Bridge pavement layer damage detection method and system based on ground detection radar

Abstract

The invention relates to the field of bridge detection, in particular to a bridge pavement layer damage detection method and system based on a ground detection radar. The method comprises the following steps: scanning a bridge pavement layer by using a ground detection radar, and collecting radar echo data; performing denoising and filtering processing to obtain a clear radar signal; judging whether a damage sign exists or not, if so, determining the radar signal as the damage sign, and entering the next step, otherwise, returning to the first step; waveform coding feature extraction is carried out on the detected damage signs, and damage features are identified; carrying out comparative analysis on the extracted damage features and a database in a normal state, and determining existence and types of the damage; according to an analysis result, utilizing an adaptive frequency modulation technology to accurately detect the damage depth and range; predicting a damage development trend; and obtaining a detection report according to the analysis result and the prediction result. The method improves the damage detection precision and detection efficiency, and achieves the accurate prediction of the damage development trend.

Description

Technical Field

[0001] The present invention relates to the field of bridge detection, and in particular to a bridge pavement damage detection method and system based on ground detection radar. Background Art

[0002] As a key transportation infrastructure, bridges play a vital role in ensuring public safety and promoting economic development. If the damage of the bridge pavement is not detected and maintained in time, it will not only accelerate the degradation of the bridge structure, but also may cause serious safety accidents and cause immeasurable losses. Therefore, it is particularly urgent to develop efficient and accurate bridge damage detection technology.

[0003] Traditional bridge damage detection methods mainly rely on manual inspections, which are not only inefficient and subjective, but also have obvious deficiencies in discovering hidden damage. With technological advances, automated detection technologies such as acoustic wave detection and ultrasonic detection are gradually being applied to bridge damage detection. Nevertheless, these technologies often have high requirements for detection conditions and have limited ability to identify certain types of damage.

[0004] In recent years, ground detection radar technology has attracted widespread attention because of its ability to penetrate the pavement and detect internal structures. This technology can evaluate the damage of the bridge pavement by analyzing the reflection and propagation characteristics of radar signals. However, existing radar detection methods still have some limitations, such as the lack of accuracy of signal processing algorithms, which makes it difficult to distinguish between minor damage and background noise; the single method of damage feature extraction cannot effectively identify multiple types of damage; the lack of intelligence in the prediction of damage development trends cannot provide sufficient scientific basis for maintenance decisions. These problems limit the widespread application and effectiveness of radar technology in bridge damage detection.

[0005] The existing patent CN1 15824816A proposes a non-destructive detection method for effective prestress of bridges based on ground penetrating radar. However, this technology mainly focuses on the prestress detection of bridges and does not involve the continuity analysis and prediction of damage detection.

[0006] Existing patent CN106289691B discloses a bridge block impact vibration detection method and detection device based on a microwave radar device. Although this technology provides a new method for bridge vibration testing, it is mainly aimed at analyzing the structural dynamic characteristics of the bridge rather than directly detecting damage.

[0007] In view of the problems and limitations of the existing technology, there is an urgent need for an improved bridge pavement damage detection method based on ground detection radar. Summary of the invention

[0008] The purpose of the present invention is to provide a bridge pavement damage detection method and system based on ground detection radar to solve the problems of insufficient damage detection accuracy, low detection efficiency and inability to accurately predict damage development trends in the prior art.

[0009] To achieve the above purpose, the following technical solution is adopted.

[0010] A bridge pavement damage detection method based on ground detection radar, comprising:

[0011] S 1. Use ground detection radar to scan the bridge pavement and collect radar echo data;

[0012] S2, denoising and filtering the radar echo data collected by S1 to obtain a clear radar signal;

[0013] S3, judging whether there is a sign of damage. If the amplitude change of the radar signal or the signal arrival time exceeds the preset threshold, the radar signal is determined to be a sign of damage and the process goes to S4; otherwise, the current state is recorded and the process goes back to S1, and the scan is repeated according to the preset period.

[0014] S4, extracting waveform coding features from the detected damage signs and identifying damage features;

[0015] S5, comparing and analyzing the extracted damage features with the normal state database to determine the existence and type of damage;

[0016] S6. Based on the analysis results, the adaptive frequency modulation technology is used to accurately detect the damage depth and range;

[0017] S7. Based on the accurate detection results, the intelligent damage evolution tracking algorithm is used to predict the damage development trend;

[0018] S8. Obtain a test report based on the analysis results and prediction results.

[0019] Optionally, the specific steps of S1 include:

[0020] Select an operating frequency that matches the material characteristics and depth of the bridge pavement to ensure that radar waves penetrate and are reflected effectively; configure the radar transmitter and receiver according to the selected operating frequency, set an appropriate pulse repetition frequency and signal bandwidth, and fix the radar antenna on the scanning device, which is used to move along a predetermined path on the bridge pavement;

[0021] Start the scanning device so that the radar antenna scans the bridge pavement point by point in a straight line or grid pattern to achieve full coverage; during the scanning process, continuously transmit radar signals and receive echo signals reflected from the bridge pavement;

[0022] Using time delay and amplitude information, the echo signal of each scanning point is recorded to generate a radar signal distribution map of the bridge pavement;

[0023] The obtained radar signal distribution map is integrated to form complete radar echo data.

[0024] Optionally, step S2 specifically includes:

[0025] Import the collected radar echo data into digital signal processing software with signal denoising and filtering functions;

[0026] Using the imported radar echo data, perform fast Fourier transform to convert the radar echo data time domain signal into frequency domain signal;

[0027] In the resulting frequency domain signal, the noise frequency components other than radar signals, including abnormal peaks or background noise far from the main frequency peaks, are identified by visual inspection or automatic algorithms;

[0028] Applying a preset band-stop filter to the frequency domain signal to suppress the noise frequency component, and then performing an inverse fast Fourier transform (IFFT) to convert the processed frequency domain signal back to the time domain to obtain a denoised signal;

[0029] Applying a median filter to the resulting denoised signal further eliminates any remaining impulse noise and obtains a denoised signal;

[0030] Perform amplitude correction and gain adjustment on the signal after denoising to ensure that the amplitude of the signal after denoising is consistent with the original radar echo data, and the damage features remain clearly visible, so as to obtain the processed signal;

[0031] Perform quality assessment on the processed signal to verify the improvement in signal-to-noise ratio and ensure that the signal meets the accuracy and clarity standards required for damage detection to obtain a clear radar signal.

