Bridge underwater pile foundation stress wave detection system

Abstract

The invention discloses a bridge underwater pile foundation stress wave detection system, and relates to the technical field of stress wave detection, the system comprises a transmitting module, a receiving module, a calculation module and an analysis module, the transmitting module uses sonar to transmit underwater sound waves, the receiving module is used for receiving reflected sound waves, the calculation module is used for calculating characteristics of different underwater stress waves, and the analysis module is used for analyzing the characteristics of the different underwater stress waves. Stress wave images of all sound waves in the underwater pile foundation of the bridge are drawn, and the analysis module performs hierarchical analysis on the underwater pile foundation of the bridge according to the stress wave images and performs historical comparison detection on stress wave characteristics to obtain abnormal conditions of the bridge pile foundation corresponding to different stress waves; personalized detection of the bridge pile foundation with liquid obstruction is specially realized, and detailed position detection, complete anomaly classification and anomaly effect detection are carried out on the abnormal condition of the underwater bridge pile foundation, so that the cost can be saved, and symptomatic treatment can be rapidly carried out on the abnormal condition.

Description

Technical Field

[0001] The invention relates to the technical field of stress wave detection, and in particular to a stress wave detection system for underwater pile foundations of bridges. Background Art

[0002] The stability of underwater pile foundations is crucial to the safety of bridges. Traditional inspection methods are difficult to carry out underwater, so efficient non-destructive testing technology is needed. Stress wave detection uses the propagation of sound waves or ultrasonic waves in pile foundations to evaluate their structural integrity and detect cracks, defects or material degradation. The underwater environment is complex and fluctuations have an impact on the detection results. Specific equipment and technology are required to adapt to the conditions of underwater detection.

[0003] At present, in a Chinese invention patent with publication number CN110487917A, a stress wave detection and analysis system based on neural network is proposed, which supports active maintenance, can discover and handle equipment failures more timely and effectively, and reduce the cost of repair and replacement of equipment. The neural network can help make more automated and intelligent judgments on the health status and fault type of the equipment reflected by the stress wave signal, and increase the applicability of the device. However, the existing technology cannot realize personalized detection of bridge pile foundations with liquid obstructions, and cannot perform detailed position detection, complete abnormal classification and abnormal effect severity detection of abnormal conditions of underwater bridge pile foundations, thereby failing to save costs and quickly carry out symptomatic treatment of abnormal conditions. Summary of the invention

[0004] The technical problem solved by the present invention is that the existing technology cannot realize personalized detection of bridge pile foundations with liquid obstruction, and cannot perform detailed position detection, complete abnormality classification and abnormality severity detection of abnormal conditions of underwater bridge pile foundations, thereby failing to save costs and quickly carry out symptomatic treatment of abnormal conditions.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: A bridge underwater pile foundation stress wave detection system comprises a transmitting module, a receiving module, a calculating module and an analyzing module;

[0006] The transmitting module uses sonar to transmit underwater sound waves and collects the water-pushing shape of the underwater sound waves and the diffraction image corresponding to the sound source depth level;

[0007] The receiving module is used to receive the reflected sound waves and pre-process the reflected sound waves;

[0008] The calculation module is used to calculate the characteristics of different underwater stress waves and draw stress wave images of all sound waves inside the underwater pile foundation of the bridge;

[0009] The analysis module performs hierarchical analysis on the underwater pile foundation of the bridge according to the stress wave image, designs confidence simulation calculations, performs historical comparative detection on the stress wave characteristics obtained by the calculation module, and obtains abnormal conditions of the bridge pile foundation corresponding to different stress waves.

[0010] Preferably, the transmitting module comprises:

[0011] Adjust the transmitting frequency and beam width of the sonar, and transmit sound waves at a preset frequency. The preset frequency is set to the first transmitting frequency to the nth transmitting frequency according to the frequency size, where n>1, and collect diffraction wave information generated by sound waves penetrating different water layers when propagating underwater. The diffraction wave information includes the water depth position, diffraction wavelength and diffraction wave signal generated by the diffraction wave.

