A method and system for detecting the thickness of steel bar protective layer during bridge construction

Abstract

The present invention relates to the field of bridge detection, and specifically to a method and system for detecting the thickness of a steel bar protective layer during bridge construction. The method includes using a mobile robot platform to perform a preliminary range scan covering the entire bridge platform to obtain preliminary thickness distribution data; performing multi-angle ultrasonic signal excitation and reception on key areas to collect thickness data of key areas; using a preset deep learning algorithm to identify the thickness of the steel bar protective layer, and recording the thickness data of each measuring point to obtain a processing result; when the processing result identifies an abnormal area, executing an enhanced detection program to perform key detection to obtain a key detection result; comparing the key detection result with a preset protective layer thickness standard to obtain a comparison result; integrating the processing result, key detection result and comparison result to obtain a complete detection report. The present invention improves detection accuracy and efficiency, and realizes comprehensive and rapid preliminary and key detection. The detection speed is improved, and the accuracy of the detection results is improved.

Description

Technical Field

[0001] The present invention relates to the field of bridge detection, and in particular to a method and system for detecting the thickness of a steel bar protective layer during bridge construction. Background Art

[0002] In the field of bridge construction, ensuring that the thickness of the steel bar cover meets the design requirements is crucial for the safety and durability of the structure. Traditional methods, such as manual tapping or drilling sampling, are not only inefficient, but may also cause damage to the structure itself. With the development of technology, non-destructive testing technologies such as electromagnetic acoustic transducers (EMAT) and ground penetrating radar (GPR) have begun to be used in bridge inspection. However, these technologies are usually used independently, lacking effective data fusion and intelligent processing, resulting in incomplete or erroneous test results.

[0003] Comparative document CN210288501U discloses a device for detecting the thickness of the pile body steel bar protective layer, which uses ultrasonic sensors and electromagnetic sensors to detect the pile body, and determines the thickness of the protective layer by calculating the distance difference between the two sensors and the sensor array. Although this method improves the accuracy of detection, it may be limited in actual operation by the complexity of sensor layout and operation, and lacks real-time feedback and adjustment mechanism for abnormal areas.

[0004] Comparative document CN212320625U describes a device for detecting the thickness of the steel bar protective layer of a concrete structure, which includes an overhead conveying rail, a rotating detection cylinder, an electromagnetic detector, and an ultrasonic detector, etc., and can automatically convey the detection and perform laser marking positioning. Although the device improves the detection efficiency, it is mainly designed for concrete plates and may not provide sufficient flexibility and accuracy for the complex and changeable working environment and irregularly shaped bridge structures in bridge construction.

[0005] In view of the limitations of the prior art, the present invention proposes a method and system for detecting the thickness of a steel bar protective layer during bridge construction. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for detecting the thickness of the steel bar protective layer during bridge construction, so as to solve the problems in the prior art of inaccurate control of the thickness of the steel bar protective layer, low detection efficiency, and inability to adjust the construction process in real time.

[0007] To achieve the above purpose, the following technical solution is adopted.

[0008] A method for detecting the thickness of a steel bar protective layer during bridge construction, comprising:

[0009] The mobile robot platform is equipped with an electromagnetic acoustic transducer module and moves along the bridge surface to perform a preliminary range scan covering the entire bridge platform, determine the preliminary thickness distribution of the steel bar protection layer, and obtain preliminary thickness distribution data;

[0010] According to the preliminary thickness distribution results, the multi-degree-of-freedom robotic arm on the mobile robot platform is used to adjust the position and angle of the electromagnetic acoustic transducer module, perform multi-angle ultrasonic signal excitation and reception on the key area, and collect thickness data of the key area.

[0011] Use the preset deep learning algorithm to collect preliminary thickness distribution data and key area thickness data, identify the thickness of the steel bar cover, and record the thickness data of each measuring point to obtain the processing results;

[0012] When the abnormal area is identified by the processing result, the electromagnetic acoustic transducer module is accurately positioned to the abnormal area through the mechanical arm of the mobile robot platform to perform enhanced detection procedures for key detection and obtain key detection results;

[0013] According to the key test results, the thickness of the protective layer is compared with the preset standard to determine whether the actual thickness of the protective layer meets the design requirements and obtain the comparison results;

[0014] Integrate the processing results, key test results and comparison results to obtain a complete test report.

[0015] Optionally, if the comparison result indicates that the actual protective layer thickness does not meet the design requirements, perform the following steps:

[0016] The data analysis unit on the mobile robot platform is used to conduct in-depth analysis of the comparison results to identify specific areas that do not meet the design requirements and the degree of deviation;

[0017] According to the analysis results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is automatically or manually adjusted to reposition the electromagnetic acoustic transducer module to the non-compliant area;

[0018] Dynamically adjust the construction process based on the real-time detection data of the electromagnetic acoustic transducer module and the results of the enhanced detection program, including but not limited to adjusting the steel bar layout, concrete mix or pouring technology, and correcting thickness deviations;

[0019] After adjusting the construction process, the electromagnetic acoustic transducer module carried by the mobile robot platform is used to detect the adjusted area again to verify whether the thickness of the protective layer meets the predetermined standard;

[0020] If the adjusted key test results still do not meet the design requirements, continue to iteratively adjust the construction process based on the real-time detection data of the electromagnetic acoustic transducer module until the actual protective layer thickness meets the design requirements;

[0021] After each adjustment and re-test, the complete test report is updated to record all test data, adjustment measures and verification results to form a complete quality control record of the bridge cover thickness.

[0022] Optionally, during the initial range scan and key inspection of the entire bridge platform, ground penetrating radar technology is used to perform auxiliary scanning of the bridge structure to verify the detection results of the electromagnetic acoustic transducer module and provide more information about the interior of the concrete structure;

[0023] The detection data obtained by combining the electromagnetic acoustic transducer module and the ground penetrating radar technology are fused and processed through a fusion algorithm to obtain fused data.

[0024] Optionally, a mobile robot platform is used to carry an electromagnetic acoustic transducer module and move along the bridge surface to perform a preliminary range scan covering the entire bridge platform to determine the preliminary thickness distribution of the steel bar protective layer, and the step of obtaining preliminary thickness distribution data specifically includes:

[0025] The electromagnetic acoustic transducer module of the mobile robot platform moves along a predetermined path on the bridge surface, emitting ultrasonic pulses of a specific frequency and recording the emission time;

[0026] The electromagnetic acoustic transducer module receives the ultrasonic signal reflected from the steel bar protective layer and records the receiving time;

[0027] According to the time difference between transmission and reception and the propagation speed of ultrasonic waves in the medium, the total distance that the ultrasonic waves pass through the protective layer is calculated to obtain a preliminary measurement value of the thickness of the protective layer;

[0028] Comparing the obtained preliminary measurement value with the preset calibration parameters of the electromagnetic acoustic transducer module, calibrating the measurement result, and obtaining calibrated protective layer thickness data;

[0029] The obtained calibrated protection layer thickness data is stored in a database of the mobile robot platform, and a preliminary thickness distribution map is generated as preliminary thickness distribution data.