[0032] Optionally, step S3 specifically includes:

[0033] Extracting characteristic parameters of the clear radar signal, including amplitude and arrival time, from the obtained clear radar signal;

[0034] The extracted characteristic parameters are compared with preset normal state parameters, where the preset parameters are based on the radar signal characteristics under an undamaged condition;

[0035] When the extracted amplitude variation exceeds a preset amplitude threshold, or the difference between the signal arrival time and the preset normal arrival time exceeds a preset time threshold, it is determined that there is a sign of damage;

[0036] If it is determined that there is a sign of damage, the characteristic parameters of the sign of damage are recorded and the process proceeds to step S4;

[0037] If it is determined that there is no sign of damage, the current bridge pavement state is considered normal and the current state parameters are recorded;

[0038] The recorded condition parameters will be used for comparison with future inspection results to monitor potential changes in the condition of the bridge pavement;

[0039] After confirming that the status is normal, schedule the next scan according to the predetermined period to continuously monitor the status of the bridge pavement;

[0040] When the detection cycle ends, the detection process automatically restarts from step S1 to ensure continuous and periodic monitoring.

[0041] Optionally, step S4 specifically includes:

[0042] receiving a radar signal with a determined damage indication as an input for waveform encoding feature extraction;

[0043] Perform time-frequency analysis on radar signals to identify abnormal patterns in radar signals, which indicate damaged areas;

[0044] Using the identified abnormal pattern, the radar signal is encoded using waveform coding technology to obtain an encoded signal;

[0045] The encoded signal is subjected to feature extraction to extract parameters related to the damage, including frequency offset and amplitude change, to obtain the extracted damage features.

[0046] Optionally, step S5 specifically includes:

[0047] Extract key parameters from the resulting damage characterization report, including quantitative indicators related to the damage;

[0048] Access and retrieve a pre-established normal state database, which contains radar signal characteristics of bridge pavement without damage; compare and analyze the extracted key parameters with the parameters of the retrieved normal state database one by one;

[0049] Based on the comparison results, determine whether there are parameters that deviate significantly from the normal state, which serves as a preliminary basis for judging the existence of damage;

[0050] If the damage is initially confirmed to exist, the damage characteristic parameters are further analyzed and matched with known damage types in the database;

[0051] Using the matching results, determine the type and possible cause of the injury;

[0052] Record the identified damage type and location, as well as the degree of deviation from the normal state, and form a damage assessment report.

[0053] Optionally, step S6 specifically includes:

[0054] Using the damage location and type information obtained in the step, determine the area that needs to be accurately detected;

[0055] Select the frequency modulation parameters suitable for the damage type and set the starting frequency f start and end frequency f end ;

[0056] A frequency modulator is used to generate a linearly changing frequency sweep signal. The relationship between frequency and time is:

[0057] f(t)=f start +β·t, where β is the rate of change of frequency;

[0058] Transmit the generated frequency sweep signal to the damaged area and record the timestamp r of the reflected signal;

[0059] Calculate the damage depth d using d = v t r / 2, where v is the propagation speed of radar waves in the medium;

[0060] Determine the frequency change Δf of the reflected signal. If there is a frequency change, calculate the two-dimensional size change Δr of the damaged area using the formula Δr = c·Δf / f center 2 , where c is the speed of light, f center is the center frequency of the radar signal.

[0061] Optionally, the specific steps of step S7 include:

[0062] Receive accurate detection results of current damage, including location, depth and size of the damage;

[0063] The precise detection results are fed into the pre-trained intelligent damage evolution tracking algorithm;

[0064] Intelligent damage evolution tracking algorithm analyzes the similarity between current damage characteristics and historical damage cases to determine damage type and development stage;

[0065] Based on the analysis results, the intelligent damage evolution tracking algorithm uses the established mathematical model to predict the path and speed of damage expansion;

[0066] The intelligent damage evolution tracking algorithm provides the predicted status of damage at a specific point in the future based on the prediction results, combined with environmental factors and bridge usage conditions;

[0067] Combined with the predicted status, the potential impact of damage on the safety of the bridge structure is evaluated and the prediction results are obtained.

[0068] Optionally, the development and training of the intelligent damage evolution tracking algorithm specifically includes the following steps:

[0069] Collect historical damage case data of a series of bridge pavements, including the initial state, development process and final state of the damage;

[0070] Define damage characteristic parameters, including damage area, depth, extension rate, and related environmental factors and bridge use conditions;

[0071] Using the collected and defined data, a comprehensive damage signature database is built to provide input data for algorithm training;

[0072] Choose one of the machine learning models, random forest, gradient boosting machine or deep neural network, as the basis for the intelligent damage evolution tracking algorithm;

[0073] Using the damage signature database, the selected machine learning model is trained to minimize the prediction error by adjusting the model parameters;

[0074] Validate the trained model and use cross-validation or an independent test set to evaluate the model’s predictive performance and generalization ability;

[0075] Based on the verification results, fine-tune the model, optimize the model structure and parameters, and improve the accuracy and reliability of the algorithm;

[0076] The final trained and optimized intelligent damage evolution tracking algorithm is deployed into the damage detection system for real-time analysis and prediction of damage development trends.

[0077] A bridge pavement damage detection system based on ground detection radar, comprising:

[0078] A data acquisition module is used to scan the bridge pavement using a ground detection radar and collect radar echo data;

[0079] A signal processing module, used for performing denoising and filtering on the radar echo data collected by the data acquisition module to obtain a clear radar signal;

[0080] A damage sign judgment module is used to judge whether there is a damage sign. If the amplitude change or signal arrival time of the radar signal exceeds a preset threshold, the radar signal is determined to be a damage sign. Otherwise, the current state is recorded and returned to the data acquisition module to rescan according to a predetermined period.

[0081] A feature extraction module is used to extract waveform coding features from detected damage signs and identify damage features;

[0082] The damage analysis module is used to compare and analyze the extracted damage features with the normal state database to determine the existence and type of damage;

[0083] Damage detection module, used to accurately detect damage depth and range using adaptive frequency modulation technology;

[0084] The damage prediction module is used to predict the damage development trend based on the accurate detection results obtained using the intelligent damage evolution tracking algorithm;

[0085] The report generation module is used to obtain a test report based on the analysis results and prediction results.

[0086] Compared with the prior art, the present invention has the following beneficial effects:

[0087] This application provides a bridge pavement damage detection method based on ground detection radar, which significantly improves the accuracy and resolution of damage detection by using adaptive frequency modulation technology and intelligent damage evolution tracking algorithm. Compared with traditional methods, this method can more accurately identify and locate minor damage, avoid missed detection and misjudgment, and solve the problems of insufficient damage detection accuracy, low detection efficiency, and inability to accurately predict damage development trends.

[0088] The signal processing module of this application effectively removes the noise components in the radar signal and retains the damage feature information through the application of fast Fourier transform and band-stop filter. The further application of median filter improves the signal-to-noise ratio of the signal, laying the foundation for the accurate extraction of damage features.

[0089] The collaborative work of the damage sign judgment module, feature extraction module, damage analysis module and damage detection module in this application realizes the intelligent identification of damage features and accurate detection of depth range. The intelligent damage evolution tracking algorithm can predict the development trend of damage based on historical data and current detection results, providing scientific decision-making support for bridge maintenance and repair.