[0012] Preferably, the receiving module comprises a sonar receiver for receiving the reflected stress wave signal and recording the intensity and arrival time of the reflected sound wave.

[0013] Preferably, the computing module comprises a feature extraction unit and an image rendering unit:

[0014] The feature extraction unit comprises:

[0015] Detailed information is extracted from the collected stress wave signal, the detailed information including the propagation path, reflection angle and diffraction pattern, the detailed information is standardized and then input into the convolution layer, and detailed information feature data is extracted from the detailed information.

[0016] Preferably, the image drawing unit comprises:

[0017] The method of converting the reflected stress wave signal, the diffracted wave signal and the transmitted stress wave signal into image data includes:

[0018] The collected reflected stress wave signal, diffraction wave signal and transmitted stress wave signal are converted into discrete signals, and noise and unnecessary frequency components are removed by using filters, and key features of reflected sound waves and diffraction waves are extracted, wherein the key features include signal amplitude, time delay and frequency components. According to the speed of sound and signal propagation time, the signal is converted into distance to form a time-distance graph, and the time domain signal is converted into a frequency domain signal according to Fourier transform, and a spectrum graph is generated by combining the key features. According to the time-distance graph and the spectrum graph, an A-scan graph of the reflected stress wave signal, diffraction wave signal and transmitted stress wave signal is generated by using image generation software, and a B-scan image and a C-scan image are generated by scanning multiple A-scan images of the same stress wave signal, and a three-dimensional image is generated by combining multiple B-scan images.

[0019] Preferably, the analysis module includes an image processing unit, a simulation calculation unit, a comparison detection unit and a pile foundation influence unit:

[0020] The image processing unit comprises:

[0021] According to the three-dimensional image obtained by the calculation module, a three-dimensional stereoscopic structural model of the underwater pile foundation of the bridge is established by using three-dimensional drawing software, and according to the B-scan image and C-scan image of the reflected stress wave signal and the transmitted stress wave signal, the cross section of the bridge pile foundation is divided into grid graphics by using a grid method, and represented by a three-dimensional rectangular coordinate system, and abnormal marking is performed at the corresponding position of the abnormal data of the bridge pile foundation, and the abnormal data includes abnormal values ​​in all images of the stress wave signal;

[0022] The abnormal value corresponding to the abnormal mark position is clustered according to the detailed information feature data by k-means to obtain the first abnormal feature type to the mth abnormal feature type, where m>1. The size of the three-dimensional stereoscopic structure model of the bridge pile foundation, the grid size, the sound wave intensity of the abnormal mark position and the coordinates of the abnormal mark position are normalized and imported into the LTI path forward algorithm model to obtain the path forward calculation result, which includes:

[0023] The path aggregation and path blank phenomena of the stress wave signal, the path aggregation phenomenon is the first abnormal area phenomenon, the path blank phenomenon is the second abnormal area phenomenon, and the path forward calculation result is used as the first image detection effect.

[0024] Preferably, the size, grid size, sound wave intensity of the abnormal mark position and coordinates of the abnormal mark position of the three-dimensional stereoscopic structural model of the bridge pile foundation are normalized and then imported into the IRT velocity field inversion algorithm model to calculate the sectional surface tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The sectional surface tomography results are spliced ​​and combined into a three-dimensional histogram form to obtain the overall tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The detailed abnormal severity effect of the bridge pile foundation is obtained according to the color distribution in the overall tomography result, and the detailed abnormal severity effect is used as the second image detection effect.