[0030] Optionally, according to the preliminary thickness distribution results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is used to adjust the position and angle of the electromagnetic acoustic transducer module, to perform multi-angle ultrasonic signal excitation and reception on the key area, and to collect the thickness data of the key area, specifically including:

[0031] Analyze preliminary thickness distribution data to identify unclear areas and important areas of the bridge as critical areas, including areas with abnormal cover thickness, structural stress concentration or historically vulnerable areas;

[0032] According to the identified key areas, the physical position of the electromagnetic acoustic transducer module is adjusted using the multi-degree-of-freedom robotic arm carried by the mobile robot platform, and the electromagnetic acoustic transducer module is moved to the predetermined detection point of the key area;

[0033] At each predetermined inspection point, according to the specific characteristics of the key area and the preliminary scanning data, the emission angle and receiving angle of the electromagnetic acoustic transducer module are adjusted to optimize the excitation and reception effects of the ultrasonic signal;

[0034] Excite ultrasonic signals from a set angle, receive signals reflected from the steel bar protective layer, and record the emission and reception time of the ultrasonic waves;

[0035] The recorded emission and reception times are used to calculate the propagation time of the ultrasonic wave in the protective layer, and then the thickness of the protective layer is calculated based on the known propagation speed of the ultrasonic wave in the medium;

[0036] The protective layer thickness measurement results of each critical area are recorded, including the measurement value, location and time, and these data are stored in the database of the mobile robot platform to generate a thickness distribution report for the critical area.

[0037] Optionally, the steps of collecting preliminary thickness distribution data and key area thickness data using a preset deep learning algorithm, identifying the thickness of the steel bar cover, and recording the thickness data of each measuring point to obtain the processing result include:

[0038] De-noising and normalizing the preliminary thickness distribution data and key area thickness data collected from the electromagnetic acoustic transducer module;

[0039] The generative adversarial network structure in the deep learning algorithm is used to extract features from the preprocessed data and automatically identify features related to the thickness of the steel bar cover;

[0040] The deep learning model is trained using historical thickness data and known bridge structure parameters to obtain a trained deep learning model;

[0041] The trained deep learning model is used to analyze the data after feature extraction and analyze the thickness of the steel bar cover at each measurement point;

[0042] Using the anomaly detection mechanism in the deep learning model, outliers in the data are identified and possible areas of steel bar cover defects are marked as abnormal areas.

[0043] Optionally, when the abnormal area is identified by the processing result, the electromagnetic acoustic transducer module is accurately positioned to the abnormal area by the mechanical arm of the mobile robot platform to perform an enhanced detection program for key detection, and the step of obtaining the key detection result specifically includes:

[0044] When the abnormal area is identified by the processing results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is operated to adjust the position and angle of the electromagnetic acoustic transducer module according to the position information of the abnormal area to ensure that the electromagnetic acoustic transducer module is accurately aligned with the abnormal area;

[0045] Perform enhanced detection procedures on the located abnormal areas, including increasing the excitation frequency and receiving sensitivity of the ultrasonic signal to obtain more accurate thickness data;

[0046] Collect detailed thickness data of abnormal areas through enhanced detection procedures, including but not limited to the propagation time and reflection intensity of ultrasonic waves in the protective layer;

[0047] The preset deep learning model is used to analyze the data collected by the enhanced detection, identify the actual thickness of the steel bar protection layer and potential structural defects, and obtain key detection results.

[0048] Optionally, the enhanced detection procedure further includes the following steps:

[0049] Ultrasonic signal optimization, adjusting the emission parameters of the electromagnetic acoustic transducer module, including the frequency, amplitude and pulse width of the ultrasonic wave, to adapt to the specific material characteristics and structural conditions of the abnormal area;

[0050] Synchronously collect ultrasonic signals from the electromagnetic acoustic transducer module and radar signals from the ground penetrating radar to perform multi-modal data fusion;

[0051] Adopting adaptive dynamic focusing algorithm, the focusing depth of signal processing is dynamically adjusted according to the attenuation and distortion of ultrasonic signals during propagation;

[0052] Use pre-trained convolutional neural networks to perform feature-level fusion of ultrasonic and ground-penetrating radar radar signals to automatically identify and classify features of abnormal areas, including cracks, voids, and protective layer peeling;

[0053] According to the results of AI-assisted analysis, the detection parameters of the electromagnetic acoustic transducer module and ground-penetrating radar are adjusted in real time to optimize the subsequent detection process;

[0054] Conduct multiple inspections of the abnormal area from different angles, use stereo imaging technology to reconstruct the three-dimensional structure of the abnormal area, and comprehensively evaluate the thickness and integrity of the protective layer;

[0055] Automatically adjust detection parameters based on real-time environmental monitoring data, including temperature, humidity and wind speed;

[0056] All data, analysis results and adjustment parameters of enhanced testing shall be recorded to form an enhanced testing report.

[0057] A steel bar protective layer thickness detection system during bridge construction, comprising:

[0058] A mobile robot platform, equipped with an electromagnetic acoustic transducer module, is used to move along the bridge surface and to excite and receive ultrasonic signals;

[0059] A multi-degree-of-freedom robotic arm connected to the mobile robot platform, used to adjust the position and angle of the electromagnetic acoustic transducer module to accurately detect the designated area;

[0060] A data acquisition unit, integrated with the electromagnetic acoustic transducer module and the multi-degree-of-freedom robotic arm, for collecting excitation and reception data of ultrasonic signals;

[0061] A deep learning processing unit, configured with a preset deep learning algorithm, for processing the collected data, identifying the thickness of the steel bar cover, and recording the thickness data of each measuring point;

[0062] An enhanced detection control unit, for guiding the multi-degree-of-freedom robotic arm to accurately position the electromagnetic acoustic transducer module to the abnormal area and execute an enhanced detection procedure when an abnormal area is identified;

[0063] The data recording and report generation unit is used to record all data, analysis results and adjustment parameters of the enhanced detection, and generate an enhanced detection report.