[0090] The detection method of the present application realizes the automation of the damage detection process through modular design, reduces manual intervention, and improves detection efficiency. The automated operation of the data acquisition module and the signal processing module ensures the consistency and repeatability of the detection process.

[0091] The damage prediction module and report generation module of this application can formulate periodic monitoring plans and maintenance strategies according to the damage development trend and current status, providing an effective tool for long-term health monitoring and management of bridges. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1It is a schematic flow chart of the steps of an embodiment of a bridge pavement damage detection method based on ground detection radar according to the present invention.

[0093] Figure 2 It is a schematic diagram of the module structure of an embodiment of a bridge pavement damage detection system based on ground detection radar according to the present invention. DETAILED DESCRIPTION

[0094] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0095] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0096] Example 1

[0097] like Figure 1 As shown, a bridge pavement damage detection method based on ground detection radar includes:

[0098] S 1. Use ground detection radar to scan the bridge pavement and collect radar echo data;

[0099] Specifically, in this step, the radar system selects the optimal operating frequency according to the specific material characteristics of the bridge pavement to ensure effective penetration and reflection of radar waves. The radar antenna performs a grid scan on the bridge pavement along a predetermined path to ensure full coverage, while collecting radar echo data, including but not limited to amplitude, frequency, and time delay information. This ensures comprehensive detection of the bridge pavement and provides the necessary raw data for subsequent damage analysis.

[0100] S2, denoising and filtering the radar echo data collected by S1 to obtain a clear radar signal;

[0101] Specifically, in this step, the time domain signal is first converted to the frequency domain signal using Fourier transform, then a band-stop filter is designed to remove the noise component in the frequency domain, and finally the clean signal is converted back to the time domain through inverse Fourier transform to facilitate subsequent damage sign analysis. This effectively improves the signal quality and lays the foundation for the accuracy of damage detection.

[0102] S3, judging whether there is a sign of damage. If the amplitude change of the radar signal or the signal arrival time exceeds the preset threshold, the radar signal is determined to be a sign of damage and the process goes to S4; otherwise, the current state is recorded and the process goes back to S1, and the scan is repeated according to the preset period.

[0103] Specifically, in this step, by comparing the key parameters of the radar signal with the preset threshold, possible signs of damage are quickly identified. If signs of damage are detected, the process will continue with a detailed analysis of the damage characteristics; if no signs of damage are detected, the system will record the current status and re-execute step S1 after a preset time period. This judgment mechanism can quickly identify potential damage areas and provide targets for further detailed analysis.

[0104] S4, extracting waveform coding features from the detected damage signs and identifying damage features;

[0105] Specifically, in this step, the radar signal of the suspected damage area is further analyzed in time and frequency, and the waveform coding technology is used to extract key parameters reflecting the damage characteristics, such as frequency offset and amplitude attenuation, to provide a basis for accurate identification of the damage. Feature extraction provides key information for quantitative analysis of damage and enhances the accuracy of damage identification.

[0106] S5, comparing and analyzing the extracted damage features with the normal state database to determine the existence and type of damage;

[0107] Specifically, in this step, by comparing the extracted damage features with the pre-established normal state database, statistical analysis or pattern recognition algorithms, such as cluster analysis or classification algorithms, are used to determine the existence, type and severity of the damage. Through comparative analysis, the nature of the damage can be accurately identified, providing a basis for the classification and assessment of the damage.

[0108] S6. Based on the analysis results, the adaptive frequency modulation technology is used to accurately detect the damage depth and range;

[0109] Specifically, in this step, according to the damage type and characteristics determined in step S5, the frequency of the radar signal is adjusted, and the adaptive frequency modulation technology is used to more finely detect the damaged area to determine the specific depth and range of the damage. According to the analysis results, the adaptive frequency modulation technology is used to accurately detect the depth and range of the damage; the application of adaptive technology improves the accuracy of detection and makes the damage location more accurate.

[0110] S7. Based on the accurate detection results, the intelligent damage evolution tracking algorithm is used to predict the damage development trend;

[0111] Specifically, in this step, the intelligent damage evolution tracking algorithm predicts the development trend and potential risks of damage based on the damage depth and range data obtained in step S6, combined with the historical damage data of the bridge and environmental factors. The application of the prediction algorithm provides forward-looking guidance for bridge maintenance and helps to take maintenance measures in a timely manner.

[0112] S8. Obtain a test report based on the analysis results and prediction results.

[0113] Specifically, in this step, the analysis and prediction results from steps S5 to S7 are combined to generate a detailed inspection report. The report includes the location, type, degree, development trend forecast of the damage, and recommended maintenance and repair measures. The final inspection report provides users with comprehensive damage information and suggestions to help users make scientific maintenance decisions.

[0114] Example 2

[0115] As a preferred example, the specific steps of S1 include:

[0116] Select an operating frequency that matches the material characteristics and depth of the bridge pavement to ensure that radar waves penetrate and are reflected effectively; configure the radar transmitter and receiver according to the selected operating frequency, set an appropriate pulse repetition frequency and signal bandwidth, and fix the radar antenna on the scanning device, which is used to move along a predetermined path on the bridge pavement;

[0117] Start the scanning device so that the radar antenna scans the bridge pavement point by point in a straight line or grid pattern to achieve full coverage; during the scanning process, continuously transmit radar signals and receive echo signals reflected from the bridge pavement;

[0118] Using time delay and amplitude information, the echo signal of each scanning point is recorded to generate a radar signal distribution map of the bridge pavement;

[0119] The obtained radar signal distribution map is integrated to form complete radar echo data.

[0120] Specifically, S1.1, select an operating frequency that matches the material characteristics and depth of the bridge pavement. In this embodiment, the optimal frequency range suitable for different types of pavement materials is pre-determined through laboratory testing and data analysis. For example, for a concrete pavement, a frequency range of 1 GHz to 3 GHz is selected to ensure that radar waves can effectively penetrate and be reflected;

[0121] S1.2. Configure the radar transmitter and receiver according to the operating frequency selected in S1.1, and set the appropriate pulse repetition frequency (PRI) and signal bandwidth to maximize the transmission efficiency and resolution of the signal. The setting of the pulse repetition frequency takes into account the thickness of the bridge pavement and the propagation speed of the radar wave, and the setting of the signal bandwidth is used to ensure sufficient frequency resolution to distinguish small signal changes;

[0122] S1.3. The radar antenna is fixed to a scanning device, which is designed to move along a predetermined path on the bridge pavement. The scanning device can be a robotic arm, a track system or other automated platform to ensure that the radar antenna maintains a constant distance from the pavement surface to obtain consistent scanning quality;

[0123] S1.4. Start the scanning equipment so that the radar antenna scans the bridge pavement point by point in a straight line or grid pattern to achieve full coverage. The selection of scanning mode is based on the layout and size of the bridge pavement. The grid pattern is suitable for large or irregularly shaped bridges, while the straight line pattern is suitable for smaller or regularly shaped bridges.

