Tunnel muon imaging system and method based on compressed sensing

Abstract

The application discloses a tunnel muon imaging system and method based on compressed sensing, relates to the technical field of tunnel muon imaging, and comprises a plurality of fixed muon detector arrays, a mobile muon detection device, a data acquisition and preprocessing module and a tunnel muon imaging server. The application realizes efficient, high-precision and low-cost detection of tunnel structure defects by fusing the strong penetration of muon imaging, the high-efficiency data processing capacity of compressed sensing and the fixed and mobile cooperative detection architecture, and solves the problems of limited penetration depth, high cost and low efficiency in traditional technologies. Through the cooperation of fixed and mobile detection, the real-time data verification and supplement mechanism and the fusion of compressed sensing theory, the problems of low efficiency, limited coverage and high cost in traditional tunnel detection are solved, and efficient, high-precision three-dimensional imaging of internal defects of a tunnel structure is realized.

Description

Technical Field

[0001] This invention relates to the field of tunnel muon imaging technology, and in particular to a tunnel muon imaging system and method based on compressed sensing. Background Technology

[0002] Traditional tunnel inspection technologies mainly include railcar ground-penetrating radar, railcar ultrasonic testing, and railcar X-ray / gamma-ray testing;

[0003] The limitations of the railcar-based ground-penetrating radar are: limited penetration depth and insufficient resolution for high-density concrete structures.

[0004] Ultrasonic testing using railcars requires contact operation, making it difficult to cover a large area of ​​the tunnel surface.

[0005] Railcar X-ray / gamma-ray technology poses radiation risks and has high equipment costs. Summary of the Invention

[0006] To address the technical challenges of tunnel detection, this invention provides a tunnel muon imaging system and method based on compressed sensing. The technical solution adopted is as follows:

[0007] A tunnel muon imaging system based on compressed sensing includes multiple fixed muon detector arrays, a mobile muon detection device, a data acquisition and preprocessing module, and a tunnel muon imaging server. The multiple fixed muon detector arrays are installed at equal intervals at the top and bottom of the tunnel. The mobile muon detection device moves within the tunnel and wirelessly communicates with the muon detection end transceiver modules of each of the fixed muon detector arrays to acquire and store muon detection data. The data acquisition and preprocessing module is installed on the mobile muon detection device and preprocesses the muon detection data from the current fixed muon detector array. In the processing and analysis, when the muon detection data of the current fixed muon detector array is determined to be unqualified, the mobile muon detection device continues to move directly below the fixed muon detector array. The mobile muon detector arrays built into the top and bottom of the mobile muon detector device replace the current fixed muon detector array for detection, and the muon detection data is saved. After the mobile muon detection device completes the collection of muon detection data for the entire tunnel, it transmits all the muon detection data to the tunnel muon imaging server. The tunnel muon imaging server outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology.

[0008] Optionally, the fixed muon detector array includes a tunnel top muon detector array, a tunnel bottom muon detector array, a muon detector data acquisition unit, and a muon detector end wireless transceiver module. The tunnel top muon detector array is installed at the top of the tunnel, the tunnel bottom muon detector array is installed at the bottom of the tunnel, and the muon detector end wireless transceiver module is installed between two adjacent fixed muon detector arrays. The signal input terminal of the muon detector end wireless transceiver module is communicatively connected to the signal output terminals of the tunnel top muon detector array and the tunnel bottom muon detector array, respectively.

[0009] By adopting the above technical solution, the strong penetrability of natural muons (muons) in cosmic rays (which can penetrate rock layers at the kilometer level) is utilized. When muons pass through the tunnel structure, their attenuation and scattering characteristics are related to the density distribution inside the structure (e.g., the density is low in voids and cracked areas, and the attenuation of muons is weak).

[0010] The muon detector array at the top of the tunnel and the muon detector array at the bottom of the tunnel form a dual-plane detection structure. By recording the incident direction, energy loss and trajectory of muons, the density distribution of the area to be measured can be inverted.

[0011] Multiple fixed muon detector arrays are arranged at intervals at the top and bottom of the tunnel to achieve large-scale, routine data collection. The collected muon data is transmitted in real time to the mobile muon detection device through the wireless transceiver module of the muon detector end.

[0012] The mobile muon detection device moves via a railcar and uses an onboard wireless transceiver module to receive data from the fixed muon detector array in a close-range manner, avoiding data loss during data acquisition within the data tunnel. The data acquisition and preprocessing module also performs real-time assessment of data quality.

[0013] When the data from the fixed muon detector array is unqualified, the mobile muon detector array (roof / under-vehicle array) collects supplementary data at the corresponding location to ensure data integrity and provide reliable input for subsequent compressed sensing processing.

[0014] Internal defects in tunnel structures (such as cavities and cracks) exhibit sparsity under wavelet basis. The tunnel muon imaging server efficiently recovers the three-dimensional density distribution from the acquired muon data through compressed sensing algorithms (such as random Gaussian matrix sampling and L1 norm minimization reconstruction), thereby achieving defect imaging.

[0015] By integrating the strong penetration of muon imaging, the efficient data processing capabilities of compressed sensing, and the fixed-mobile collaborative detection architecture, efficient, high-precision, and low-cost detection of tunnel structural defects has been achieved, solving the problems of limited penetration depth, high cost, and low efficiency in traditional technologies.