[0064] Optionally, the enhanced detection control unit includes:

[0065] A signal optimization module, which is used to adjust the emission parameters of the electromagnetic acoustic transducer module, including the frequency, amplitude and pulse width of the ultrasonic wave, to adapt to the specific material characteristics and structural conditions of the abnormal area;

[0066] A multimodal data fusion module, used for synchronously collecting ultrasonic signals from the electromagnetic acoustic transducer module and radar signals from the ground penetrating radar, and performing data fusion processing;

[0067] A dynamic focusing control module is used to dynamically adjust the focusing depth of signal processing according to the attenuation and distortion of the ultrasonic signal during propagation by using an adaptive dynamic focusing algorithm;

[0068] A feature recognition module that uses a pre-trained convolutional neural network to perform feature-level fusion of ultrasonic and ground-penetrating radar signals to automatically identify and classify features of abnormal areas, including cracks, voids, and protective layer peeling;

[0069] A real-time parameter adjustment module, used to adjust the detection parameters of the electromagnetic acoustic transducer module and the ground penetrating radar in real time according to the analysis results of the feature recognition module, so as to optimize the subsequent detection process;

[0070] Stereoscopic imaging module, used to conduct multiple inspections of abnormal areas from different angles, reconstruct the three-dimensional structure of the abnormal area using stereoscopic imaging technology, and comprehensively evaluate the thickness and integrity of the protective layer;

[0071] Environmental adaptability module, used to automatically adjust detection parameters based on real-time environmental monitoring data, including temperature, humidity and wind speed.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] Improve detection accuracy and efficiency. The present invention solves the problems of low efficiency and possible damage to the structure of traditional detection methods. The present invention uses a mobile robot platform equipped with an EMAT module to perform a range scan covering the entire bridge platform, and combines a multi-degree-of-freedom robotic arm to accurately detect key areas, thereby achieving comprehensive and rapid preliminary and key detection. This automated detection process not only improves the detection speed, but also improves the accuracy of the detection results by processing the collected data through a preset deep learning algorithm.

[0074] Another beneficial effect of the present invention is intelligent data processing. The preliminary thickness distribution data and key area thickness data are processed by a deep learning algorithm to automatically identify the thickness of the steel bar cover and record the data of each measurement point. This intelligent processing not only reduces manual operation errors, but also improves the efficiency and accuracy of data processing.

[0075] When the test results show that the actual protective layer thickness does not meet the design requirements, the present invention can use the data analysis unit on the mobile robot platform to conduct in-depth analysis and dynamically adjust the construction process. This real-time feedback and adjustment mechanism enables the construction team to respond quickly to the test results and correct the thickness deviation in time, thereby improving the construction quality and efficiency.

[0076] The accuracy and reliability of the detection system are further enhanced. By using EMAT and GPR technology simultaneously and combining the data obtained by fusion algorithm processing, more comprehensive internal information of the structure is provided. This multimodal data fusion technology makes the detection results more comprehensive and accurate, providing a more reliable basis for construction adjustments.

[0077] The present invention also includes an environmental adaptability module that can automatically adjust the detection parameters according to real-time environmental monitoring data. This environmental adaptability ensures the stability and accuracy of the detection results under different environmental conditions. At the same time, the automated parameter adjustment reduces the need for manual operation and improves the intelligence level of the system.

[0078] In summary, the present invention not only improves the accuracy and efficiency of steel bar protection layer thickness detection during bridge construction, but also provides strong support for construction quality control by integrating multiple advanced technologies, intelligent data processing and real-time construction adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 The present invention is a schematic diagram of the steps of an embodiment of a method for detecting the thickness of a steel bar protective layer during bridge construction.

[0080] Figure 2 The module diagram of an embodiment of a steel bar protective layer thickness detection system during bridge construction of the present invention. DETAILED DESCRIPTION

[0081] 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.

[0082] 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.

[0083] Example 1

[0084] like Figure 1 As shown in the figure, a method for detecting the thickness of the steel bar protective layer during the bridge construction process uses a mobile robot platform equipped with an electromagnetic acoustic transducer (EMAT) module to perform a preliminary range scan covering the entire bridge platform to determine the preliminary thickness distribution of the steel bar protective layer, and then perform multi-angle ultrasonic signal excitation and reception on the key area to collect the thickness data of the key area. The method also includes using a preset deep learning algorithm to process the collected data, identify the thickness of the steel bar protective layer, and record the thickness data of each measurement point, and finally integrate them to obtain a complete detection report.

[0085] The mobile robot platform equipped with the EMAT module moves along a predetermined path on the bridge surface, emitting ultrasonic pulses of a specific frequency and recording the emission time.

[0086] The EMAT module receives the ultrasonic signal reflected from the steel bar cover and records the receiving time.

[0087] Based on the time difference between transmission and reception and the propagation speed of ultrasound in the medium, the total distance the ultrasound passes through the protective layer is calculated to obtain a preliminary measurement value of the thickness of the protective layer.

[0088] The obtained preliminary measurement value is compared with the preset calibration parameters of the EMAT module, the measurement result is calibrated, and the calibrated protection layer thickness data is obtained.

[0089] The obtained calibrated protection layer thickness data is stored in a database of the mobile robot platform, and a preliminary thickness distribution map is generated as preliminary thickness distribution data.

[0090] Analyze preliminary thickness distribution data to identify unclear areas and important areas of the bridge as critical areas, including areas with abnormal cover thickness, structural stress concentrations or historically vulnerable areas.

[0091] According to the identified key areas, the physical position of the EMAT module is adjusted using the multi-degree-of-freedom robotic arm carried by the mobile robot, and the EMAT module is moved to the predetermined detection point in the key area.

[0092] At each predetermined detection point, the emission angle and receiving angle of the EMAT module are adjusted according to the specific characteristics of the key area and the preliminary scanning data to optimize the excitation and reception effects of the ultrasonic signal.

[0093] The ultrasonic signal is excited from the set angle, and the signal reflected from the steel bar protective layer is received, and the emission and reception time of the ultrasonic wave is recorded.

[0094] The recorded emission and reception times are used to calculate the propagation time of the ultrasonic wave in the protective layer, and then the thickness of the protective layer is calculated based on the known propagation speed of the ultrasonic wave in the medium.

[0095] The protective layer thickness measurement results of each critical area are recorded, including the measurement value, location and time, and these data are stored in the database of the mobile robot platform to generate a thickness distribution report for the critical area.

[0096] The preliminary thickness distribution data and key area thickness data collected from the EMAT module are denoised and normalized.

[0097] The generative adversarial network structure in the deep learning algorithm is used to extract features from the preprocessed data and automatically identify features related to the thickness of the steel bar cover.

[0098] The deep learning model is trained using historical thickness data and known bridge structure parameters to obtain a trained deep learning model.

[0099] The trained deep learning model is used to analyze the data after feature extraction and analyze the thickness of the steel bar protective layer at each measuring point.

[0100] Using the anomaly detection mechanism in the deep learning model, outliers in the data are identified and possible areas of steel bar cover defects are marked as abnormal areas.