[0025] Preferably, the simulation calculation unit comprises:

[0026] The first image detection effect and the second image detection effect are calculated by confidence simulation to obtain the abnormal sound wave intensity confidence interval of the bridge pile foundation, and the abnormal sound wave intensity confidence interval represents the internal quality detection condition of the bridge pile foundation. The confidence simulation model calculation process includes:

[0027] The abnormal feature types corresponding to the first image detection effect and the second image detection effect are used as statistical samples for statistical analysis. The statistical analysis includes mean calculation and standard deviation calculation. The mean and standard deviation of the reflected stress wave signal intensity are obtained respectively. The normal stress wave signal threshold is set, and the confidence prediction interval of the reflected stress wave signal is calculated. The mathematical expression for calculating the confidence prediction interval is:

[0028] ;

[0029] in, is the confidence interval, A is the statistical parameter corresponding to the probability distribution type, is the mean value of the reflected stress wave signal intensity, σ is the standard deviation of the reflected stress wave signal intensity, a is the normal reflected stress wave signal threshold, and the probability distribution types include normal distribution, binomial distribution, Poisson distribution and uniform distribution;

[0030] The reflected stress wave signal strength that is higher or lower than the confidence prediction interval is an abnormal sound wave, and the corresponding abnormal sound wave strength is recorded at the corresponding abnormal mark position;

[0031] The abnormal characteristic type of the abnormal sound wave is also recorded at the corresponding abnormal mark position;

[0032] When the number of grids divided in the overall tomographic imaging result of the three-dimensional structure model of the bridge pile foundation is greater than a preset division threshold, a statistical parameter is selected according to the probability distribution type. ;

[0033] If a is smaller, the statistical parameter A is larger, the confidence prediction interval is higher, and the second abnormal area has fewer abnormal phenomena;

[0034] If a is larger, the statistical parameter A is smaller, the confidence prediction interval is lower, and the second abnormal area has more abnormal phenomena.

[0035] Preferably, the comparison detection unit comprises:

[0036] Collect a three-dimensional structural model of the underwater part of the historical bridge pile foundation and the received historical stress wave signal data, the historical stress wave signal data including the frequency spectrum and sound wave intensity of the reflected stress wave signal, the diffraction wave signal and the transmitted stress wave signal, compare the first image detection effect and the second image detection effect currently measured with the historical stress wave signal data through the calculation module, the image processing unit and the simulation calculation unit, and obtain the distinguishing features of the current bridge pile foundation and the underwater part of the historical bridge pile foundation, the distinguishing features including undetected abnormal conditions of the bridge pile foundation, and obtain the third image detection effect by passing the undetected abnormal conditions of the bridge pile foundation through the image processing unit and the simulation calculation unit.

[0037] Preferably, the pile foundation influencing unit comprises:

[0038] Perform k-means clustering on the abnormal feature types corresponding to the detection effects included in the third image detection effect except the first image detection effect and the second image detection effect, to obtain the m+1th abnormal feature type ~ and the kth abnormal feature type, where k>m+1, m>1;

[0039] The first image detection effect, the second image detection effect, the third image detection effect and the corresponding abnormal feature type recorded at the abnormal mark position are recorded in a dictionary form, and the pointer of the address in the dictionary form is pointed to the memory address position where the three-dimensional coordinates of the abnormal mark position are stored, and the head nodes of the chain queue are set to be the first abnormal feature type~the mth abnormal feature type~the kth abnormal feature type respectively, and the memory address positions where the three-dimensional coordinates of all abnormal mark positions are stored are respectively connected to the corresponding chain queue head nodes to form a chain queue form;

[0040] All abnormal feature types represent a bridge pile foundation impact effect, which includes material defects, pile foundation structure defects and underwater biological attachment corrosion.

[0041] The beneficial effects of the present invention are as follows: it realizes personalized detection of bridge pile foundations with liquid obstruction, and performs detailed position detection, complete abnormality classification and abnormality severity detection on abnormal conditions of underwater bridge pile foundations, thereby saving costs and quickly carrying out symptomatic treatment of abnormal conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic diagram of the basic flow of a bridge underwater pile foundation stress wave detection system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0044] Reference Figure 1 , which is an embodiment of the present invention, provides a bridge underwater pile foundation stress wave detection system, characterized in that it includes a transmitting module, a receiving module, a calculating module and an analyzing module;