[0016] Optionally, the fixed muon detector array also includes a locator for calibrating the location of the wireless transceiver module at the muon detector end.

[0017] Optionally, the mobile muon detection device includes a railcar, a mobile muon detector array, a positioning receiver, an onboard wireless transceiver module, an onboard memory, a data acquisition and preprocessing module, and a chip-based railcar controller. Tracks are installed on both sides of the tunnel, and the railcar travels along these tracks on both sides of the tunnel. The mobile muon detector array includes a roof-mounted muon detector array and an under-mounted muon detector array, which are respectively installed on the top and bottom of the railcar. The positioning receiver is installed on the railcar to detect the position of the locator. The onboard wireless transceiver module wirelessly communicates with the muon detection terminal wireless transceiver module to exchange muon detection data. The onboard memory is communicatively connected to the onboard wireless transceiver module. The data acquisition and preprocessing module is communicatively connected to the memory. The railcar controller is communicatively connected to both the positioning receiver and the data acquisition and preprocessing module, and controls the actions of the railcar, the mobile muon detector array, and the onboard wireless transceiver module.

[0018] Optionally, the positioning receiver receives the position data from the locator. When the railcar controller determines that the current position is less than the position of the corresponding fixed muon detector array, the railcar controller controls the on-board wireless transceiver module to receive the muon detection data sent by the wireless transceiver module of the corresponding fixed muon detector array.

[0019] Optionally, the data acquisition and preprocessing module includes a buffer and a data analysis chip. The buffer is communicatively connected to the data output terminal of the vehicle-mounted wireless transceiver module, and the data analysis chip is communicatively connected to the buffer and interacts with the railcar controller to analyze the results.

[0020] By adopting the above technical solution, multiple fixed muon detector arrays are arranged at a preset interval at the top and bottom of the tunnel. Each array includes a muon detector array at the top of the tunnel and a muon detector array at the bottom of the tunnel, which can capture information such as muon trajectory and energy loss passing through the tunnel structure in real time.

[0021] The locator is used to calibrate the position of the wireless transceiver module of the muon detector, ensuring that the data is accurately correlated with the tunnel mileage coordinates, and providing a benchmark for spatial positioning of subsequent 3D imaging.

[0022] The wireless transceiver module at the muon detector wirelessly transmits the collected raw data to the mobile muon detector, achieving wide-area, contactless data coverage.

[0023] The mobile muon detection device's railcar travels along the tracks on both sides of the tunnel. The positioning receiver receives the position signal from the locator in real time. When the railcar controller determines that the distance between the current position and the nth group of fixed muon detector arrays is less than a set threshold, it controls the onboard wireless transceiver module to establish a connection with the wireless transceiver module of the muon detector end of that group, receives data, and stores it in the onboard memory.

[0024] The data acquisition and preprocessing module's buffer temporarily stores the received data, while the data analysis chip performs real-time preprocessing on the data (such as noise reduction and trajectory integrity verification) and feeds back the quality judgment result (qualified / unqualified) to the track vehicle controller.

[0025] If the data analysis chip determines that the data from the fixed array is unqualified, the track vehicle controller drives the track vehicle to stop at the corresponding position and activates the mobile muon detector array (roof muon detector array and under-vehicle muon detector array) to supplement the data collection and ensure that there is no missing data in the key areas.

[0026] After the full data stored in the vehicle-mounted memory (including qualified data from the fixed array and supplementary data from the mobile array 21) is transmitted to the tunnel muon imaging server, the server performs compressed sampling and reconstruction of the data based on compressed sensing theory (utilizing the sparsity of the tunnel structure), and finally outputs a three-dimensional density distribution image to achieve defect localization.

[0027] By combining fixed and mobile detection, and employing a real-time data verification and supplementation mechanism, and integrating compressed sensing theory, the problems of low efficiency, limited coverage, and high cost in traditional tunnel detection have been solved, achieving efficient and high-precision three-dimensional imaging of internal defects in tunnel structures.

[0028] Optionally, the mobile muon detection device also includes a data output interface, which is communicatively connected to an on-board storage device for outputting muon detection data to a tunnel muon imaging server.

[0029] Optionally, the tunnel muon imaging server includes a data input interface, a server-side memory, a computer, and a display. The server-side memory is communicatively connected to the data input interface. When the data input interface is connected to the data output interface, the server-side memory collects all muon detection data stored in the vehicle-mounted memory. The computer is communicatively connected to the server-side memory and outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology. The display is communicatively connected to the computer and displays the three-dimensional reconstructed image of the tunnel.

[0030] By adopting the above technical solution, the on-board memory of the mobile muon detection device stores all muon detection data (including qualified data from multiple fixed muon detector arrays and supplementary data from the mobile muon detector array). The data output interface serves as the external interaction port of the on-board memory, responsible for exporting the stored data.

[0031] After the data input interface and data output interface of the tunnel muon imaging server are connected, the server-side storage reads all the data in the vehicle-mounted storage through this link, completing the data migration from mobile acquisition to fixed processing and ensuring that the original data is completely transmitted to the server.