[0101] When the abnormal area is identified in the processing results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is operated to adjust the position and angle of the EMAT module according to the position information of the abnormal area to ensure that the EMAT module is accurately aligned with the abnormal area.

[0102] An enhanced detection procedure is performed on the located abnormal area, including increasing the excitation frequency and receiving sensitivity of the ultrasonic signal to obtain more accurate thickness data.

[0103] The enhanced detection procedure collects detailed thickness data of abnormal areas, including but not limited to the propagation time and reflection intensity of ultrasonic waves in the protective layer.

[0104] The preset deep learning model is used to analyze the data collected by the enhanced detection, identify the actual thickness of the steel bar protection layer and potential structural defects, and obtain key detection results.

[0105] According to the key test results, it is compared with the preset protective layer thickness standard to determine whether the actual protective layer thickness meets the design requirements and obtain the comparison result.

[0106] Integrate processing results, key test results and comparison results to obtain a complete test report

[0107] Example 2

[0108] The present invention takes the following steps to conduct in-depth analysis and dynamically adjust the construction process:

[0109] The comparison results are analyzed in depth using the data analysis unit on the mobile robot platform to identify specific areas that do not meet the design requirements and the degree of deviation. This step involves statistical analysis of the collected data to determine the distribution and degree of deviation.

[0110] Based on the analysis results, the multi-degree-of-freedom robotic arm on the mobile robot platform is automatically or manually adjusted to reposition the electromagnetic acoustic transducer module to the non-compliant area. The adjustment process may involve programming the motion path of the robotic arm to ensure that the EMAT module is accurately positioned in the critical area.

[0111] Based on the real-time detection data from the electromagnetic acoustic transducer module and the results of the enhanced detection program, the construction process is dynamically adjusted. This may include adjusting the reinforcement layout, concrete mix or pouring technology to correct thickness deviations. The determination of adjustment measures is based on the analysis of real-time data and the compliance assessment of construction standards.

[0112] After adjusting the construction process, the electromagnetic acoustic transducer module carried by the mobile robot platform is used again to detect the adjusted area to verify whether the thickness of the protective layer meets the predetermined standard. This step ensures the effectiveness of the construction adjustment and provides a basis for subsequent quality control.

[0113] If the adjusted key test results still do not meet the design requirements, continue to iteratively adjust the construction process based on the real-time detection data of the electromagnetic acoustic transducer module until the actual protective layer thickness meets the design requirements. This process may involve multiple iterations until all quality standards are met.

[0114] After each adjustment and re-test, the complete test report is updated to record all test data, adjustment measures and verification results to form a complete quality control record of the bridge cover thickness. Report updates ensure transparency and traceability of the construction process.

[0115] Through the above steps, the present invention not only provides an accurate method for detecting the thickness of the steel bar protective layer during bridge construction, but also provides a construction adjustment and quality control mechanism based on real-time data, which ensures the construction quality and meets the bridge design and safety requirements.

[0116] Example 3

[0117] In the present invention, the steel bar protection layer thickness detection method during bridge construction combines electromagnetic acoustic transducer (EMAT) technology and ground penetrating radar (GPR) technology to achieve comprehensive detection of bridge structures. The following is a detailed description of the specific implementation steps, including:

[0118] The mobile robot platform equipped with the EMAT module moves along the predetermined path on the bridge surface, and simultaneously activates the GPR device integrated with the robot platform to ensure that the scanning coverage areas of the EMAT and GPR are completely overlapped. The EMAT module emits ultrasonic pulses of a specific frequency and records the emission time t emit , and GPR equipment transmits high-frequency electromagnetic waves and receives reflected signals.

[0119] The ultrasonic signals generated by the EMAT module and the radar signals generated by the GPR device are collected and used for subsequent data fusion processing. The propagation time Δt of the ultrasonic wave in the protective layer can be calculated by the following formula:

[0120] Δt=t receive -t emit where t receive It is the receiving time.

[0121] The EMAT and GPR data are fused at the feature level using a deep learning-based fusion algorithm. The feature map size calculation formula in the convolutional neural network (CNN) is: w′=s(w+2p-k) / s+1, where w is the width of the input feature map, p is the number of zero-padding layers, k is the convolution kernel size, and s is the stride.

[0122] The EMAT data is analyzed to determine the propagation time of the ultrasonic wave in the protective layer, and then the thickness d of the protective layer is calculated: d = (Δt×v) / 2, where v is the propagation speed of the ultrasonic wave in the medium.

[0123] Combining the high resolution of EMAT technology and the deep detection capability of GPR technology, the detection results are optimized through a fusion algorithm to achieve accurate assessment of the thickness of the steel bar cover.

[0124] Compare the fused data with the individual EMAT test results to verify the accuracy of the fused data. If necessary, adjust the transmission frequency of the EMAT module or the scanning parameters of the GPR device to optimize the subsequent inspection process.

[0125] Pre-trained deep learning models, such as convolutional neural networks (CNNs), are used to extract features and automatically identify fused data to improve the accuracy of assessment of steel bar cover thickness.

[0126] The processing results, verification results and adjustment parameters of the fused data are documented and a detailed enhanced inspection report is generated, providing a record for complete quality control of the bridge cover thickness.

[0127] Through the above specific implementation methods, the present invention provides an accurate method for detecting the thickness of the steel bar protective layer during bridge construction. By combining EMAT and GPR technologies and using deep learning algorithms for data fusion and analysis, the construction quality is ensured and the bridge design and safety requirements are met.

[0128] Example 4

[0129] In the method for detecting the thickness of the steel bar protective layer during the bridge construction process of the present invention, a mobile robot platform is used to carry an electromagnetic acoustic transducer (EMAT) module to move along the surface of the bridge, and a range scan that initially covers the entire bridge platform is performed to determine the preliminary thickness distribution of the steel bar protective layer, and the specific steps of obtaining preliminary thickness distribution data are performed.

[0130] The mobile robot platform was deployed on the bridge, and the EMAT module on board was configured to move along a predetermined path on the bridge surface. The path was pre-planned according to the design drawings and construction requirements of the bridge to ensure that all critical areas were covered.

[0131] The EMAT module emits an ultrasonic pulse of a specific frequency. This frequency is selected based on the characteristics of the bridge material and the expected thickness of the protective layer to ensure optimal signal penetration and reflection. The frequency of the emitted ultrasonic pulse is f0 and the emission time is t emit .

[0132] The EMAT module receives the ultrasonic signal reflected from the steel bar protective layer and records the receiving time t receive The received signal strength and time will be used for subsequent data analysis.

[0133] According to the time difference between transmission and reception Δt=t receive -t emit The total distance that the ultrasonic wave passes through the protective layer is calculated based on the propagation speed v of the ultrasonic wave in the medium. The calculation formula of the protective layer thickness d is: d = (Δt × v) / 2, where v is the propagation speed of the ultrasonic wave in the concrete, which is usually 30m / s to 50m / s, depending on the density and elastic modulus of the concrete.

