increased from weekly to daily, ensuring real-time data updates and supporting subsequent decision-making. This approach forms a complete closed-loop system, from data collection to severity assessment and business linkage. The entire process is automated by the system, ensuring assessment accuracy and efficiency.

[0194] In step S7, when processing the preliminary coordinate data of the damage location, assuming that the initial damage location coordinates are (x=25.5, y=30.2), the system first divides the area with a radius of 5.0 meters around the coordinates into 100 small grid cells using a preset grid division algorithm. Each cell has an area of ​​0.25 square meters. Then, the vibration signal data in the area is extracted and the signal-to-noise ratio (SNR) of each cell is calculated. The formula is: ,in is the signal power, is the noise power. Assuming that the signal power of a unit is 0.8 watts and the noise power is 0.2 watts, then ,The system sorts the SNR values ​​of all units and selects the units with SNR lower than 5.0 decibels as potential abnormal areas.

[0195] Next, the system enters the secondary analysis stage of the local vibration signal. It performs time-frequency analysis on the screened low SNR units (assuming there are 10 of them). Using the short-time Fourier transform (STFT) algorithm, it sets the time window length to 0.1 seconds and the frequency resolution to 10 Hz to extract the spectral characteristics of the vibration signal. It is found that 8 of the units have significant peaks at 50 Hz, with amplitudes exceeding 0.5 volts, indicating that these areas may have abnormal vibrations caused by damage. The remaining 2 units have no obvious spectral abnormalities and may be noise interference.

[0196] Finally, the system combined the coordinate ranges of these eight units with the results of the secondary analysis, determining the final damage area as a circular area with a radius of 3.2 meters centered on the initial coordinates, covering an area of ​​approximately 32.17 square meters. The results were then compared with the historical damage database, confirming that this area matched past damage patterns with an 85% accuracy, further verifying the accuracy of the judgment. Through the aforementioned signal processing and algorithmic analysis, the system automated the process from initial coordinates to final damage range determination, ensuring the scientific and reliable nature of the results.

[0197] The present invention provides a municipal bridge damage defect detection system, which uses the municipal bridge damage defect detection method to perform defect detection. The system includes:

[0198] The data acquisition module is used to obtain the vibration signal data set of the key parts of the bridge. The data set is collected in real time by the acceleration sensor and contains the original vibration response information;

[0199] The signal denoising module is used to perform multi-scale decomposition and reconstruction on the vibration signal dataset using a denoising method based on wavelet transform, filter out environmental noise interference, and obtain a denoised vibration signal set;

[0200] The frequency domain analysis module is used to perform Fourier transform processing based on the denoised vibration signal set using the frequency domain analysis method to extract the frequency components and amplitude characteristics and determine the frequency distribution characteristics of the bridge structure;

[0201] The modal identification module is used to analyze the natural frequency change and damping ratio data based on the frequency distribution characteristics using the modal parameter identification algorithm, build a bridge dynamic characteristics model, and determine the modal parameter distribution;

[0202] The stiffness assessment module is used to compare historical data with current data to calculate the natural frequency deviation and the degree of damping ratio anomaly if the modal parameter distribution deviates from the preset threshold range, thereby obtaining evidence of structural stiffness degradation;

[0203] The damage quantification module is used to quantify the deviation value and abnormality degree based on the evidence of structural stiffness degradation using the damage threshold standard, generate a damage probability index, and determine the potential damage severity;

[0204] The positioning analysis module is used to analyze the spatial distribution characteristics of vibration signals based on damage probability indicators using a damage location accuracy improvement algorithm, and then obtain preliminary coordinates of the damage location based on the bridge structure model.

[0205] The area determination module is used to extract the signal-to-noise ratio value of the corresponding area signal based on the preliminary coordinates of the damage location, perform secondary analysis of the local vibration signal, and determine the final damage area range.