[0032] The computer retrieves data from the server-side memory and processes the data based on compressed sensing theory: by utilizing the sparsity of the tunnel structure (such as concrete matrix and defect areas) under wavelet basis, the data is compressed and sampled using a random Gaussian matrix (to reduce the amount of data), and then the OMP algorithm combined with Monte Carlo simulation is used to reconstruct the sparse signal and invert the density distribution inside the tunnel.

[0033] The reconstructed three-dimensional density distribution image is visualized on a monitor, enabling spatial positioning and morphological display of defects (such as cavities and cracks) inside the tunnel.

[0034] The tunnel muon imaging method based on compressed sensing uses a tunnel muon imaging system based on compressed sensing to perform three-dimensional imaging of the target tunnel, including the following steps:

[0035] Step 1: Place the mobile muon detection device on the track at the tunnel starting point. The track moves at a constant speed along the tunnel track. The positioning receiver receives the position signals of multiple locators from multiple fixed muon detector arrays in real time.

[0036] Step 2: When the railcar controller determines that the distance between the positioning receiver and the locator is less than the set minimum distance threshold, it controls the on-board wireless transceiver module to receive the muon detection data sent by the wireless transceiver module of the muon detection end of the fixed muon detector array corresponding to the locator, and transmits it to the on-board memory for storage.

[0037] Step 3: The data analysis chip reads the raw data of the corresponding fixed muon detector array from the buffer, performs preprocessing, removes electron interference signals in cosmic rays based on energy thresholds, verifies the integrity of the trajectory, matches the incident and outgoing trajectories by timestamps, determines whether there is a break or abnormal deflection, and transmits the preprocessing results to the track vehicle controller.

[0038] Step 4: If the judgment result in Step 3 is that there is no breakage or abnormal deflection, the track vehicle controller records the corresponding fixed muon detector array data as qualified and controls the track vehicle to continue to the next fixed muon detector array until all muon detection data is collected.

[0039] Step 5: The data input interface and data output interface of the tunnel muon imaging server are connected. The server-side storage collects all the muon detection data stored in the vehicle-mounted storage. The computer outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology.

[0040] Optionally, if the result of step 3 indicates the presence of a break or abnormal deflection, then a mobile sampling step is performed:

[0041] The railcar controller uses a closed-loop position control system with a positioner and a position receiver to park the railcar directly below the corresponding fixed muon detector array.

[0042] The mobile muon detector array is activated to replace the corresponding fixed muon detector array in collecting muon data; the supplementary collected data is written into the on-board memory, and the data analysis chip repeats step 3 to preprocess the supplementary data to confirm that there is no breakage or abnormal deflection. The track vehicle continues to move and continues to execute steps 4 and 5.

[0043] In summary, the present invention has at least one of the following beneficial technical effects:

[0044] This invention provides a tunnel muon imaging system and method based on compressed sensing. Utilizing the strong penetrating power of natural muons in cosmic rays, the attenuation and scattering characteristics of muons as they pass through a tunnel structure are related to the density distribution within the structure. A muon detector array at the top and bottom of the tunnel forms a dual-plane detection structure. By recording the incident direction, energy loss, and trajectory of the muons, the density distribution of the area under test can be inverted.

[0045] Multiple fixed muon detector arrays are arranged at intervals at the top and bottom of the tunnel to achieve large-scale, routine data collection. The collected muon data is transmitted in real time to the mobile muon detection device through the wireless transceiver module of the muon detector end.

[0046] By integrating the strong penetration of muon imaging, the efficient data processing capabilities of compressed sensing, and the fixed-mobile collaborative detection architecture, efficient, high-precision, and low-cost detection of tunnel structural defects has been achieved, solving the problems of limited penetration depth, high cost, and low efficiency in traditional technologies.

[0047] By combining fixed and mobile detection, and employing a real-time data verification and supplementation mechanism, and integrating compressed sensing theory, the problems of low efficiency, limited coverage, and high cost in traditional tunnel detection have been solved, achieving efficient and high-precision three-dimensional imaging of internal defects in tunnel structures. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the electrical component connection principle of the tunnel muon imaging system based on compressed sensing of the present invention;

[0049] Figure 2 This is a schematic diagram illustrating the structural principle of the tunnel muon imaging system based on compressed sensing, as described in this invention, when deployed inside a tunnel.

[0050] Explanation of reference numerals in the attached figures: 1. Fixed muon detector array; 11. Tunnel top muon detector array; 12. Tunnel bottom muon detector array; 13. Muon detector data acquisition unit; 14. Muon detector wireless transceiver module; 15. Positioner; 2. Mobile muon detection device; 211. Vehicle roof muon detector array; 212. Vehicle bottom muon detector array; 22. Railcar; 23. Positioning receiver; 24. Vehicle-mounted wireless transceiver module; 25. Vehicle-mounted storage; 26. Railcar controller; 27. Data output interface; 3. Data acquisition and preprocessing module; 31. Buffer; 32. Data analysis chip; 4. Tunnel muon imaging server; 41. Data input interface; 42. Server-side storage; 43. Computer; 44. Display. Detailed Implementation

[0051] The present invention will be further described in detail below with reference to the accompanying drawings.

[0052] This invention discloses a tunneling muon imaging system and method based on compressed sensing.