[0134] The obtained preliminary measurement value is compared with the preset calibration parameters of the EMAT module to calibrate the measurement result. The calibration parameters include system delay, signal attenuation coefficient and temperature compensation factor to ensure the accuracy of the measurement result. The calibrated protective layer thickness d' can be calculated by the following formula: d' = d × (1 + α × ΔT) where α is the temperature coefficient and ΔT is the difference from the reference temperature.

[0135] The calibrated protective layer thickness data d' is stored in the database of the mobile robot platform. These data include the thickness value, position coordinates and measurement time of each measurement point.

[0136] The stored data is used to generate a preliminary thickness distribution map. The distribution map is generated by GIS (Geographic Information System) software or dedicated data analysis software to visually display the thickness distribution of the bridge protective layer.

[0137] Example 5

[0138] In the method for detecting the thickness of the steel bar protective layer during the bridge construction process of the present invention, according to the preliminary thickness distribution results, a multi-degree-of-freedom robotic arm carried by a mobile robot platform is used to adjust the position and angle of the electromagnetic acoustic transducer (EMAT) module, perform multi-angle ultrasonic signal excitation and reception on the key area, and collect the specific steps of the thickness data of the key area.

[0139] The preliminary thickness distribution data collected from the EMAT module is analyzed using the data analysis unit on the mobile robot platform to identify unclear areas and important areas of the bridge as critical areas. These areas may include areas with abnormal cover thickness, structural stress concentration or historically vulnerable areas.

[0140] According to the identified key areas, the multi-degree-of-freedom robotic arm on the mobile robot platform is operated to accurately adjust the physical position of the EMAT module and move the EMAT module to the predetermined detection point in the key area. This step ensures that the EMAT module can accurately detect the key areas.

[0141] At each predetermined inspection point, the EMAT module’s transmission and reception angles are adjusted based on the specific features of the key area and preliminary scanning data to optimize the excitation and reception effects of the ultrasonic signal. This step improves the accuracy and efficiency of the inspection.

[0142] The ultrasonic signal is excited from the set angle, and the signal reflected from the steel bar cover is received, and the emission and reception time of the ultrasonic wave is recorded. These time data will be used for subsequent thickness calculation.

[0143] The recorded transmission and reception times are used to calculate the propagation time of the ultrasonic wave in the protective layer, and then the thickness of the protective layer is calculated based on the known propagation speed of the ultrasonic wave in the medium. The thickness calculation formula is: d = (v × Δt) / 2, where v is the propagation speed of the ultrasonic wave in concrete, and Δt is the total time for the ultrasonic wave to pass through the protective layer.

[0144] The protective layer thickness measurement results of each critical area, including the measurement value, location and time, are recorded and stored in the database of the mobile robot platform. These data will be used to generate a thickness distribution report for the critical area.

[0145] The stored data is used to generate a thickness distribution report for critical areas. The report details the thickness data for each measurement point, providing a record for complete quality control of the bridge cover thickness.

[0146] Example 6

[0147] In the method for detecting the thickness of the steel bar protective layer during the bridge construction process of the present invention, a preset deep learning algorithm is used to process the collected preliminary thickness distribution data and key area thickness data, identify the thickness of the steel bar protective layer, and record the thickness data of each measuring point to obtain specific steps of the processing results.

[0148] The preliminary thickness distribution data and key area thickness data collected from the electromagnetic acoustic transducer module are first denoised to eliminate data interference caused by environmental noise or equipment errors. The denoised data is normalized to scale the data to a uniform numerical range for easy input into the deep learning model.

[0149] The generative adversarial network (GAN) structure in the deep learning algorithm is used to extract features from the preprocessed data. GAN consists of two parts: the generator (responsible for generating data features) and the discriminator (responsible for distinguishing between real data and generated features). This step can automatically identify features related to the thickness of the steel bar cover.

[0150] The deep learning model was trained using historical thickness data and known bridge structural parameters. During the model training process, the model parameters were continuously adjusted so that the model could accurately identify and predict the thickness of the steel cover.

[0151] The feature-extracted data is analyzed using a trained deep learning model. The model learns the complex patterns in the data, analyzes the thickness of the steel cover at each measurement point, and provides accurate thickness measurements.

[0152] Use anomaly detection mechanisms in deep learning models, such as the Local Outlier Factor (LOF) algorithm, to identify outliers in the data. The LOF algorithm identifies anomalies by comparing the local density difference between a data point and its neighboring points, marking possible areas of steel cover defects as outliers.

[0153] The protective layer thickness measurement results of each key area are recorded, including the measurement value, location and time, and these data are stored in the database of the mobile robot platform. Based on the stored data, a thickness distribution report of the key area is generated, providing a detailed record for the complete quality control of the bridge protective layer thickness.

[0154] Example 7

[0155] In the method for detecting the thickness of the steel bar protective layer during the bridge construction process of the present invention, when an abnormal area is identified in the processing results, the electromagnetic acoustic transducer module is accurately positioned to the abnormal area by the mechanical arm of the mobile robot platform to execute an enhanced detection program for key detection, thereby obtaining specific steps for key detection results.

[0156] After the preliminary thickness distribution data and critical area thickness data are processed using a preset deep learning algorithm, the identified abnormal areas will be marked. These areas may include areas with abnormal protective layer thickness, structural stress concentration, or historically vulnerable areas.

[0157] Operate the multi-degree-of-freedom robotic arm carried by the mobile robot platform, and adjust the position and angle of the electromagnetic acoustic transducer module through a precise positioning system according to the location information of the abnormal area, to ensure that the electromagnetic acoustic transducer module is accurately aligned with the abnormal area.

[0158] Perform an enhanced detection procedure on the located abnormal area, which includes increasing the excitation frequency and receiving sensitivity of the ultrasonic signal to obtain more accurate thickness data. This step may involve adjusting the transmission parameters of the EMAT module, such as frequency, amplitude and pulse width, to adapt to the specific material properties and structural conditions of the abnormal area.

[0159] The enhanced detection procedure collects detailed thickness data of abnormal areas, including but not limited to the propagation time and reflection intensity of ultrasonic waves in the protective layer. The data collected by the enhanced detection is analyzed using a preset deep learning model to identify the actual thickness of the steel bar protective layer and potential structural defects, and obtain key detection results.

[0160] Record the protective layer thickness measurement results of each key area, including the measurement value, location and time, and store these data in the database of the mobile robot platform. Based on the stored data, generate a thickness distribution report for key areas, and update the complete inspection report after each adjustment and re-inspection, recording all inspection data, adjustment measures and verification results to form a complete quality control record of the bridge protective layer thickness.