[0045] The transmitting module uses sonar to transmit underwater sound waves and collects the water-pushing shape of the underwater sound waves and the diffraction image corresponding to the sound source depth level;

[0046] The receiving module is used to receive the reflected sound waves and pre-process the reflected sound waves;

[0047] The preprocessing includes measuring the signal strength, distance, time and angle of the reflected sound wave;

[0048] The calculation module is used to calculate the characteristics of different underwater stress waves and draw stress wave images of all sound waves inside the underwater pile foundation of the bridge;

[0049] The analysis module performs hierarchical analysis on the underwater pile foundation of the bridge according to the stress wave image, designs confidence simulation calculations, performs historical comparative detection on the stress wave characteristics obtained by the calculation module, and obtains abnormal conditions of the bridge pile foundation corresponding to different stress waves.

[0050] It can realize personalized detection of bridge pile foundations with liquid obstruction, and carry out detailed location detection, complete abnormality classification and abnormal effect severity detection of abnormal conditions of underwater bridge pile foundations.

[0051] The transmitting module comprises:

[0052] Adjust the sonar transmission frequency and beam width, low frequency for long-distance detection, high frequency for high-resolution images. Transmit sound waves at a preset frequency, the preset frequency is set to the first transmission frequency ~ nth transmission frequency according to the frequency size, where n>1, collect the diffraction wave information generated by the sound wave penetrating different water layers when propagating underwater, the diffraction wave information includes the water depth position, diffraction wavelength and diffraction wave signal generated by the diffraction wave.

[0053] The receiving module comprises a sonar receiver for receiving the reflected stress wave signal and recording the intensity and arrival time of the reflected sound wave.

[0054] The computing module includes a feature extraction unit and an image rendering unit:

[0055] The feature extraction unit comprises:

[0056] Detailed information is extracted from all collected stress wave signals, including propagation path, reflection angle and diffraction pattern, and the detailed information is input into the convolution layer after being standardized, and detailed information feature data is extracted from the detailed information.

[0057] The image drawing unit comprises:

[0058] Converting reflected stress wave signals, diffracted wave signals and transmitted stress wave signals into image data includes:

[0059] The collected reflected stress wave signal, diffraction wave signal and transmitted stress wave signal are converted into discrete signals, and Nyquist theorem is used to avoid aliasing. Filters are used to remove noise and unnecessary frequency components to improve signal quality, and key features of reflected sound waves and diffraction waves are extracted. The key features include signal amplitude, time delay and frequency components. According to the speed of sound and signal propagation time, the signal is converted into distance to form a time-distance graph, which is helpful to locate the position of reflection and diffraction. The time domain signal is converted into a frequency domain signal according to Fourier transform, and a spectrum graph is generated in combination with the key features. The spectrum graph can show the intensity of different frequency components and is used to further analyze signal characteristics. According to the time-distance The A-scan image of the reflected stress wave signal, the diffraction wave signal and the transmitted stress wave signal is generated by using image generation software, and the A-scan image shows the change of the signal intensity of a single channel over time. By scanning multiple A-scan images of the same stress wave signal, a B-scan image is generated. The B-scan image is a two-dimensional image, which shows the propagation path and reflection position of the sound wave in the medium, forms a cross-sectional view, and generates a C-scan image. The C-scan image is an image based on a surface perspective, which shows the intensity of the reflected signal at a specific depth or position. A three-dimensional image is generated by combining multiple B-scan images to show the three-dimensional distribution of the reflected signal.