[0053] Reference Figure 1 and Figure 2Example 1: A tunnel muon imaging system based on compressed sensing includes multiple fixed muon detector arrays 1, a mobile muon detection device 2, a data acquisition and preprocessing module 3, and a tunnel muon imaging server 4. The multiple fixed muon detector arrays 1 are installed at equal intervals at the top and bottom of the tunnel. The mobile muon detection device 2 moves within the tunnel and wirelessly communicates with the muon detection end wireless transceiver modules 14 of the multiple fixed muon detector arrays 1 to acquire and store muon detection data. The data acquisition and preprocessing module 3 is installed on the mobile muon detection device 2 and processes the muon detection data of the current fixed muon detector array 1. Preprocessing analysis is performed. When the muon detection data of the current fixed muon detector array 1 is determined to be unqualified, the mobile muon detection device 2 continues to move directly below the fixed muon detector array 1. The mobile muon detector arrays on the top and bottom of the mobile muon detection device 2 replace the current fixed muon detector array 1 for detection and save the muon detection data. After the mobile muon detection device 2 completes the collection of muon detection data for the entire tunnel, it transmits all the muon detection data to the tunnel muon imaging server 4. The tunnel muon imaging server 4 outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology.

[0054] Example 2: The fixed muon detector array 1 includes a tunnel top muon detector array 11, a tunnel bottom muon detector array 12, a muon detector data acquisition unit 13, and a muon detector end wireless transceiver module 14. The tunnel top muon detector array 11 is installed at the top of the tunnel, the tunnel bottom muon detector array 12 is installed at the bottom of the tunnel, and the muon detector end wireless transceiver module 14 is installed between two adjacent fixed muon detector arrays 1. The signal input terminal of the muon detector end wireless transceiver module 14 is communicatively connected to the signal output terminals of the tunnel top muon detector array 11 and the tunnel bottom muon detector array 12, respectively.

[0055] Taking advantage of the strong penetrability of natural muons (muons) in cosmic rays (which can penetrate rock layers at the kilometer level), when muons pass through a tunnel structure, their attenuation and scattering characteristics are related to the density distribution inside the structure (e.g., the density is low in voids and cracks, and the attenuation of muons is weak).

[0056] The muon detector array 11 at the top of the tunnel and the muon detector array 12 at the bottom of the tunnel form a dual-plane detection structure. By recording the incident direction, energy loss and trajectory of muons, the density distribution of the area to be measured is inverted.

[0057] Multiple fixed muon detector arrays 1 are arranged at intervals at the top and bottom of the tunnel to achieve large-scale, routine data collection. The collected muon data is transmitted in real time to the mobile muon detection device 2 through the wireless transceiver module 14 of the muon detection end.

[0058] The mobile muon detection device 2 moves via a railcar 22 and uses the onboard wireless transceiver module 24 to receive data from the fixed muon detector array 1 in a close-range manner, avoiding data loss in the data tunnel. The data quality is judged in real time through the data acquisition and preprocessing module 3.

[0059] When the data from the fixed muon detector array 1 is unqualified, the mobile muon detector array (roof / under-vehicle array) collects supplementary data at the corresponding location to ensure data integrity and provide reliable input for subsequent compressed sensing processing.

[0060] Internal defects in tunnel structures (such as cavities and cracks) exhibit sparsity under wavelet basis. The tunnel muon imaging server 4 efficiently recovers the three-dimensional density distribution from the collected muon data through compressed sensing algorithms (such as random Gaussian matrix sampling and L1 norm minimization reconstruction), thereby achieving defect imaging.

[0061] By integrating the strong penetration of muon imaging, the efficient data processing capabilities of compressed sensing, and the fixed-mobile collaborative detection architecture, efficient, high-precision, and low-cost detection of tunnel structural defects has been achieved, solving the problems of limited penetration depth, high cost, and low efficiency in traditional technologies.

[0062] In Example 3, the fixed muon detector array 1 also includes a locator 15, which is used to calibrate the position of the wireless transceiver module 14 at the muon detector end.

[0063] Example 4: The mobile muon detection device 2 includes a railcar 22, a mobile muon detector array, a positioning receiver 23, an onboard wireless transceiver module 24, an onboard memory 25, a data acquisition and preprocessing module 3, and a chip-based railcar controller 26. Tracks are installed on both sides of the tunnel, and the railcar 22 travels on the tracks on both sides of the tunnel. The mobile muon detector array includes a roof-mounted muon detector array 211 and a floor-mounted muon detector array 212, which are respectively installed on the tracks. The top and bottom of the vehicle 22 are equipped with positioning receivers 23, which are used to detect the position of the locator 15. The vehicle-mounted wireless transceiver module 24 wirelessly communicates with the muon detector wireless transceiver module 14 to exchange muon detection data. The vehicle-mounted memory 25 is connected to the vehicle-mounted wireless transceiver module 24. The data acquisition and preprocessing module 3 is connected to the memory 25. The vehicle controller 26 is connected to the positioning receiver 23 and the data acquisition and preprocessing module 3 respectively, and controls the actions of the vehicle 22, the mobile muon detector array and the vehicle-mounted wireless transceiver module 24.