[0161] Example 8

[0162] In the method for detecting the thickness of the steel bar protective layer during the bridge construction process of the present invention, a detection procedure is enhanced, and the procedure includes the following steps:

[0163] Adjust the emission parameters of the electromagnetic acoustic transducer (EMAT) module, including the frequency f, amplitude A and pulse width τ of the ultrasonic wave, to adapt to the specific material properties and structural conditions of the abnormal area. The adjustment of the emission parameters is based on the following formula: E = Afτ, where E represents the emission energy, ensuring effective excitation and reception of signals under different material properties.

[0164] The ultrasonic signal from the EMAT module and the radar signal from the ground penetrating radar (GPR) are collected synchronously. The two signals are fused at the feature level using multimodal data fusion technology to improve the accuracy and reliability of the detection results. The data fusion algorithm can be based on a deep learning framework as shown below:

[0165] F combined =αF EMAT +(1-α)F GPR where F combined is the fused feature, F EMAT and F GPR are the characteristics of EMAT and GPR respectively, and α is the weight coefficient, which is dynamically adjusted according to the signal-to-noise ratio of the signal.

[0166] Adopting adaptive dynamic focusing algorithm, according to the attenuation and distortion of ultrasonic signal in the process of propagation, the focusing depth D of signal processing is dynamically adjusted. The adjustment of focusing depth is based on the following formula: D = D0·e -β·d Where D0 is the initial focus depth, d is the signal propagation distance, and β is the attenuation coefficient, which is adjusted according to material properties and environmental conditions.

[0167] The pre-trained convolutional neural network (CNN) is used to perform feature-level fusion of ultrasonic and GPR signals to automatically identify and classify the features of abnormal areas, including cracks, cavities, and protective layer peeling. The training of the CNN model is based on the following loss function:

[0168] L=L classification +λL regression

[0169] Where L classification is the classification loss, L regression is the regression loss, and λ is the balance coefficient, which is adjusted according to the actual detection requirements.

[0170] According to the results of AI-assisted analysis, the detection parameters of the EMAT module and GPR are adjusted in real time to optimize the subsequent detection process. The parameter adjustment is based on the following optimization algorithms:

[0171] Where P new is the new parameter, P old are the old parameters, η is the learning rate, and VL is the gradient of the loss function.

[0172] The abnormal area is inspected multiple times at the same angle, and the three-dimensional structure of the abnormal area is reconstructed using stereo imaging technology to comprehensively evaluate the thickness and integrity of the protective layer. Stereo imaging technology is based on the following geometric relationship: Z = f·ab, where Z is the depth information, f is the focal length, and a and b are the coordinate differences of the corresponding points in the left and right images.

[0173] Automatically adjust the detection parameters based on real-time environmental monitoring data, including temperature T, humidity H and wind speed w. Parameter adjustment is based on the following environmental impact model: env =f(T, H, W) where P env is the parameter adjusted according to environmental conditions, and f is the influence function, which is established based on historical data and environmental science principles.

[0174] All data, analysis results and adjustment parameters of enhanced detection are recorded to form an enhanced detection report. Report generation is based on the following data integration process: R = {D, A, P, E}, where R is the enhanced detection report, D is the detection data, A is the analysis result, P is the adjustment parameter, and E is the environmental monitoring data.

[0175] Example 9

[0176] like Figure 2 As shown, the present invention provides a steel bar protective layer thickness detection system during bridge construction, the system specifically includes the following components and steps:

[0177] The mobile robot platform is equipped with an electromagnetic acoustic transducer module and is able to move autonomously along a preset path on the bridge surface. The EMAT module emits ultrasonic pulses of a specific frequency according to a predetermined scanning strategy and records the emission time t emit .

[0178] Multi-DOF Robotic Arm: The multi-DOF robotic arm connected to the mobile robot platform is used to adjust the position and angle of the EMAT module to ensure accurate inspection of the designated area. The robotic arm can move along three main axes (X, Y, Z axis) to cover different parts of the bridge.

[0179] Data acquisition unit: The data acquisition unit is integrated with the EMAT module and the multi-degree-of-freedom robotic arm to collect the excitation and reception data of the ultrasonic signal. The unit records the time difference Δt=t from the emission to the reception of the ultrasonic wave. receive -t emit , as well as the strength and waveform of the signal.

[0180] The deep learning processing unit is configured with a preset deep learning algorithm to process the collected data. The unit includes a trained neural network model that can identify the thickness of the steel bar cover and record the thickness data at each measurement point.

[0181] When the deep learning processing unit identifies an abnormal area, the enhanced detection control unit guides the multi-degree-of-freedom robotic arm to accurately locate the EMAT module to the abnormal area and execute the enhanced detection procedure. The unit adjusts the detection parameters according to the preset strategy and real-time feedback.

[0182] The data recording and report generation unit is used to record all data, analysis results and adjustment parameters of the enhanced inspection, and generate an enhanced inspection report. The report includes the thickness data of each measurement point, the identification results of abnormal areas, and construction adjustment suggestions.

[0183] Specifically, after the system is started, the mobile robot platform moves along the predetermined path on the bridge surface, and the EMAT module starts to transmit and receive ultrasonic signals. The data acquisition unit collects the ultrasonic signal data excited and received by the EMAT module, including the emission time, reception time and signal strength. The deep learning processing unit receives the collected data and processes it using a preset deep learning model to identify the thickness of the steel bar protective layer and record the thickness data of each measuring point. If the deep learning processing unit identifies an abnormal area, the enhanced detection control unit will instruct the multi-degree-of-freedom robotic arm to adjust the position and angle of the EMAT module to accurately locate the abnormal area. The enhanced detection control unit executes the enhanced detection procedure, which may include increasing the excitation frequency and receiving sensitivity of the ultrasonic signal to obtain more accurate thickness data. The data recording and report generation unit records all data, analysis results and adjustment parameters of the enhanced detection, and generates an enhanced detection report to provide a basis for subsequent construction adjustments and quality control.

[0184] Example 10

[0185] The reinforcement protection layer thickness detection system in the bridge construction process of the present invention includes the following modules and implementation steps:

[0186] Signal optimization module: This module is responsible for adjusting the emission parameters of the EMAT module, including the frequency f, amplitude A and pulse width τ of the ultrasonic wave, to adapt to the specific material properties and structural conditions of the abnormal area. The adjustment of the emission parameters is based on the following formula: E = A·f·τ, where E represents the emission energy, ensuring the effective excitation and reception of the signal under different material properties.