[0060] Combining multiple B-scan images to generate a 3D image includes:

[0061] An ultrasonic detector is used to collect several B-scan images at different angles. The B-scan images represent the depth reflection data on the corresponding scanning path. Noise removal and image enhancement are performed on each B-scan image in turn to improve the accuracy of subsequent processing and ensure that the spatial positions of all B-scan images are correctly aligned. An image registration algorithm is used to adjust the spatial position of the image. Depth information and intensity values ​​are extracted from each B-scan image. The depth information is calculated by signal propagation time. The depth data of each B-scan image is converted into three-dimensional coordinates (X, Y, Z) to form point cloud data. The coordinates of each point correspond to the position and depth in the image. The point cloud data of all B-scan images are merged to form a complete three-dimensional model. Point cloud processing tools are used for splicing to ensure that the point cloud data in different images are correctly aligned. A three-dimensional mesh model is generated based on the point cloud data, and a triangulation algorithm is used to convert the point cloud into a continuous three-dimensional surface.

[0062] The analysis module includes an image processing unit, a simulation calculation unit, a comparison detection unit and a pile foundation influence unit:

[0063] The image processing unit comprises:

[0064] According to the three-dimensional image obtained by the calculation module, a three-dimensional stereoscopic structural model of the underwater pile foundation of the bridge is established by using three-dimensional drawing software, and according to the B-scan image and C-scan image of the reflected stress wave signal and the transmitted stress wave signal, the cross section of the bridge pile foundation is divided into grid graphics by using a grid method, and represented by a three-dimensional rectangular coordinate system, and abnormal marking is performed at the corresponding position of the abnormal data of the bridge pile foundation, and the abnormal data includes abnormal values ​​in all images of the stress wave signal;

[0065] The abnormal value corresponding to the abnormal mark position is subjected to k-means clustering according to the detailed information feature data to obtain the first abnormal feature type to the mth abnormal feature type, wherein m>1, and m is the number of abnormal feature types obtained by k-means clustering. The size of the three-dimensional stereoscopic structure model of the bridge pile foundation, the grid size, the acoustic wave intensity of the abnormal mark position, and the coordinates of the abnormal mark position are normalized and imported into the LTI path forward algorithm model to obtain the path forward calculation result, which includes:

[0066] The path aggregation and path blank phenomena of the stress wave signal, the path aggregation phenomenon is the first abnormal area phenomenon, and the path blank phenomenon is the second abnormal area phenomenon. The path aggregation and path blank phenomena show the trend of the stress wave signal propagating along the direction of the abnormal area, which further explains the cause of the inversion image artifacts. The path forward calculation result is used as the first image detection effect.

[0067] The size, grid size, sound wave intensity of the abnormal mark position and coordinates of the abnormal mark position of the three-dimensional stereoscopic structural model of the bridge pile foundation are normalized and then imported into the IRT velocity field inversion algorithm model to calculate the sectional surface tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The sectional surface tomography results are spliced ​​and combined into a three-dimensional histogram form to obtain the overall tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The detailed abnormal severity effect of the bridge pile foundation is obtained according to the color distribution in the overall tomography result, and the detailed abnormal severity effect is used as the second image detection effect.

[0068] The simulation calculation unit comprises:

[0069] The first image detection effect and the second image detection effect are calculated by confidence simulation to obtain the abnormal sound wave intensity confidence interval of the bridge pile foundation, and the abnormal sound wave intensity confidence interval represents the internal quality detection condition of the bridge pile foundation. The confidence simulation model calculation process includes:

[0070] The abnormal feature types corresponding to the first image detection effect and the second image detection effect are used as statistical samples for statistical analysis. The statistical analysis includes mean calculation and standard deviation calculation. The mean and standard deviation of the reflected stress wave signal intensity are obtained respectively. The normal stress wave signal threshold is set, and the confidence prediction interval of the reflected stress wave signal is calculated. The mathematical expression for calculating the confidence prediction interval is:

[0071] ;

[0072] in, is the confidence prediction interval, A is the statistical parameter corresponding to the probability distribution type, is the mean value of the reflected stress wave signal intensity, σ is the standard deviation of the reflected stress wave signal intensity, a is the normal reflected stress wave signal threshold, and the probability distribution types include normal distribution, binomial distribution, Poisson distribution and uniform distribution;

[0073] The reflected stress wave signal strength that is higher or lower than the confidence prediction interval is an abnormal sound wave, and the corresponding abnormal sound wave strength is recorded at the corresponding abnormal mark position;