[0064] In Example 5, the positioning receiver 23 receives the position data of the locator 15. When the railcar controller 26 determines that the current position is less than the position of the corresponding fixed muon detector array 1, the railcar controller 26 controls the on-board wireless transceiver module 24 to receive the muon detection data sent by the muon detection end wireless transceiver module 14 of the corresponding fixed muon detector array 1.

[0065] Example 6: The data acquisition and preprocessing module 3 includes a buffer 31 and a data analysis chip 32. The buffer 31 is communicatively connected to the data output terminal of the vehicle-mounted wireless transceiver module 24. The data analysis chip 32 is communicatively connected to the buffer 31 and interacts with the railcar controller 26 to analyze the results.

[0066] Multiple fixed muon detector arrays 1 are arranged at preset intervals at the top and bottom of the tunnel. Each array includes a muon detector array 11 at the top of the tunnel and a muon detector array 12 at the bottom of the tunnel. They can capture information such as muon trajectories and energy losses passing through the tunnel structure in real time.

[0067] The locator 15 is used to calibrate the position of the wireless transceiver module 14 of the muon detector, ensuring that the data is accurately correlated with the tunnel mileage coordinates, and providing a benchmark for spatial positioning of subsequent three-dimensional imaging.

[0068] The wireless transceiver module 14 of the muon detector wirelessly transmits the collected raw data to the mobile muon detector 2, achieving wide-range, contactless data coverage.

[0069] The track vehicle 22 of the mobile muon detection device 2 travels along the tracks on both sides of the tunnel. The positioning receiver 23 receives the position signal of the locator 15 in real time. When the track vehicle controller 26 determines that the distance between the current position and the nth group of fixed muon detector array 1 is less than the set threshold, it controls the on-board wireless transceiver module 24 to establish a connection with the wireless transceiver module 14 of the muon detection end of that group, receives data and stores it in the on-board memory 25.

[0070] The buffer 31 of the data acquisition and preprocessing module 3 temporarily stores the received data, the data analysis chip 32 performs real-time preprocessing on the data (such as noise reduction and trajectory integrity verification), and feeds back the quality judgment result (qualified / unqualified) to the track vehicle controller 26.

[0071] If the data analysis chip 32 determines that the data of the fixed array 1 is unqualified, the track vehicle controller 26 drives the track vehicle 22 to stop at the corresponding position and starts the mobile muon detector array (the muon detector array 211 on the roof and the muon detector array 212 on the bottom of the vehicle) to supplement the data collection and ensure that there is no missing data in the key areas.

[0072] After the full data (including qualified data from the fixed array 1 and supplementary data from the mobile array 21) stored in the vehicle-mounted memory 25 is transmitted to the tunnel muon imaging server 4, the server performs compressed sampling and reconstruction of the data based on the compressed sensing theory (utilizing the sparsity of the tunnel structure), and finally outputs a three-dimensional density distribution image to achieve defect localization.

[0073] By combining fixed and mobile detection, and employing a real-time data verification and supplementation mechanism, and integrating compressed sensing theory, the problems of low efficiency, limited coverage, and high cost in traditional tunnel detection have been solved, achieving efficient and high-precision three-dimensional imaging of internal defects in tunnel structures.

[0074] In Example 7, the mobile muon detection device 2 also includes a data output interface 27, which is communicatively connected to the on-board memory 25 and is used to output muon detection data to the tunnel muon imaging server 4.

[0075] Example 8: The tunnel muon imaging server 4 includes a data input interface 41, a server-side memory 42, a computer 43, and a display 44. The server-side memory 42 is communicatively connected to the data input interface 41. When the data input interface 41 is connected to the data output interface 27, the server-side memory 42 collects all muon detection data stored in the vehicle-mounted memory 25. The computer 43 is communicatively connected to the server-side memory 42. The computer 43 outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology. The display 44 is communicatively connected to the computer 43 and displays the three-dimensional reconstructed image of the tunnel.

[0076] The on-board storage 25 of the mobile muon detection device 2 stores all muon detection data (including qualified data from multiple fixed muon detector arrays 1 and supplementary data from the mobile muon detector array). The data output interface 27 serves as the external interaction port of the on-board storage 25, responsible for exporting the stored data.

[0077] After the data input interface 41 of the tunnel muon imaging server 4 is connected to the data output interface 27, the server-side memory 42 reads all the data in the vehicle-mounted memory 25 through this link, completing the data migration from mobile acquisition to fixed processing, and ensuring that the original data is completely transmitted to the server.

[0078] Computer 43 retrieves data from server-side memory 42 and processes the data based on compressed sensing theory: utilizing the sparsity of tunnel structure (such as concrete matrix, defect area) under wavelet basis, the data is compressed and sampled using a random Gaussian matrix (reducing the amount of data), and then the OMP algorithm combined with Monte Carlo simulation is used to reconstruct the sparse signal and invert the density distribution inside the tunnel.

[0079] The reconstructed three-dimensional density distribution image is visualized on a monitor 44, enabling spatial positioning and morphological display of defects (such as cavities and cracks) inside the tunnel.

[0080] Example 9: A tunnel muon imaging method based on compressed sensing. This method uses a tunnel muon imaging system based on compressed sensing to perform three-dimensional imaging of a target tunnel, including the following steps:

[0081] Step 1: Place the mobile muon detection device track vehicle 22 on the track at the tunnel starting point. The track vehicle 22 moves at a constant speed along the tunnel track. The positioning receiver 23 receives the position signals of multiple locators 15 of multiple sets of fixed muon detector array 1 in real time.