[0206] This invention provides a municipal bridge damage and defect detection system to address the problem of vibration signals being easily affected by environmental noise, such as vehicle loads and wind loads, which can hinder the quality and accuracy of collected data. The system can identify damage to bridges based on vibration signals, thereby improving the signal-to-noise ratio of monitoring data and the accuracy of damage location. This system provides bridge management departments with more accurate and reliable decision-making, ultimately effectively guiding bridge maintenance and reinforcement efforts.

[0207] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A municipal bridge damage defect detection method, characterized in that: include: Obtain a data set of vibration signals from key parts of the bridge, perform multi-scale decomposition and reconstruction on the vibration signals, filter out environmental noise, and generate a denoised vibration signal set; Extract the frequency components and amplitude characteristics of the denoised vibration signal set to determine the frequency distribution characteristics of the bridge; Based on the frequency distribution characteristics, the natural frequency change and damping ratio data are analyzed to build a dynamic characteristic model, which is used to determine the distribution of modal parameters; When the modal parameter distribution deviates from the preset threshold range, the natural frequency deviation value and the degree of damping ratio abnormality are calculated by comparing historical data to generate evidence of structural stiffness degradation; Quantitatively evaluate the deviation value and abnormality degree of structural stiffness degradation evidence to generate damage probability indicators; Based on the damage probability index, combined with the spatial distribution characteristics of the vibration signal and the bridge structure model, the preliminary coordinates of the damage location are determined; The signal-to-noise ratio value of the signal in the area corresponding to the damage location coordinates is extracted, and a secondary analysis of the local vibration signal is performed to determine the final damage area range.

2. A municipal bridge damage defect detection method according to claim 1, characterized in that: Collect vibration signals from key parts of the bridge, calibrate the timestamps of the vibration signals, and obtain original vibration records containing time series and amplitude information; Denoising and filtering are performed on the original vibration records to decompose the original vibration records into multi-scale sub-signals to obtain pre-processed vibration data; Extract frequency, amplitude and phase features from the preprocessed vibration data, generate feature vectors of the vibration data, and obtain a feature set for analysis; The feature set is classified and processed, and the key parameters in the feature set are compared using a preset threshold. If the key parameters exceed the preset threshold, it is determined that the bridge vibration is abnormal, and the vibration status results of the key parts of the bridge are obtained.

3. A municipal bridge damage defect detection method according to claim 1, characterized in that: The vibration signal dataset is subjected to multi-scale decomposition and reconstruction, and the influence of environmental noise is removed by filtering to obtain a denoised vibration signal set. According to the denoised vibration signal set, the time series characteristics of the signal are extracted, the periodic changes of the signal are decomposed and the periodic distribution characteristics of the signal are obtained; The stability of the vibration signal is analyzed by the periodic distribution characteristics of the signal. If the stability parameter deviates from the preset threshold, the signal is further segmented to determine the range of abnormal fluctuations. Obtain the local amplitude change characteristics of the signal within the range of abnormal fluctuations, transform the local signal based on frequency domain analysis, and obtain the frequency distribution result of the local signal; Based on the frequency distribution results of the local signal, the frequency anomaly of the vibration signal in a specific interval is determined. If the frequency distribution exceeds the preset threshold range, the interval is marked as a potential abnormal segment, and a set of marked signal segments is obtained; Extract the phase feature information of the potential abnormal segments in the marked signal segment set, perform difference analysis between the abnormal segments and the normal segments, and determine the phase offset of the vibration signal; By using the time series characteristics and frequency distribution results, a comprehensive vibration feature vector is constructed for the phase shift of the vibration signal to obtain a feature data set for subsequent analysis.

4. A municipal bridge damage defect detection method according to claim 1, characterized in that: Preprocess the denoised vibration signal set to obtain pure vibration data; Based on the clean vibration data, the frequency domain analysis method is implemented, the Fourier transform is used to decompose the signal and extract the frequency components and amplitude characteristics; Analyze the distribution rules of the extracted frequency components and amplitude characteristics to determine the frequency distribution characteristics of the bridge structure; If there are outliers in the frequency distribution characteristics, they are compared with preset thresholds to determine whether there are potential structural problems. Based on the judgment results, the specific location information of the abnormal frequency components is obtained, and the source of the abnormality is analyzed in combination with the bridge structure characteristics. By analyzing the source of anomalies and comparing structural characteristics with frequency distribution, we can determine whether the anomalies affect the overall stability, thereby obtaining stability analysis results. The stability analysis results are combined with the data extracted from the frequency components and amplitude features to generate the health status assessment data of the bridge structure.