[0074] The abnormal characteristic type of the abnormal sound wave is also recorded at the corresponding abnormal mark position;

[0075] When the number of grids divided in the overall tomographic imaging result of the three-dimensional structure model of the bridge pile foundation is greater than a preset division threshold, a statistical parameter is selected according to the probability distribution type. ;

[0076] If a is smaller, the statistical parameter A is larger, the confidence prediction interval is higher, and the second abnormal area has fewer abnormal phenomena;

[0077] If a is larger, the statistical parameter A is smaller, the confidence prediction interval is lower, and the second abnormal area has more abnormal phenomena.

[0078] The value of a is adjusted to obtain the required confidence prediction interval, so that the detection effects of the first image and the second image are clearer and the values ​​are more specific.

[0079] The comparison detection unit comprises:

[0080] Collect a three-dimensional structural model of the underwater part of the historical bridge pile foundation and the received historical stress wave signal data, the historical stress wave signal data including the frequency spectrum and sound wave intensity of the reflected stress wave signal, the diffraction wave signal and the transmitted stress wave signal, compare the first image detection effect and the second image detection effect currently measured with the historical stress wave signal data through the calculation module, the image processing unit and the simulation calculation unit, and obtain the distinguishing features of the current bridge pile foundation and the underwater part of the historical bridge pile foundation, the distinguishing features including undetected abnormal conditions of the bridge pile foundation, and obtain the third image detection effect by passing the undetected abnormal conditions of the bridge pile foundation through the image processing unit and the simulation calculation unit.

[0081] The pile foundation impact unit comprises:

[0082] Perform k-means clustering on the abnormal feature types corresponding to the detection effects included in the third image detection effect except the first image detection effect and the second image detection effect, to obtain the m+1th abnormal feature type to the kth abnormal feature type, where k>m+1, m>1, and k is the total value of all abnormal feature types;

[0083] The first image detection effect, the second image detection effect, the third image detection effect and the corresponding abnormal feature type recorded at the abnormal mark position are recorded in a dictionary form, the pointer of the address in the dictionary form is pointed to the memory address position where the three-dimensional coordinates of the abnormal mark position are stored, the head nodes of the chain queue are set to be the first abnormal feature type, the second abnormal feature type, ..., the mth abnormal feature type, ... and the kth abnormal feature type respectively, and the memory address positions where the three-dimensional coordinates of all abnormal mark positions are stored are respectively combined with the corresponding chain queue head nodes to form a chain queue form;

[0084] All abnormal feature types represent a bridge pile foundation impact effect, which includes material defects, pile foundation structure defects and underwater biological attachment corrosion.

[0085] It can realize personalized detection of bridge pile foundations with liquid obstruction, conduct detailed location detection, complete abnormality classification and abnormality severity detection of abnormal conditions of underwater bridge pile foundations, so as to save costs and quickly deal with abnormal conditions symptomatically.

[0086] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. Bridge underwater pile foundation stress wave detection system, characterized by: It includes a transmitting module, a receiving module, a calculating module and an analyzing module; The transmitting module uses sonar to transmit underwater sound waves and collects the water-pushing shape of the underwater sound waves and the diffraction image corresponding to the sound source depth level; The receiving module is used to receive the reflected sound waves and pre-process the reflected sound waves; The calculation module is used to calculate the characteristics of different underwater stress waves and draw stress wave images of all sound waves inside the underwater pile foundation of the bridge; The analysis module performs hierarchical analysis on the underwater pile foundation of the bridge according to the stress wave image, designs confidence simulation calculations, performs historical comparative detection on the stress wave characteristics obtained by the calculation module, and obtains abnormal conditions of the bridge pile foundation corresponding to different stress waves.