[0082] Step 2: When the railcar controller 26 determines that the distance between the positioning receiver 23 and the locator 15 is less than the set minimum distance threshold, it controls the on-board wireless transceiver module 24 to receive the muon detection data sent by the muon detection end wireless transceiver module 14 of the fixed muon detector array 1 corresponding to the locator 15, and transmits it to the on-board memory 25 for storage.

[0083] Step 3: The data analysis chip 32 reads the raw data of the corresponding fixed muon detector array 1 from the buffer 31, performs preprocessing, removes electron interference signals in cosmic rays based on energy threshold, verifies the integrity of the trajectory, matches the incident and outgoing trajectories by timestamp, determines whether there is a break or abnormal deflection, and transmits the preprocessing results to the track vehicle controller 26.

[0084] Step 4: If the judgment result in step 3 is that there is no breakage or abnormal deflection, then the track vehicle controller 26 records that the data of the corresponding fixed muon detector array 1 is qualified, and controls the track vehicle 22 to continue to the next set of fixed muon detector array 1 until all muon detection data is collected.

[0085] Step 5: The data input interface 41 of the tunnel muon imaging server 4 is connected to the data output interface 27. The server-side memory 42 collects all the muon detection data stored in the vehicle-mounted memory 25. The computer 43 outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology.

[0086] Example 10: If the judgment result in step 3 indicates the presence of a break or abnormal deflection, then the mobile sampling step is executed:

[0087] The track vehicle controller 26 uses the position closed-loop control of the positioner 15 and the position receiver 23 to park the track vehicle 22 directly below the corresponding fixed muon detector array 1;

[0088] The mobile muon detector array is activated to replace the corresponding fixed muon detector array 1 in collecting muon data; the supplementary collected data is written into the on-board memory 25, and at the same time, the data analysis chip 32 repeats step 3 to preprocess the supplementary data to confirm that there is no breakage or abnormal deflection. The track vehicle 22 continues to move and continues to execute steps 4 and 5.

[0089] The following specific embodiments illustrate the implementation principle of the present invention:

[0090] Taking the inspection of a railway tunnel section from K1+000 to K1+500 as an example;

[0091] The target tunnel is a double-track railway tunnel, 500 meters long and 8 meters in diameter. It is necessary to inspect its interior for defects such as cavities and water seepage. A tunnel muon imaging system based on compressed sensing will be used. The specific implementation steps are as follows:

[0092] System deployment and data acquisition:

[0093] 1. Fixed array deployment:

[0094] Ten fixed muon detector arrays (1) are deployed every 50 meters at the top and bottom of the tunnel. Each array contains:

[0095] The tunnel has a muon detector array 11 at the top (8 detector units, spaced 1 meter apart) and a muon detector array 12 at the bottom (8 detector units, spaced 1 meter apart). The muon detectors are high-sensitivity muon detectors, which can be structured as scintillators plus photomultiplier tubes.

[0096] Muon detector wireless transceiver module 14 (deployed between two adjacent arrays);

[0097] Locator 15 (calibrates the tunnel mileage coordinates of each array group, such as K1+000 for the first group and K1+050 for the second group).

[0098] 2. Preparation of mobile equipment:

[0099] The track vehicle 22 of the mobile muon detection device 2 is placed on the track at the tunnel starting point (K1+000). The muon detector array 211 (6 units) on the top of the vehicle and the muon detector array 212 (6 units) on the bottom of the vehicle are started and preheated. The buffer 31 of the data acquisition and preprocessing module 3 is cleared, and the data analysis chip 32 is loaded with the data quality threshold (signal-to-noise ratio ≥20dB, trajectory integrity ≥90%).

[0100] 3. Data Acquisition and Supplementary Exploration:

[0101] The track vehicle 22 travels at a speed of 2km / h. The positioning receiver 23 receives the signal from the locator 15. When the distance to the nth fixed array 1 is ≤5 meters, the vehicle-mounted wireless transceiver module 24 connects with the muon detector wireless transceiver module 14, receives the muon data (including incident direction, energy loss, and timestamp), stores it in the vehicle-mounted memory 25, and synchronously transmits it to the buffer 31.

[0102] The data analysis chip 32 preprocesses the data in the buffer 31: it removes cosmic ray electron interference signals (signals with energy loss <100MeV) and verifies the integrity of the trajectory. The data from the third fixed array 1 is deemed unqualified due to trajectory breakage (integrity 82%). The track vehicle controller 26 drives the track vehicle 22 to stop at K1+150 (directly below the third array), and starts the roof muon detector array 211 and the undercarriage muon detector array 212 to supplement the data acquisition for 1 hour. The data is stored in the on-board memory 25 (labeled as supplementary_K1+150).

[0103] III. Data transmission to the server:

[0104] After the track vehicle 22 returns to the starting point, it connects the data output interface 27 (using a fiber optic interface) of the mobile muon detection device 2 with the data input interface 41 of the tunnel muon imaging server 4. The server-side memory 42 reads the full amount of data (10 sets of fixed data + 1 set of supplementary data, totaling 80GB) from the on-board memory 25.