[0124] S1.5. During the scanning process, the radar signal is continuously transmitted and the echo signal reflected from the bridge pavement is received. The radar system is designed to transmit and receive signals synchronously to ensure the timing control of the signal and data synchronization;

[0125] S1.6. Using the time delay and amplitude information, record the echo signal of each scanning point and generate the radar signal distribution map of the bridge pavement. The time delay is used to calculate the propagation depth of the radar signal in the pavement, and the amplitude information reflects the characteristics of the pavement material and possible damage.

[0126] S1.7. Perform data integration on the obtained radar signal distribution diagram to form complete radar echo data. Data integration includes signal digitization, storage and preprocessing to facilitate subsequent signal processing and damage analysis.

[0127] Through the above implementation modes, the present invention can efficiently and accurately collect radar echo data of the bridge pavement layer, providing a solid foundation for subsequent damage detection and analysis.

[0128] Example 3

[0129] As a preferred example, step S2 specifically includes:

[0130] Import the collected radar echo data into digital signal processing software with signal denoising and filtering functions;

[0131] Using the imported radar echo data, perform fast Fourier transform to convert the radar echo data time domain signal into frequency domain signal;

[0132] In the resulting frequency domain signal, the noise frequency components other than radar signals, including abnormal peaks or background noise far from the main frequency peaks, are identified by visual inspection or automatic algorithms;

[0133] Applying a preset band-stop filter to the frequency domain signal to suppress the noise frequency component, and then performing an inverse fast Fourier transform (IFFT) to convert the processed frequency domain signal back to the time domain to obtain a denoised signal;

[0134] Applying a median filter to the resulting denoised signal further eliminates any remaining impulse noise and obtains a denoised signal;

[0135] Perform amplitude correction and gain adjustment on the signal after denoising to ensure that the amplitude of the signal after denoising is consistent with the original radar echo data, and the damage features remain clearly visible, so as to obtain the processed signal;

[0136] Perform quality assessment on the processed signal to verify the improvement in signal-to-noise ratio and ensure that the signal meets the accuracy and clarity standards required for damage detection to obtain a clear radar signal.

[0137] Specifically, S2.1, importing the radar echo data collected in step S1 into digital signal processing software, wherein the software has advanced algorithms and a user interface, allowing a user to visualize and process the data;

[0138] S2.2. Using the imported radar echo data, perform fast Fourier transform (FFT) to convert the time domain signal into the frequency domain signal. The FFT is performed using an efficient algorithm to reduce the calculation time and improve the conversion accuracy.

[0139] S2.3, in the resulting frequency domain signal, identifying noise frequency components that are not radar signals by visual inspection or an automated algorithm, where the automated algorithm may be a preset threshold detection or a machine learning model to identify and classify noise in the signal;

[0140] S2.4, applying a preset band-stop filter to the frequency domain signal, wherein the band-stop filter is designed to suppress the noise frequency component identified in S2.3, and the filter parameters can be adjusted according to the noise characteristics;

[0141] S2.5, performing an inverse fast Fourier transform (IFFT) to convert the frequency domain signal processed by the band-stop filter back to the time domain to obtain a denoised signal. The execution of IFFT ensures that the time domain characteristics of the signal are preserved;

[0142] S2.6, applying a median filter to the obtained denoised signal to further eliminate any remaining impulse noise, the window size of the median filter is selected according to the signal characteristics and the expected filtering effect;

[0143] S2.7, performing amplitude correction and gain adjustment on the signal after denoising. By comparing the amplitude distribution of the original signal and the denoised signal, the gain is adjusted to ensure that the amplitude of the signal is consistent with the original data while maintaining the clarity of the damage characteristics;

[0144] S2.8. Perform quality assessment on the processed signal and verify the effect of signal processing by calculating indicators such as the signal-to-noise ratio (SNR) to ensure that the signal meets the accuracy and clarity standards required for damage detection, obtain a clear radar signal, and provide high-quality data for subsequent damage feature extraction and analysis.

[0145] Through the above implementation modes, the present invention can effectively remove noise while maintaining signal integrity, thereby improving the accuracy and reliability of damage detection.

[0146] Example 4

[0147] As a preferred example, step S3 specifically includes:

[0148] Extracting characteristic parameters of the clear radar signal, including amplitude and arrival time, from the obtained clear radar signal;

[0149] The extracted characteristic parameters are compared with preset normal state parameters, where the preset parameters are based on the radar signal characteristics under an undamaged condition;

[0150] When the extracted amplitude variation exceeds a preset amplitude threshold, or the difference between the signal arrival time and the preset normal arrival time exceeds a preset time threshold, it is determined that there is a sign of damage;

[0151] If it is determined that there is a sign of damage, the characteristic parameters of the sign of damage are recorded and the process proceeds to step S4;

[0152] If it is determined that there is no sign of damage, the current bridge pavement state is considered normal and the current state parameters are recorded;

[0153] The recorded condition parameters will be used for comparison with future inspection results to monitor potential changes in the condition of the bridge pavement;

[0154] After confirming that the status is normal, schedule the next scan according to the predetermined period to continuously monitor the status of the bridge pavement;

[0155] When the detection cycle ends, the detection process automatically restarts from step S1 to ensure continuous and periodic monitoring.

[0156] Specifically, S3.1, extracting characteristic parameters of the radar signal, including amplitude and arrival time, from the obtained clear radar signal; the extraction process adopts advanced signal processing technology, such as time-frequency analysis or short-time Fourier transform, to ensure the accuracy and reliability of the parameters.

[0157] S3.2. Compare the characteristic parameters extracted in S3.1 with preset normal state parameters; the normal state parameters are derived from a radar signal characteristic database under no damage conditions, which is established through historical data and statistical analysis to provide a benchmark for damage judgment.

[0158] S3.3. During the comparison process, when the extracted amplitude change exceeds the preset amplitude threshold, or the difference between the signal arrival time and the preset normal arrival time exceeds the preset time threshold, it is determined that there are signs of damage; the threshold is determined based on material properties, environmental factors and signal propagation theory to ensure the sensitivity and specificity of damage judgment.

[0159] S3.4: If it is determined in S3.3 that there are signs of damage, the characteristic parameters of the damage signs are recorded, including the damage location, amplitude change and signal arrival time difference, and the process proceeds to step S4 for further damage feature extraction and analysis.

[0160] S3.5. If it is determined in S3.3 that there is no sign of damage, the current state of the bridge pavement is considered normal, and the current state parameters, including the amplitude and arrival time of the radar signal, are recorded to provide a reference for the health status of the bridge pavement.

[0161] S3.6. The recorded status parameters will be used for comparison with future test results. By analyzing the trend of parameter changes and monitoring the potential changes in the status of the bridge pavement, long-term tracking and management of the health status of the bridge can be achieved.

[0162] S3.7. After confirming that the status is normal, arrange the next scan according to the predetermined period based on the use of the bridge and environmental conditions, and continuously monitor the status of the bridge pavement to ensure the safety and reliability of the bridge.