5. A municipal bridge damage defect detection method according to claim 1, characterized in that: Collect frequency distribution characteristic data of vibration signals, pre-process the vibration signals through time domain and frequency domain conversion tools, and obtain a preliminary set of dynamic characteristic parameters; Analyze the natural frequency change and damping ratio data of the preliminary dynamic characteristic parameter set and compare them with the preset threshold range. If the parameter value exceeds the threshold range, mark the abnormal data and determine the parameter subset that needs calibration; Abnormal data is obtained from the parameter subset that needs to be calibrated, and the abnormal values ​​are corrected. By comparing with historical bridge dynamic characteristic data, it is determined whether the corrected parameter values ​​are within the expected range, and a calibrated dynamic characteristic parameter set is obtained; Based on the calibrated dynamic characteristic parameter set, a dynamic characteristic distribution diagram of the bridge is constructed. By comparing the distribution of different modal parameters, the correlation between the modal parameters is determined, and the evaluation results of the overall dynamic characteristics of the bridge are obtained.

6. A municipal bridge damage defect detection method according to claim 1, characterized in that: Based on the bridge vibration signal, the modal parameters of the current data are obtained and the modal parameters are preliminarily screened. If the modal parameters exceed the preset threshold range, they are marked as abnormal parameters, and an abnormal parameter set is obtained; Extract the modal parameter distribution range from the historical data of the abnormal parameter set, calculate the deviation amplitude of the natural frequency between the current data and the historical data, and obtain the frequency deviation data set; Calculate the abnormality of the damping ratio of the frequency deviation data set. If the abnormality exceeds a preset threshold, mark it as a high-risk parameter and obtain a high-risk parameter subset. Obtain relevant data on structural stiffness from the high-risk parameter subset, and use a pre-established analytical model to quantify the basis for structural stiffness change to obtain a set of stiffness change indicators; Generate a distribution map of structural stiffness changes using a set of stiffness change indices, determine whether there are significant abnormal areas in the distribution map, and obtain an abnormal area identification set; Analyze the correlation between the natural frequency and damping ratio of the abnormal area identification set. If the correlation is lower than the preset threshold range, it is marked as a potential risk point and the potential risk point set is determined; Compare historical data of the potential risk point set to further verify the matching degree between the structural stiffness change basis and the current data, determine whether the matching degree is within the expected range, and obtain the final verification result set.

7. A municipal bridge damage defect detection method according to claim 1, characterized in that: Acquire monitoring data of structural stiffness, conduct preliminary screening for degradation evidence, and obtain significance judgment results of degradation evidence by comparing the changing trends of historical data and current data; Based on the significance judgment results of the degradation evidence, the calculation method of the deviation value and the abnormality degree is used, combined with the preset damage threshold standard, to determine the correlation assessment value between the deviation value and the abnormality degree; By evaluating the correlation between the deviation value and the abnormality degree, the damage possibility is classified and predicted to obtain preliminary indicator data of the damage possibility; Based on the preliminary indicator data of the possibility of injury and the distribution characteristics of potential injuries, the severity is graded to obtain a quantitative level of severity; By combining the quantitative severity level with the standard assessment rule base, if the quantitative level exceeds the preset threshold range, the potential damage will be marked and the key monitoring area will be determined; Based on the marking results of key monitoring areas, the structural stiffness data and updated information of degradation evidence in the relevant areas are obtained. Through data fusion processing, the dynamic changes of damage trends are judged; Time series analysis is used to make long-term predictions of the dynamic changes in damage trends and obtain updated data on the evolution path and severity of potential damage.