[0105] IV. Three-dimensional imaging processing based on compressed sensing (core steps, including component numbers):

[0106] The following steps are executed by computer 43 of tunnel muon imaging server 4, and the results are finally output through display 44:

[0107] 1. Data preprocessing (computer 43 calls data from server-side memory 42):

[0108] The original data is timestamped and synchronized (based on the coordinate information of locator 15, the data is accurately associated with the tunnel mileage).

[0109] Trajectory correction: Introducing the Earth's magnetic field model to correct the muon path deflection error (deviation ≤ 0.1°) and ensure that the trajectory matches the tunnel structure spatially.

[0110] 2. Sparse signal modeling:

[0111] The preprocessed data is subjected to wavelet transform (using db4 wavelet basis). Taking advantage of the sparsity of the tunnel structure—the concrete matrix has uniform density (non-sparse), while the density of voids and seepage areas changes abruptly (sparse feature)—the data is converted into a sparse coefficient matrix (dimension 1024×1024).

[0112] 3. Compressed sampling:

[0113] A random Gaussian matrix (512×1024 dimensions) is generated as the measurement matrix. The sparse coefficient matrix is ​​compressed and sampled to retain 50% of the key information (the data volume is compressed from 80GB to 40GB), reducing the computational burden.

[0114] 4. Signal reconstruction (OMP algorithm + Monte Carlo simulation):

[0115] The Orthogonal Matching Pursuit (OMP) algorithm is used for iterative optimization: the residual is initialized with compressed sampled data, and in each iteration, the wavelet basis atom (matching defect feature) most related to the residual is selected to update the reconstructed signal and residual. After 30 iterations, the preliminary density distribution matrix is ​​obtained.

[0116] Monte Carlo simulation correction: Simulates the scattering path of muons in tunnel concrete to compensate for edge errors (such as blurred defect boundaries) in the initial reconstruction results, thereby improving the density resolution to 0.1 meters.

[0117] 5. 3D Image Generation:

[0118] The density distribution matrix is ​​mapped to a 3D mesh model based on tunnel mileage (K1+000-K1+500) and cross-sectional coordinates (top, bottom, sidewalls), where:

[0119] The density of normal concrete is marked as 2.4 g / cm³ (blue);

[0120] The density of the void area is <1.5g / cm³ (red, such as the 0.3-meter diameter void found in the dome at K1+150).

[0121] The density of the seepage area is 2.0-2.2 g / cm³ (yellow, such as a 0.5 m × 0.8 m seepage zone on the sidewall at K1+300).

[0122] 6. Results visualization:

[0123] Computer 43 transmits the 3D mesh model to display 44, displaying it in a cross-sectional slice + 3D walkthrough mode:

[0124] Cross-sectional slices: Generate a cross-sectional image every 10 meters along the tunnel mileage, and mark the location and size of defects;

[0125] 3D walkthrough: Allows for dynamic viewing of the spatial distribution of defects throughout the tunnel, supporting zoom and rotation operations.

[0126] 1. Detection efficiency: The detection time for a 500-meter tunnel is 6 hours (including 4 hours for data acquisition and 2 hours for imaging processing), which is 67% more efficient than the traditional Muzi imaging system (24 hours);

[0127] 2. Imaging accuracy: Defect location error ≤ 0.5 meters, resolution 0.1 meters, successfully identified small cavities with a diameter of 0.3 meters;

[0128] 3. Cost advantage: The number of detection units in the fixed array 1 is reduced by 40% compared to the fully dense arrangement, resulting in a significant reduction in hardware costs.

[0129] This embodiment achieves efficient and high-precision three-dimensional imaging of tunnel defects through the coordinated acquisition of a fixed muon detector array 1 and a mobile muon detection device 2, combined with compressed sensing processing by a tunnel muon imaging server 4, thus verifying the feasibility of the technical solution.

[0130] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A tunneling muon imaging system based on compressed sensing, characterized in that: The system includes multiple fixed muon detector arrays (1), a mobile muon detection device (2), a data acquisition and preprocessing module (3), and a tunnel muon imaging server (4). The multiple fixed muon detector arrays (1) are installed at equal intervals at the top and bottom of the tunnel. The mobile muon detection device (2) moves within the tunnel and wirelessly communicates with the muon detection end wireless transceiver module (14) of the multiple fixed muon detector arrays (1) to collect and store muon detection data. The data acquisition and preprocessing module (3) is installed on the mobile muon detection device (2) and preprocesses the muon detection data of the current fixed muon detector array (1). Analysis shows that when the muon detection data of the current fixed muon detector array (1) is unqualified, the mobile muon detection device (2) continues to move to the bottom of the fixed muon detector array (1) and uses the mobile muon detector arrays on the top and bottom of the mobile muon detection device (2) to replace the current fixed muon detector array (1) for detection and save the muon detection data. After the mobile muon detection device (2) completes the collection of muon detection data for the entire tunnel, it transmits all the muon detection data to the tunnel muon imaging server (4). The tunnel muon imaging server (4) outputs a three-dimensional reconstructed image of the tunnel based on compressed sensing processing and three-dimensional imaging technology.