[0163] S3.8. When the inspection cycle ends, the inspection process will automatically restart from step S1. Through the automated control and scheduling system, continuous and periodic monitoring can be ensured to achieve real-time monitoring and management of bridge pavement damage.

[0164] Through the above-mentioned implementation mode, the present invention can accurately determine whether there is damage to the bridge pavement layer, and realize periodic monitoring and management of the health status of the bridge, effectively preventing and reducing the occurrence of bridge safety accidents.

[0165] Example 5

[0166] As a preferred example, step S4 specifically includes:

[0167] receiving a radar signal with a determined damage indication as an input for waveform encoding feature extraction;

[0168] Perform time-frequency analysis on radar signals to identify abnormal patterns in radar signals, which indicate damaged areas;

[0169] Using the identified abnormal pattern, the radar signal is encoded using waveform coding technology to obtain an encoded signal;

[0170] The encoded signal is subjected to feature extraction to extract parameters related to the damage, including frequency offset and amplitude change, to obtain the extracted damage features.

[0171] Specifically, S4.1, receiving the radar signal with damage signs determined in step S3 as input for waveform coding feature extraction; the radar signal includes characteristic parameters such as amplitude changes and signal arrival time differences identified in step S3, providing basic data for damage feature extraction.

[0172] S4.2. Perform time-frequency analysis on the radar signal to identify abnormal patterns in the radar signal; the time-frequency analysis uses techniques such as wavelet transform or short-time Fourier transform to identify abnormal patterns in the signal that indicate damaged areas, such as sudden changes in frequency components or abnormal peaks in amplitude.

[0173] S4.3. Using the abnormal pattern identified in S4.2, waveform coding technology is applied to encode the radar signal; the waveform coding technology includes but is not limited to Hilbert-Huang transform, empirical mode decomposition or Wigner-Ville distribution, etc., to enhance the identifiability of damage characteristics.

[0174] S4.4. Perform feature extraction on the signal encoded in S4.3 to extract parameters related to the damage; the parameters include frequency offset, amplitude change, phase change, etc. These parameters can quantitatively describe the characteristics and degree of the damage.

[0175] S4.5. Classify and identify the damage type by combining the characteristic parameters extracted by S4.4 with a machine learning algorithm or an expert system; the algorithm or system realizes automatic matching and identification of damage characteristics based on a pre-trained damage pattern database.

[0176] S4.6. Based on the damage type identified in S4.5, further analyze the severity and development trend of the damage; evaluate the impact of the damage on the safety of the bridge structure by comparing the extracted characteristic parameters with the threshold or model of the damage severity.

[0177] S4.7. Integrate the damage characteristics, damage type, severity and development trend information extracted from S4.4 and S4.6 to form a detailed damage characteristics report; the report provides a scientific basis for subsequent damage assessment and maintenance decisions.

[0178] Through the above implementation modes, the present invention can effectively extract damage features from radar signals, realize accurate identification and evaluation of bridge pavement damage, and provide accurate data support for bridge maintenance and management.

[0179] Example 6

[0180] As a preferred example, step S5 specifically includes:

[0181] Extract key parameters from the resulting damage characterization report, including quantitative indicators related to the damage;

[0182] Access and retrieve a pre-established normal state database, which contains radar signal characteristics of bridge pavement without damage; compare and analyze the extracted key parameters with the parameters of the retrieved normal state database one by one;

[0183] Based on the comparison results, determine whether there are parameters that deviate significantly from the normal state, which serves as a preliminary basis for judging the existence of damage;

[0184] If the damage is initially confirmed to exist, the damage characteristic parameters are further analyzed and matched with known damage types in the database;

[0185] Using the matching results, determine the type and possible cause of the injury;

[0186] Record the identified damage type and location, as well as the degree of deviation from the normal state, and form a damage assessment report.

[0187] Specifically, S5.1, extract key parameters from the damage characteristic report obtained in step S4.7. These parameters include but are not limited to quantitative indicators such as frequency offset, amplitude change, phase change, etc. These indicators can quantitatively describe the characteristics and degree of the damage.

[0188] S5.2. Access and retrieve a pre-established normal state database containing radar signal characteristics of the bridge pavement in an undamaged state. These characteristics serve as a benchmark for assessing damage. The database is established through historical data collection and analysis.

[0189] S5.3. Compare and analyze the key parameters extracted in S5.1 with the parameters of the normal state database retrieved in S5.2 one by one. The comparative analysis uses statistical methods or pattern recognition technology to ensure the accuracy and repeatability of the analysis.

[0190] S5.4. Based on the comparison results in S5.3, determine whether there are parameters that deviate significantly from the normal state. If so, it will be used as a preliminary basis for judging the existence of damage. The threshold of significant deviation is determined based on engineering judgment and statistical analysis.

[0191] S5.5. If the preliminary judgment in S5.4 confirms the existence of damage, the damage characteristic parameters are further analyzed and matched with the known damage types in the database. The matching process uses machine learning algorithms, such as support vector machines (SVM) or decision trees, to improve the accuracy of damage type identification.

[0192] S5.6. Using the matching results in S5.5, determine the type of damage and possible causes. Damage types include but are not limited to cracks, voids, honeycombs, etc. Cause analysis takes into account material aging, environmental factors, usage loads, etc.

[0193] S5.7. Record the damage type, location, and degree of deviation from normal conditions determined in S5.6 and prepare a damage assessment report that describes the specific circumstances of the damage in detail to provide decision support for bridge maintenance and management.

[0194] S5.8. The damage assessment report further includes a damage severity assessment and repair recommendations, the severity assessment is based on comparing damage parameters with predefined severity standards, and the repair recommendations are made based on the damage type and severity.

[0195] Through the above-mentioned implementation modes, the present invention can accurately identify and evaluate the damage of the bridge pavement layer, and provide a scientific and accurate basis for the maintenance and management of the bridge.

[0196] Example 7

[0197] As a preferred example, step S6 specifically includes:

[0198] Using the damage location and type information obtained in the step, determine the area that needs to be accurately detected;

[0199] Select the frequency modulation parameters suitable for the damage type and set the starting frequency f start and end frequency f end ;

[0200] A frequency modulator is used to generate a linearly changing frequency sweep signal. The relationship between frequency and time is:

[0201] f(t)=f start +β·t, where β is the rate of change of frequency;

[0202] The generated frequency sweep signal is transmitted to the damaged area, and the timestamp t of the reflected signal is recorded. r ;

[0203] Calculate the damage depth d using d = v t r / 2, where v is the propagation speed of radar waves in the medium;

[0204] Determine the frequency change Δf of the reflected signal. If there is a frequency change, calculate the two-dimensional size change Δr of the damaged area using the formula Δr = c·Δf / f center 2 , where c is the speed of light, f center is the center frequency of the radar signal.