8. A municipal bridge damage defect detection method according to claim 1, characterized in that: Acquiring vibration signal data of the bridge structure to obtain an original vibration signal, performing denoising and filtering on the original vibration signal to obtain a processed first vibration signal data set; Analyzing the spatial distribution characteristics of the processed first vibration signal data set, and determining the spatial distribution regularity characteristics of the vibration signal using a pre-established distribution characteristic analysis model; Based on the spatial distribution characteristics of the vibration signal, a finite element analysis is performed on the pre-built bridge structure model to obtain the stress distribution data corresponding to the distribution characteristics in the structural model; If there are abnormal areas in the stress distribution data, the preset stress threshold range is compared to determine whether the abnormal area is a potential damage location, and a preliminary set of damage locations is obtained; further calibrating the spatial coordinates of each potential damage location in the preliminary damage location set to determine a calibrated second damage location set; By performing multi-dimensional verification on the second damage location set and combining the spatial distribution characteristics of the vibration signal with the geometric parameters of the bridge structure model, the final damage location coordinate data is obtained; If the final damage location coordinate data does not match the preset damage possibility index range, the first vibration signal data set is traced back to adjust the signal processing parameters and re-analyze to obtain more accurate damage location coordinates.

9. A municipal bridge damage defect detection method according to claim 1, characterized in that: By obtaining relevant information about the damage location from the preliminary coordinates, the data of the corresponding area is preliminarily extracted to obtain the initial distribution of the signal-to-noise ratio; Based on the initial distribution of the signal-to-noise ratio, the local vibration signal is decomposed into features, and the abnormal points are screened using a preset threshold to determine the key abnormal areas in the vibration signal. If the signal strength in a critical abnormal area exceeds a preset threshold, the vibration signal in that area is deeply processed to obtain more detailed signal data and determine the potential damage range; By analyzing the frequency domain of the fine signal data, the characteristic parameters of the local vibration are extracted and combined with the preliminary coordinates of the damage location to determine the boundary range of the final area; If there is a significant deviation between the boundary range of the final area and the preliminary coordinates, the signal data is corrected to obtain the corrected distribution characteristics and determine the accuracy of the damage range; According to the corrected distribution characteristics, the support vector machine algorithm is used to classify and confirm the damage range to obtain the final damage area division result; By comparing and analyzing the final damage area division results and combining them with the historical signal data of the corresponding area, the dynamic change trend of the damage location is obtained and the key areas for long-term monitoring are determined.

10. A municipal bridge damage and defect detection system, applying the municipal bridge damage and defect detection method according to any one of claims 1 to 9 to perform defect detection, the system comprising: The data acquisition module is used to obtain the vibration signal data set of the key parts of the bridge. The data set is collected in real time by the acceleration sensor and contains the original vibration response information; The signal denoising module is used to perform multi-scale decomposition and reconstruction on the vibration signal dataset using a denoising method based on wavelet transform, filter out environmental noise interference, and obtain a denoised vibration signal set; The frequency domain analysis module is used to perform Fourier transform processing based on the denoised vibration signal set using the frequency domain analysis method to extract the frequency components and amplitude characteristics and determine the frequency distribution characteristics of the bridge structure; The modal identification module is used to analyze the natural frequency change and damping ratio data based on the frequency distribution characteristics using the modal parameter identification algorithm, build a bridge dynamic characteristics model, and determine the modal parameter distribution; The stiffness assessment module is used to compare historical data with current data to calculate the natural frequency deviation and the degree of damping ratio anomaly if the modal parameter distribution deviates from the preset threshold range, thereby obtaining evidence of structural stiffness degradation; The damage quantification module is used to quantify the deviation value and abnormality degree based on the evidence of structural stiffness degradation using the damage threshold standard, generate a damage probability index, and determine the potential damage severity; The positioning analysis module is used to analyze the spatial distribution characteristics of vibration signals based on damage probability indicators using a damage location accuracy improvement algorithm, and then obtain preliminary coordinates of the damage location based on the bridge structure model. The area determination module is used to extract the signal-to-noise ratio value of the corresponding area signal based on the preliminary coordinates of the damage location, perform secondary analysis of the local vibration signal, and determine the final damage area range.

Images