2. The bridge underwater pile foundation stress wave detection system according to claim 1, characterized in that: The transmitting module comprises: Adjust the transmitting frequency and beam width of the sonar, and transmit sound waves at a preset frequency. The preset frequency is set to the first transmitting frequency to the nth transmitting frequency according to the frequency size, where n>1, and collect diffraction wave information generated by sound waves penetrating different water layers when propagating underwater. The diffraction wave information includes the water depth position, diffraction wavelength and diffraction wave signal generated by the diffraction wave.

3. The bridge underwater pile foundation stress wave detection system according to claim 2, characterized in that: The receiving module comprises a sonar receiver for receiving the reflected stress wave signal and recording the intensity and arrival time of the reflected sound wave.

4. The bridge underwater pile foundation stress wave detection system according to claim 3, characterized in that: The computing module includes a feature extraction unit and an image rendering unit: The feature extraction unit comprises: Detailed information is extracted from the collected stress wave signal, the detailed information including the propagation path, reflection angle and diffraction pattern, the detailed information is standardized and then input into the convolution layer, and detailed information feature data is extracted from the detailed information.

5. The bridge underwater pile foundation stress wave detection system according to claim 4, characterized in that: The image drawing unit comprises: The method of converting the reflected stress wave signal, the diffracted wave signal and the transmitted stress wave signal into image data includes: The collected reflected stress wave signal, diffraction wave signal and transmitted stress wave signal are converted into discrete signals, and noise and unnecessary frequency components are removed by using filters, and key features of reflected sound waves and diffraction waves are extracted, wherein the key features include signal amplitude, time delay and frequency components. According to the speed of sound and signal propagation time, the signal is converted into distance to form a time-distance graph, and the time domain signal is converted into a frequency domain signal according to Fourier transform, and a spectrum graph is generated by combining the key features. According to the time-distance graph and the spectrum graph, an A-scan graph of the reflected stress wave signal, diffraction wave signal and transmitted stress wave signal is generated by using image generation software, and a B-scan image and a C-scan image are generated by scanning multiple A-scan images of the same stress wave signal, and a three-dimensional image is generated by combining multiple B-scan images.

6. The bridge underwater pile foundation stress wave detection system according to claim 5, characterized in that: The analysis module includes an image processing unit, a simulation calculation unit, a comparison detection unit and a pile foundation influence unit: The image processing unit comprises: According to the three-dimensional image obtained by the calculation module, a three-dimensional stereoscopic structural model of the underwater pile foundation of the bridge is established by using three-dimensional drawing software, and according to the B-scan image and C-scan image of the reflected stress wave signal and the transmitted stress wave signal, the cross section of the bridge pile foundation is divided into grid graphics by using a grid method, and represented by a three-dimensional rectangular coordinate system, and abnormal marking is performed at the corresponding position of the abnormal data of the bridge pile foundation, and the abnormal data includes abnormal values ​​in all images of the stress wave signal; The abnormal value corresponding to the abnormal mark position is clustered according to the detailed information feature data by k-means to obtain the first abnormal feature type to the mth abnormal feature type, where m>1. The size of the three-dimensional stereoscopic structure model of the bridge pile foundation, the grid size, the sound wave intensity of the abnormal mark position and the coordinates of the abnormal mark position are normalized and imported into the LTI path forward algorithm model to obtain the path forward calculation result, which includes: The path aggregation and path blank phenomena of the stress wave signal, the path aggregation phenomenon is the first abnormal area phenomenon, the path blank phenomenon is the second abnormal area phenomenon, and the path forward calculation result is used as the first image detection effect.

7. The bridge underwater pile foundation stress wave detection system according to claim 6, characterized in that: The size, grid size, sound wave intensity of the abnormal mark position and coordinates of the abnormal mark position of the three-dimensional stereoscopic structural model of the bridge pile foundation are normalized and then imported into the IRT velocity field inversion algorithm model to calculate the sectional surface tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The sectional surface tomography results are spliced ​​and combined into a three-dimensional histogram form to obtain the overall tomography results of the three-dimensional stereoscopic structural model of the bridge pile foundation. The detailed abnormal severity effect of the bridge pile foundation is obtained according to the color distribution in the overall tomography result, and the detailed abnormal severity effect is used as the second image detection effect.