[0187] Multimodal data fusion module: This module synchronously collects ultrasonic signals from the EMAT module and radar signals from the GPR, and performs data fusion processing. The data fusion algorithm can be based on a deep learning framework as shown below: combined =αF EMAT +(1-α)F GPR Among them, F combined is the fused feature, F EMAT and F GPR are the characteristics of EMAT and GPR respectively, and α is the weight coefficient, which is dynamically adjusted according to the signal-to-noise ratio of the signal.

[0188] Dynamic focus control module: This module uses an adaptive dynamic focus algorithm to dynamically adjust the focus depth D of the signal processing according to the attenuation and distortion of the ultrasonic signal during propagation. The adjustment of the focus depth is based on the following formula: D = D0·e -β·d Where D0 is the initial focus depth, d is the signal propagation distance, and β is the attenuation coefficient, which is adjusted according to material properties and environmental conditions.

[0189] Feature recognition module: This module uses a pre-trained convolutional neural network (CNN) to perform feature-level fusion of ultrasonic and GPR signals to automatically identify and classify the features of abnormal areas, including cracks, cavities, and protective layer peeling. The training of the CNN model is based on the following loss function:

[0190] L=L classiiication +λLL regression

[0191] Where L classification is the classification loss, L regression is the regression loss, and λ is the balance coefficient, which is adjusted according to the actual detection requirements.

[0192] Real-time parameter adjustment module This module adjusts the detection parameters of the EMAT module and GPR in real time according to the analysis results of the feature recognition module to optimize the subsequent detection process. The parameter adjustment is based on the following optimization algorithms: Where P new is the new parameter, P old is the old parameter, η is the learning rate, is the gradient of the loss function.

[0193] Stereoscopic imaging module: This module performs multiple inspections of abnormal areas from different angles, reconstructs the three-dimensional structure of the abnormal area using stereoscopic imaging technology, and comprehensively evaluates the thickness and integrity of the protective layer. Stereoscopic imaging technology is based on the following geometric relationship: Z = f·ab, where Z is the depth information, f is the focal length, and a and b are the coordinate differences of the corresponding points in the left and right images.

[0194] Environmental adaptability module: This module automatically adjusts the detection parameters based on real-time environmental monitoring data, including temperature T, humidity H and wind speed w. Parameter adjustment is based on the following environmental impact model: P env =f(T, H, W) where P env is the parameter adjusted according to environmental conditions, and f is the influence function, which is established based on historical data and environmental science principles.

[0195] 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 method for detecting the thickness of a steel bar protective layer during bridge construction, characterized in that: include, The mobile robot platform is equipped with an electromagnetic acoustic transducer module and moves along the bridge surface to perform a preliminary range scan covering the entire bridge platform, determine the preliminary thickness distribution of the steel bar protection layer, and obtain preliminary thickness distribution data; According to the preliminary thickness distribution results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is used to adjust the position and angle of the electromagnetic acoustic transducer module, perform multi-angle ultrasonic signal excitation and reception on the key area, collect the thickness data of the key area, use the preset deep learning algorithm to process the collected preliminary thickness distribution data and key area thickness data, identify the thickness of the steel bar protective layer, and record the thickness data of each measuring point to obtain the processing results; When the abnormal area is identified by the processing result, the electromagnetic acoustic transducer module is accurately positioned to the abnormal area through the mechanical arm of the mobile robot platform to perform enhanced detection procedures for key detection and obtain key detection results; According to the key test results, the thickness of the protective layer is compared with the preset standard to determine whether the actual thickness of the protective layer meets the design requirements and obtain the comparison results; Integrate the processing results, key test results and comparison results to obtain a complete test report.

2. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: If the comparison results show that the actual cover thickness does not meet the design requirements, perform the following steps: The data analysis unit on the mobile robot platform is used to conduct in-depth analysis of the comparison results to identify specific areas that do not meet the design requirements and the degree of deviation; According to the analysis results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is automatically or manually adjusted to reposition the electromagnetic acoustic transducer module to the non-compliant area; Dynamically adjust the construction process based on the real-time detection data of the electromagnetic acoustic transducer module and the results of the enhanced detection program, including but not limited to adjusting the steel bar layout, concrete mix or pouring technology, and correcting thickness deviations; After adjusting the construction process, the electromagnetic acoustic transducer module carried by the mobile robot platform is used to detect the adjusted area again to verify whether the thickness of the protective layer meets the predetermined standard; If the adjusted key test results still do not meet the design requirements, continue to iteratively adjust the construction process based on the real-time detection data of the electromagnetic acoustic transducer module until the actual protective layer thickness meets the design requirements; After each adjustment and re-test, the complete test report is updated to record all test data, adjustment measures and verification results to form a complete quality control record of the bridge cover thickness.

3. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: During the initial range scan and key inspection covering the entire bridge platform, ground penetrating radar technology was used to perform auxiliary scans of the bridge structure to verify the detection results of the electromagnetic acoustic transducer module and provide more information about the interior of the concrete structure; The detection data obtained by combining the electromagnetic acoustic transducer module and the ground penetrating radar technology are fused and processed through a fusion algorithm to obtain fused data.

4. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: The steps of using a mobile robot platform equipped with an electromagnetic acoustic transducer module to move along the bridge surface to perform a preliminary range scan covering the entire bridge platform, determine the preliminary thickness distribution of the steel bar protective layer, and obtain preliminary thickness distribution data include: The electromagnetic acoustic transducer module of the mobile robot platform moves along a predetermined path on the bridge surface, emitting ultrasonic pulses of a specific frequency and recording the emission time; The electromagnetic acoustic transducer module receives the ultrasonic signal reflected from the steel bar protective layer and records the receiving time; According to the time difference between transmission and reception and the propagation speed of ultrasonic waves in the medium, the total distance that the ultrasonic waves pass through the protective layer is calculated to obtain a preliminary measurement value of the thickness of the protective layer; Comparing the obtained preliminary measurement value with the preset calibration parameters of the electromagnetic acoustic transducer module, calibrating the measurement result, and obtaining calibrated protective layer thickness data; The obtained calibrated protection layer thickness data is stored in a database of the mobile robot platform, and a preliminary thickness distribution map is generated as preliminary thickness distribution data.

5. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: According to the preliminary thickness distribution results, the multi-degree-of-freedom robotic arm on the mobile robot platform is used to adjust the position and angle of the electromagnetic acoustic transducer module, to excite and receive multi-angle ultrasonic signals in the key area, and to collect thickness data of the key area. Specifically, Analyze preliminary thickness distribution data to identify unclear areas and important areas of the bridge as critical areas, including areas with abnormal cover thickness, structural stress concentration or historically vulnerable areas; According to the identified key areas, the physical position of the electromagnetic acoustic transducer module is adjusted using the multi-degree-of-freedom robotic arm carried by the mobile robot platform, and the electromagnetic acoustic transducer module is moved to the predetermined detection point of the key area; At each predetermined inspection point, according to the specific characteristics of the key area and the preliminary scanning data, the emission angle and receiving angle of the electromagnetic acoustic transducer module are adjusted to optimize the excitation and reception effects of the ultrasonic signal; Excite ultrasonic signals from a set angle, receive signals reflected from the steel bar protective layer, and record the emission and reception time of the ultrasonic waves; The recorded emission and reception times are used to calculate the propagation time of the ultrasonic wave in the protective layer, and then the thickness of the protective layer is calculated based on the known propagation speed of the ultrasonic wave in the medium; The protective layer thickness measurement results of each critical area are recorded, including the measurement value, location and time, and these data are stored in the database of the mobile robot platform to generate a thickness distribution report for the critical area.

6. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: The steps of collecting preliminary thickness distribution data and key area thickness data using a preset deep learning algorithm, identifying the thickness of the steel bar cover, and recording the thickness data of each measuring point to obtain the processing results include: De-noising and normalizing the preliminary thickness distribution data and key area thickness data collected from the electromagnetic acoustic transducer module; The generative adversarial network structure in the deep learning algorithm is used to extract features from the preprocessed data and automatically identify features related to the thickness of the steel bar cover; The deep learning model is trained using historical thickness data and known bridge structure parameters to obtain a trained deep learning model; The trained deep learning model is used to analyze the data after feature extraction and analyze the thickness of the steel bar cover at each measurement point; Using the anomaly detection mechanism in the deep learning model, outliers in the data are identified and possible areas of steel bar cover defects are marked as abnormal areas.

7. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: When the abnormal area is identified by the processing result, the electromagnetic acoustic transducer module is accurately positioned to the abnormal area by the mechanical arm of the mobile robot platform to perform the enhanced detection program for key detection, and the steps of obtaining the key detection results specifically include: When the abnormal area is identified by the processing results, the multi-degree-of-freedom robotic arm carried by the mobile robot platform is operated to adjust the position and angle of the electromagnetic acoustic transducer module according to the position information of the abnormal area to ensure that the electromagnetic acoustic transducer module is accurately aligned with the abnormal area; Perform enhanced detection procedures on the located abnormal areas, including increasing the excitation frequency and receiving sensitivity of the ultrasonic signal to obtain more accurate thickness data; Collect detailed thickness data of abnormal areas through enhanced detection procedures, including but not limited to the propagation time and reflection intensity of ultrasonic waves in the protective layer; The preset deep learning model is used to analyze the data collected by the enhanced detection, identify the actual thickness of the steel bar protection layer and potential structural defects, and obtain key detection results.

8. The method for detecting the thickness of the steel bar protective layer during bridge construction according to claim 1, characterized in that: The enhanced testing procedure also includes the following steps: Ultrasonic signal optimization, adjusting the emission parameters of the electromagnetic acoustic transducer module, including the frequency, amplitude and pulse width of the ultrasonic wave, to adapt to the specific material characteristics and structural conditions of the abnormal area; Synchronously collect ultrasonic signals from the electromagnetic acoustic transducer module and radar signals from the ground penetrating radar to perform multi-modal data fusion; Adopting adaptive dynamic focusing algorithm, the focusing depth of signal processing is dynamically adjusted according to the attenuation and distortion of ultrasonic signals during propagation; Use pre-trained convolutional neural networks to perform feature-level fusion of ultrasonic and ground-penetrating radar radar signals to automatically identify and classify features of abnormal areas, including cracks, voids, and protective layer peeling; According to the results of AI-assisted analysis, the detection parameters of the electromagnetic acoustic transducer module and ground-penetrating radar are adjusted in real time to optimize the subsequent detection process; Conduct multiple inspections of the abnormal area from different angles, use stereo imaging technology to reconstruct the three-dimensional structure of the abnormal area, and comprehensively evaluate the thickness and integrity of the protective layer; Automatically adjust detection parameters based on real-time environmental monitoring data, including temperature, humidity and wind speed; All data, analysis results and adjustment parameters of enhanced testing shall be recorded to form an enhanced testing report.

9. A steel bar protective layer thickness detection system during bridge construction, based on a steel bar protective layer thickness detection method during bridge construction according to any one of claims 1 to 8, characterized in that: include, A mobile robot platform, equipped with an electromagnetic acoustic transducer module, is used to move along the bridge surface and to excite and receive ultrasonic signals; A multi-degree-of-freedom robotic arm connected to the mobile robot platform, used to adjust the position and angle of the electromagnetic acoustic transducer module to accurately detect the designated area; A data acquisition unit, integrated with the electromagnetic acoustic transducer module and the multi-degree-of-freedom robotic arm, for collecting excitation and reception data of ultrasonic signals; A deep learning processing unit, configured with a preset deep learning algorithm, for processing the collected data, identifying the thickness of the steel bar cover, and recording the thickness data of each measuring point; An enhanced detection control unit, for guiding the multi-degree-of-freedom robotic arm to accurately position the electromagnetic acoustic transducer module to the abnormal area and execute an enhanced detection procedure when an abnormal area is identified; The data recording and report generation unit is used to record all data, analysis results and adjustment parameters of the enhanced detection, and generate an enhanced detection report.

10. A steel bar protective layer thickness detection system in a bridge construction process according to claim 9, characterized in that: The enhanced detection control unit comprises: A signal optimization module, which is used to adjust the emission parameters of the electromagnetic acoustic transducer module, including the frequency, amplitude and pulse width of the ultrasonic wave, to adapt to the specific material characteristics and structural conditions of the abnormal area; A multimodal data fusion module, used for synchronously collecting ultrasonic signals from the electromagnetic acoustic transducer module and radar signals from the ground penetrating radar, and performing data fusion processing; A dynamic focusing control module is used to dynamically adjust the focusing depth of signal processing according to the attenuation and distortion of the ultrasonic signal during propagation by using an adaptive dynamic focusing algorithm; A feature recognition module that uses a pre-trained convolutional neural network to perform feature-level fusion of ultrasonic and ground-penetrating radar signals to automatically identify and classify features of abnormal areas, including cracks, voids, and protective layer peeling; A real-time parameter adjustment module, used to adjust the detection parameters of the electromagnetic acoustic transducer module and the ground penetrating radar in real time according to the analysis results of the feature recognition module, so as to optimize the subsequent detection process; Stereoscopic imaging module, used to conduct multiple inspections of abnormal areas from different angles, reconstruct the three-dimensional structure of the abnormal area using stereoscopic imaging technology, and comprehensively evaluate the thickness and integrity of the protective layer; Environmental adaptability module, used to automatically adjust detection parameters based on real-time environmental monitoring data, including temperature, humidity and wind speed.

Images