[0205] Specifically, S6.1, using the damage location and type information obtained in step S5.7, the area that needs to be accurately detected is accurately demarcated through a geographic information system (GIS) or similar technology. This step ensures the pertinence and efficiency of the detection work.

[0206] S6.2. Select appropriate frequency modulation parameters according to the type of injury. For example, for deep injuries, select a lower starting frequency f start For better penetration, select a higher frequency for surface damage to improve resolution. End frequency f end The setting takes into account the thickness of the bridge pavement and the attenuation characteristics of the radar waves.

[0207] S6.3, use the frequency modulator to generate a linear frequency sweep signal. In this process, the frequency increases linearly with time, and the formula is f(t) = f start +β·t, where β is the rate of change of frequency, adjusted according to the required detection depth and resolution.

[0208] S6.4, the generated frequency scanning signal is transmitted to the damaged area. The signal is sent by the radar transmitter and coupled into the bridge pavement through the antenna. The reflected signal is received by the same antenna and the timestamp is t r Record the return time of the reflected signal to provide a basis for calculating the damage depth.

[0209] S6.5. Calculate the damage depth d. According to the propagation speed v of the radar wave in the medium and the timestamp tr of the reflected signal, the damage depth is calculated using the formula d=v·tr / 2. In this step, the propagation speed v is predetermined according to the material properties of the bridge pavement.

[0210] S6.6. Determine the frequency change Δf of the reflected signal. The frequency change is determined by analyzing the frequency content of the reflected signal, which reflects the scattering characteristics of the damaged area to the radar wave.

[0211] S6.7. If there is a frequency change, calculate the two-dimensional size change Δr of the damaged area. Use the formula Δr = c.Δf / f center^2Calculate the size change, where c is the speed of light, f center is the center frequency of the radar signal. This step provides the extension of the damaged area in the plane.

[0212] S6.8. Through the above steps, the precise location, depth and size information of the damage are obtained, providing detailed and accurate data support for damage assessment and repair decisions.

[0213] Example 8

[0214] As a preferred example, the specific steps of step S7 include:

[0215] Receive accurate detection results of current damage, including location, depth and size of the damage;

[0216] The precise detection results are fed into the pre-trained intelligent damage evolution tracking algorithm;

[0217] Intelligent damage evolution tracking algorithm analyzes the similarity between current damage characteristics and historical damage cases to determine damage type and development stage;

[0218] Based on the analysis results, the intelligent damage evolution tracking algorithm uses the mathematical model established by v to predict the path and speed of damage expansion;

[0219] The intelligent damage evolution tracking algorithm provides the predicted status of damage at a specific point in the future based on the prediction results, combined with environmental factors and bridge usage conditions;

[0220] Combined with the predicted status, the potential impact of damage on the safety of the bridge structure is evaluated and the prediction results are obtained.

[0221] Specifically, S7.1, receive the accurate detection results of the current damage obtained in step S6.8, including the location, depth and size of the damage. These results provide initial input data for the intelligent damage evolution tracking algorithm, ensuring the accuracy of the algorithm analysis.

[0222] S7.2. Input the precise detection results in S7.1 into a pre-trained intelligent damage evolution tracking algorithm, which has been trained to recognize and predict the evolution pattern of damage through machine learning techniques such as neural networks or support vector machines.

[0223] S7.3, Intelligent Damage Evolution Tracking Algorithm analyzes the similarity between current damage characteristics and historical damage cases to determine the damage type and development stage. The algorithm identifies the damage pattern and development trend by comparing the current damage characteristics with historical data.

[0224] S7.4. Based on the analysis results of S7.3, the intelligent damage evolution tracking algorithm uses established mathematical models to predict the path and speed of damage extension. These models may include time series-based prediction models or physical damage evolution models.

[0225] S7.5, Intelligent Damage Evolution Tracking Algorithm Based on the prediction results of S7.4, combined with environmental factors and bridge usage conditions, it provides a predicted state of damage at a specific point in the future. The algorithm considers environmental factors such as temperature, humidity, traffic load, as well as the frequency of use and load history of the bridge.

[0226] S7.6, combined with the predicted state in S7.5, evaluate the potential impact of the damage on the safety of the bridge structure and obtain a prediction result. This step involves a risk assessment model that compares the predicted damage state with the structural integrity standard of the bridge to determine the potential risk level.

[0227] S7.7. The intelligent damage evolution tracking algorithm outputs a damage prediction report, including the damage expansion path, speed, future state, and impact assessment on bridge safety. The report provides a scientific basis for bridge maintenance and management decisions.

[0228] S7.8. Based on the prediction report in S7.7, if the damage extension rate exceeds the safety threshold or the prediction status indicates an impending structural problem, the algorithm will trigger an early warning mechanism and recommend immediate repair or reinforcement measures.

[0229] Through the above-mentioned implementation modes, the present invention can realize the intelligent prediction and evaluation of bridge pavement damage, and provide an efficient and accurate tool for bridge maintenance and management.

[0230] Example 9

[0231] As a preferred example, the development and training of the intelligent damage evolution tracking algorithm specifically includes the following steps:

[0232] Collect historical damage case data of a series of bridge pavements, including the initial state, development process and final state of the damage;

[0233] Define damage characteristic parameters, including damage area, depth, extension rate, and related environmental factors and bridge use conditions;

[0234] Using the collected and defined data, a comprehensive damage signature database is built to provide input data for algorithm training;

[0235] Choose one of the machine learning models, random forest, gradient boosting machine or deep neural network, as the basis for the intelligent damage evolution tracking algorithm;

[0236] Using the damage signature database, the selected machine learning model is trained to minimize the prediction error by adjusting the model parameters;

[0237] Validate the trained model and use cross-validation or an independent test set to evaluate the model’s predictive performance and generalization ability;

[0238] Based on the verification results, fine-tune the model, optimize the model structure and parameters, and improve the accuracy and reliability of the algorithm;

[0239] The final trained and optimized intelligent damage evolution tracking algorithm is deployed into the damage detection system for real-time analysis and prediction of damage development trends.

[0240] Specifically, S9.1. Collect a series of historical damage case data of bridge pavement, including the initial state, development process and final state of the damage. The data collection process involves on-site inspections, regular monitoring reports and historical maintenance records to ensure the comprehensiveness and diversity of the data.

[0241] S9.2. Define damage characterization parameters, including but not limited to damage area, depth, growth rate, and relevant environmental factors such as temperature, humidity, and bridge usage conditions such as traffic volume and load type. These parameters provide a quantitative description of the damage characteristics.

[0242] S9.3. Using the data collected and defined in S9.1 and S9.2, establish a comprehensive damage signature database. The database provides input data for algorithm training, including radar signal characteristics in normal state and various damage states.

[0243] S9.4. Select an appropriate machine learning model, such as random forest, gradient boosting machine, or deep neural network, as the basis for the intelligent damage evolution tracking algorithm. The model is selected based on its ability to handle time and sequence data, and its ability to model nonlinear relationships.