8. The bridge underwater pile foundation stress wave detection system according to claim 7, characterized in that: The simulation calculation unit comprises: The first image detection effect and the second image detection effect are calculated by confidence simulation to obtain the abnormal sound wave intensity confidence interval of the bridge pile foundation, and the abnormal sound wave intensity confidence interval represents the internal quality detection condition of the bridge pile foundation. The confidence simulation model calculation process includes: The abnormal feature types corresponding to the first image detection effect and the second image detection effect are used as statistical samples for statistical analysis. The statistical analysis includes mean calculation and standard deviation calculation. The mean and standard deviation of the reflected stress wave signal intensity are obtained respectively. The normal stress wave signal threshold is set, and the confidence prediction interval of the reflected stress wave signal is calculated. The mathematical expression for calculating the confidence prediction interval is: ； in, is the confidence prediction interval, A is the statistical parameter corresponding to the probability distribution type, is the mean value of the reflected stress wave signal intensity, σ is the standard deviation of the reflected stress wave signal intensity, a is the normal reflected stress wave signal threshold, and the probability distribution types include normal distribution, binomial distribution, Poisson distribution and uniform distribution; The reflected stress wave signal strength that is higher or lower than the confidence prediction interval is an abnormal sound wave, and the corresponding abnormal sound wave strength is recorded at the corresponding abnormal mark position; The abnormal characteristic type of the abnormal sound wave is also recorded at the corresponding abnormal mark position; When the number of grids divided in the overall tomographic imaging result of the three-dimensional structural model of the bridge pile foundation is greater than a preset division threshold, a statistical parameter is selected according to the probability distribution type. ; If a is smaller, the statistical parameter A is larger, the confidence prediction interval is higher, and the second abnormal area has fewer abnormal phenomena; If a is larger, the statistical parameter A is smaller, the confidence prediction interval is lower, and the second abnormal area has more abnormal phenomena.

9. The bridge underwater pile foundation stress wave detection system according to claim 7, characterized in that: The comparison detection unit comprises: Collect a three-dimensional structural model of the underwater part of the historical bridge pile foundation and the received historical stress wave signal data, the historical stress wave signal data including the frequency spectrum and sound wave intensity of the reflected stress wave signal, the diffraction wave signal and the transmitted stress wave signal, compare the first image detection effect and the second image detection effect currently measured with the historical stress wave signal data through the calculation module, the image processing unit and the simulation calculation unit, and obtain the distinguishing features of the current bridge pile foundation and the underwater part of the historical bridge pile foundation, the distinguishing features including undetected abnormal conditions of the bridge pile foundation, and obtain the third image detection effect by passing the undetected abnormal conditions of the bridge pile foundation through the image processing unit and the simulation calculation unit.

10. The bridge underwater pile foundation stress wave detection system according to claim 6, characterized in that: The pile foundation impact unit comprises: Perform k-means clustering on the abnormal feature types corresponding to the detection effects included in the third image detection effect except the first image detection effect and the second image detection effect, to obtain the m+1th abnormal feature type ~ and the kth abnormal feature type, where k>m+1, m>1, and k is the total value of all abnormal feature types; The first image detection effect, the second image detection effect, the third image detection effect and the corresponding abnormal feature type recorded at the abnormal mark position are recorded in a dictionary form, and the pointer of the address in the dictionary form is pointed to the memory address position where the three-dimensional coordinates of the abnormal mark position are stored, and the head nodes of the chain queue are set to be the first abnormal feature type~the mth abnormal feature type~the kth abnormal feature type respectively, and the memory address positions where the three-dimensional coordinates of all abnormal mark positions are stored are respectively connected to the corresponding chain queue head nodes to form a chain queue form; All abnormal feature types represent a bridge pile foundation impact effect, which includes material defects, pile foundation structure defects and underwater biological attachment corrosion.

Images