[0244] S9.5. Use the damage signature database to train the selected machine learning model. During the training process, the model parameters are adjusted to minimize the prediction error to ensure that the model can accurately learn and simulate the damage evolution process.

[0245] S9.6. Validate the trained model and use cross-validation or an independent test set to evaluate the model’s predictive performance and generalization ability. The validation process ensures the model’s reliability and potential for application on unknown data.

[0246] S9.7. Based on the verification results in S9.6, fine-tune the model, optimize the model structure and parameters, and improve the accuracy and reliability of the algorithm. Fine-tuning may include adjusting the number of network layers, the number of neurons, the learning rate, and other learning parameters.

[0247] S9.8. Deploy the final trained and optimized intelligent damage evolution tracking algorithm into the damage detection system. The deployed algorithm can receive new detection data in real time, analyze and predict damage development trends, and provide maintenance and repair suggestions.

[0248] Through the above implementation, the intelligent damage evolution tracking algorithm provided by the present invention can accurately predict the development of bridge pavement damage and provide a scientific basis for bridge maintenance and safety management.

[0249] Example 10

[0250] like Figure 2 As shown, a bridge pavement damage detection system based on ground detection radar includes:

[0251] A data acquisition module is used to scan the bridge pavement using a ground detection radar and collect radar echo data;

[0252] Specifically, the module includes a radar transmitter, a receiver and a positioning system to ensure that the radar antenna accurately scans the bridge pavement according to the predetermined path.

[0253] The signal processing module is used to perform denoising and filtering on the radar echo data collected by the data acquisition module to obtain a clear radar signal; the module adopts digital signal processing technology, including Fourier transform, wavelet transform and band-stop filtering, to improve signal quality.

[0254] The damage sign judgment module is used to judge whether there are signs of damage. If the amplitude change of the radar signal or the signal arrival time exceeds the preset threshold, the radar signal is determined to be a sign of damage. Otherwise, the current state is recorded and returned to the data acquisition module to rescan according to the preset period.

[0255] The feature extraction module is used to extract waveform coding features of detected damage signs and identify damage characteristics; this module applies time-frequency analysis and waveform coding technology to extract damage-related parameters such as frequency offset and amplitude attenuation.

[0256] The damage analysis module is used to compare and analyze the extracted damage features with the normal state database to determine the existence and type of damage; this module uses pattern recognition and machine learning algorithms to classify and identify damage features.

[0257] The damage detection module is used to accurately detect the depth and range of damage using adaptive frequency modulation technology; this module dynamically adjusts the frequency of the radar signal according to the damage characteristics to achieve high-precision detection of the damaged area.

[0258] The damage prediction module is used to predict the damage development trend based on the precise detection results obtained using an intelligent damage evolution tracking algorithm; this module uses a machine learning model to analyze the damage evolution trend and predict the future state.

[0259] The report generation module is used to obtain the inspection report based on the analysis results and prediction results. This module integrates the output of all modules and generates a detailed damage inspection report, including damage location, type, severity, development trend and maintenance recommendations.

[0260] Specifically, the data acquisition module may further include: an antenna positioning submodule, used to ensure that the radar antenna moves accurately along a predetermined path on the bridge pavement; and a signal acquisition submodule, used to receive the echo signal reflected by the radar and perform digital processing.

[0261] The signal processing module may further include: a denoising submodule, which uses a fast Fourier transform (FFT) to perform frequency domain analysis on the signal and uses a band-stop filter to remove noise; a filtering submodule, which uses an inverse fast Fourier transform (IFFT) to convert the signal back to the time domain and applies a median filter to further eliminate pulse noise.

[0262] The damage sign judgment module may further include: a threshold setting submodule, which sets the thresholds of amplitude change and signal arrival time according to historical data and material properties; a damage judgment submodule, which compares the signal characteristics with the thresholds, determines whether there are signs of damage, and records the status for periodic rescanning.

[0263] The feature extraction module may further include: a time-frequency analysis submodule, which uses wavelet transform or short-time Fourier transform to identify abnormal patterns in radar signals; and a feature encoding submodule, which uses waveform coding technology to encode the signal and extract damage feature parameters.

[0264] The damage analysis module may further include: a database matching submodule, which accesses the normal state database and performs comparative analysis with the extracted features; and a damage type determination submodule, which uses a pattern recognition algorithm to determine the existence and type of damage.

[0265] The damage detection module may further include: a frequency modulation submodule, which selects appropriate frequency modulation parameters to generate a linearly changing frequency scanning signal; and a depth detection submodule, which calculates the timestamp of the reflection signal to determine the damage depth and two-dimensional size change.

[0266] The damage prediction module may further include: an evolution tracking submodule, which receives damage detection results and inputs an intelligent damage evolution tracking algorithm for analysis; and a trend prediction submodule, which uses an established mathematical model to predict the path and speed of damage expansion.

[0267] The report generation module can further include a result integration submodule to summarize the output of each module, including damage location, type, severity and development trend; and a report output submodule to generate and output a detailed inspection report and provide maintenance and repair suggestions. Through the collaborative work of the above modules and their submodules, this system can achieve comprehensive, automated and intelligent detection of bridge pavement damage, providing scientific and accurate decision support for bridge maintenance and management.

[0268] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

Claims

1. A bridge pavement damage detection method based on ground detection radar, characterized in that: include, S1. Use ground detection radar to scan the bridge pavement and collect radar echo data; S2, denoising and filtering the radar echo data collected by S1 to obtain a clear radar signal; S3, judging whether there is a sign of damage. If the amplitude change of the radar signal or the signal arrival time exceeds the preset threshold, the radar signal is determined to be a sign of damage and the process goes to S4; otherwise, the current state is recorded and the process goes back to S1, and the scan is repeated according to the preset period. S4, extracting waveform coding features from the detected damage signs and identifying damage features; S5, comparing and analyzing the extracted damage features with the normal state database to determine the existence and type of damage; S6. Based on the analysis results, the adaptive frequency modulation technology is used to accurately detect the damage depth and range; S7. Based on the accurate detection results, the intelligent damage evolution tracking algorithm is used to predict the damage development trend; S8. Obtain a test report based on the analysis results and prediction results.

2. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: The specific steps of S1 include: Select an operating frequency that matches the material characteristics and depth of the bridge pavement to ensure that radar waves penetrate and are reflected effectively; Configure the radar transmitter and receiver according to the selected operating frequency, and set the appropriate pulse repetition frequency and signal bandwidth; The radar antenna is fixed to a scanning device, and the scanning device is used to move along a predetermined path on the bridge pavement; Start the scanning equipment so that the radar antenna scans the bridge pavement point by point in a straight line or grid pattern to achieve full coverage; During the scanning process, radar signals are continuously emitted and echo signals reflected from the bridge pavement are received; Using time delay and amplitude information, the echo signal of each scanning point is recorded to generate a radar signal distribution map of the bridge pavement; The obtained radar signal distribution map is integrated to form complete radar echo data.

3. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: Step S2 specifically includes: Import the collected radar echo data into digital signal processing software with signal denoising and filtering functions; Using the imported radar echo data, perform fast Fourier transform to convert the radar echo data time domain signal into frequency domain signal; In the resulting frequency domain signal, the noise frequency components other than radar signals, including abnormal peaks or background noise far from the main frequency peaks, are identified by visual inspection or automatic algorithms; Applying a preset band-stop filter to the frequency domain signal to suppress the noise frequency component, and then performing an inverse fast Fourier transform to convert the processed frequency domain signal back to the time domain to obtain a denoised signal; Applying a median filter to the resulting denoised signal further eliminates any remaining impulse noise and obtains a denoised signal; Perform amplitude correction and gain adjustment on the signal after denoising to ensure that the amplitude of the signal after denoising is consistent with the original radar echo data, and the damage features remain clearly visible, so as to obtain the processed signal; Perform quality assessment on the processed signal to verify the improvement in signal-to-noise ratio and ensure that the signal meets the accuracy and clarity standards required for damage detection to obtain a clear radar signal.

4. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: Step S3 specifically includes: Extracting characteristic parameters of the clear radar signal, including amplitude and arrival time, from the obtained clear radar signal; The extracted characteristic parameters are compared with preset normal state parameters, where the preset parameters are based on the radar signal characteristics under an undamaged condition; When the extracted amplitude variation exceeds a preset amplitude threshold, or the signal arrival time and the preset normal arrival time exceed a preset time threshold, it is determined that there is a sign of damage; If it is determined that there is a sign of damage, the characteristic parameters of the sign of damage are recorded and the process proceeds to step S4; If it is determined that there is no sign of damage, the current bridge pavement state is considered normal and the current state parameters are recorded; The recorded condition parameters will be used for comparison with future inspection results to monitor potential changes in the condition of the bridge pavement; After confirming that the status is normal, schedule the next scan according to the predetermined period to continuously monitor the status of the bridge pavement; When the detection cycle ends, the detection process automatically restarts from step S1 to ensure continuous and periodic monitoring.

5. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: Step S4 specifically includes: receiving a radar signal with a determined damage indication as an input for waveform encoding feature extraction; Perform time-frequency analysis on radar signals to identify abnormal patterns in radar signals, which indicate damaged areas; Using the identified abnormal pattern, the radar signal is encoded using waveform coding technology to obtain an encoded signal; Feature extraction is performed on the encoded signal to extract parameters related to the damage, including frequency offset and amplitude change, to obtain the extracted damage features.

6. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: Step S5 specifically includes: Extract key parameters from the resulting damage characterization report, including quantitative indicators related to the damage; Access and retrieve a pre-established normal state database containing radar signatures of bridge pavement without damage; Compare and analyze the extracted key parameters with the parameters of the retrieved normal state database one by one; Based on the comparison results, determine whether there are parameters that deviate significantly from the normal state, which serves as a preliminary basis for judging the existence of damage; If the damage is initially confirmed to exist, the damage characteristic parameters are further analyzed and matched with known damage types in the database; Using the matching results, determine the type and possible cause of the injury; Record the identified damage type and location, as well as the degree of deviation from the normal state, and form a damage assessment report.

7. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: Step S6 specifically includes: Using the damage location and type information obtained in the step, determine the area that needs to be accurately detected; Select the frequency modulation parameters suitable for the damage type and set the starting frequency f start and end frequency f end ; A frequency modulator is used to generate a linearly changing frequency sweep signal. The relationship between frequency and time is f(t) = f start +β·t, where β is the rate of change of frequency; The generated frequency sweep signal is transmitted to the damaged area, and the timestamp t of the reflected signal is recorded. r ; Calculate the damage depth d using d = v·t r / 2, where v is the propagation speed of radar waves in the medium; Determine the frequency change Δf of the reflected signal. If there is a frequency change, calculate the two-dimensional size change Δr of the damaged area using the formula Δr = c·Δf / f center 2 , where c is the speed of light, f center is the center frequency of the radar signal.

8. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: The specific steps of step S7 are: include, Receive accurate detection results of current damage, including location, depth and size of the damage; The precise detection results are fed into the pre-trained intelligent damage evolution tracking algorithm; Intelligent damage evolution tracking algorithm analyzes the similarity between current damage characteristics and historical damage cases to determine damage type and development stage; Based on the analysis results, the intelligent damage evolution tracking algorithm uses the established mathematical model to predict the path and speed of damage expansion; The intelligent damage evolution tracking algorithm provides the predicted status of damage at a specific point in the future based on the prediction results, combined with environmental factors and bridge usage conditions; Combined with the predicted status, the potential impact of damage on the safety of the bridge structure is evaluated and the prediction results are obtained.

9. The bridge pavement damage detection method based on ground detection radar according to claim 1 is characterized in that: The development and training of the intelligent damage evolution tracking algorithm, specifically The following steps are included: Collect historical damage case data of a series of bridge pavements, including the initial state, development process and final state of the damage; Define damage characteristic parameters, including damage area, depth, extension rate, and related environmental factors and bridge use conditions; Using the collected and defined data, a comprehensive damage signature database is built to provide input data for algorithm training; Choose one of the machine learning models, random forest, gradient boosting machine or deep neural network, as the basis for the intelligent damage evolution tracking algorithm; Using the damage signature database, the selected machine learning model is trained to minimize the prediction error by adjusting the model parameters; Validate the trained model and use cross-validation or an independent test set to evaluate the model’s predictive performance and generalization ability; Based on the verification results, fine-tune the model, optimize the model structure and parameters, and improve the accuracy and reliability of the algorithm; The final trained and optimized intelligent damage evolution tracking algorithm is deployed into the damage detection system for real-time analysis and prediction of damage development trends.

10. A bridge pavement damage detection system based on ground detection radar, based on a bridge pavement damage detection method based on ground detection radar according to any one of claims 1 to 9, characterized in that: include, A data acquisition module is used to scan the bridge pavement using a ground detection radar and collect radar echo data; A signal processing module, used for performing denoising and filtering on the radar echo data collected by the data acquisition module to obtain a clear radar signal; A damage sign judgment module is used to judge whether there is a damage sign. If the amplitude change or signal arrival time of the radar signal exceeds a preset threshold, the radar signal is determined to be a damage sign; otherwise, the current state is recorded and returned to the data acquisition module to rescan according to a predetermined period; A feature extraction module is used to extract waveform coding features from detected damage signs and identify damage features; A damage analysis module is used to compare and analyze the extracted damage features with the database of normal states to determine the existence and type of damage; Damage detection module, used to accurately detect damage depth and range using adaptive frequency modulation technology; The damage prediction module is used to predict the damage development trend based on the accurate detection results obtained using the intelligent damage evolution tracking algorithm; The report generation module is used to obtain a test report based on the analysis results and prediction results.

Images