2. The tunneling muon imaging system based on compressed sensing according to claim 1, characterized in that: The fixed muon detector array (1) includes a tunnel top muon detector array (11), a tunnel bottom muon detector array (12), a muon detector data acquisition unit (13), and a muon detector end wireless transceiver module (14). The tunnel top muon detector array (11) is installed at the top of the tunnel, the tunnel bottom muon detector array (12) is installed at the bottom of the tunnel, and the muon detector end wireless transceiver module (14) is installed between two adjacent fixed muon detector arrays (1). The signal input terminal of the muon detector end wireless transceiver module (14) is connected to the signal output terminals of the tunnel top muon detector array (11) and the tunnel bottom muon detector array (12), respectively.

3. The tunneling muon imaging system based on compressed sensing according to claim 2, characterized in that: The fixed muon detector array (1) also includes a locator (15) for calibrating the position of the muon detector wireless transceiver module (14).

4. The tunneling muon imaging system based on compressed sensing according to claim 3, characterized in that: The mobile muon detection device (2) includes a railcar (22), a mobile muon detector array, a positioning receiver (23), an onboard wireless transceiver module (24), an onboard memory (25), a data acquisition and preprocessing module (3), and a chip-based railcar controller (26). Tracks are installed on both sides of the tunnel, and the railcar (22) travels on the tracks on both sides of the tunnel. The mobile muon detector array includes a roof-mounted muon detector array (211) and a floor-mounted muon detector array (212), which are respectively installed on the top and bottom of the railcar (22). At the bottom, the positioning receiver (23) is installed on the railcar (22) to detect the position of the locator (15). The vehicle-mounted wireless transceiver module (24) wirelessly interacts with the muon detector wireless transceiver module (14) to exchange muon detection data. The vehicle-mounted memory (25) is connected to the vehicle-mounted wireless transceiver module (24). The data acquisition and preprocessing module (3) is connected to the vehicle-mounted memory (25). The railcar controller (26) is connected to the positioning receiver (23) and the data acquisition and preprocessing module (3) respectively, and controls the actions of the railcar (22), the mobile muon detector array and the vehicle-mounted wireless transceiver module (24).

5. The tunneling muon imaging system based on compressed sensing according to claim 4, characterized in that: The positioning receiver (23) receives the position data of the locator (15). When the railcar controller (26) determines that the current position is less than the set minimum distance threshold of the corresponding fixed muon detector array (1), the railcar controller (26) controls the on-board wireless transceiver module (24) to receive the muon detection data sent by the muon detection end wireless transceiver module (14) of the corresponding fixed muon detector array (1).

6. The tunneling muon imaging system based on compressed sensing according to claim 5, characterized in that: The data acquisition and preprocessing module (3) includes a buffer (31) and a data analysis chip (32). The buffer (31) is connected to the data output terminal of the vehicle-mounted wireless transceiver module (24). The data analysis chip (32) is connected to the buffer (31) and interacts with the railcar controller (26) to analyze the results.

7. The tunneling muon imaging system based on compressed sensing according to claim 6, characterized in that: The mobile muon detection device (2) also includes a data output interface (27), which is communicatively connected to the vehicle-mounted memory (25) and is used to output muon detection data to the tunnel muon imaging server (4).

8. The tunneling muon imaging system based on compressed sensing according to claim 7, characterized in that: The tunnel muon imaging server (4) includes a data input interface (41), a server-side memory (42), a computer (43), and a display (44). The server-side memory (42) is communicatively connected to the data input interface (41). When the data input interface (41) is connected to the data output interface (27), the server-side memory (42) collects all muon detection data stored in the vehicle-mounted memory (25). The computer (43) is communicatively connected to the server-side memory (42). The computer (43) outputs a tunnel three-dimensional reconstruction image based on compressed sensing processing and three-dimensional imaging technology. The display (44) is communicatively connected to the computer (43) and displays the tunnel three-dimensional reconstruction image.

9. A tunneling muon imaging method based on compressed sensing, characterized in that, The method of using the tunnel muon imaging system based on compressed sensing as described in claim 8 to perform three-dimensional imaging of a target tunnel includes the following steps: Step 1: Place the track vehicle (22) of the mobile muon detection device (2) on the track at the starting point of the tunnel. The track vehicle (22) moves at a constant speed along the tunnel track. The positioning receiver (23) receives the position signals of multiple locators (15) of multiple fixed muon detector arrays (1) in real time. Step 2: When the railcar controller (26) determines that the distance between the positioning receiver (23) and the locator (15) is less than the set minimum distance threshold, it controls the vehicle-mounted wireless transceiver module (24) to receive the muon detection data sent by the muon detection end wireless transceiver module (14) of the fixed muon detector array (1) corresponding to the locator (15), and transmits it to the vehicle-mounted memory (25) for storage. Step 3: The data analysis chip (32) reads the original data of the corresponding fixed muon detector array (1) from the buffer (31), performs preprocessing, removes electron interference signals in cosmic rays based on energy threshold, performs trajectory integrity verification, matches the incident and outgoing trajectories by timestamp, determines whether there is a break or abnormal deflection, and transmits the preprocessing results to the track vehicle controller (26). Step 4: If the judgment result in step 3 is that there is no breakage or abnormal deflection, then the track vehicle controller (26) records the corresponding fixed muon detector array (1) as qualified and controls the track vehicle (22) to continue to the next fixed muon detector array (1) until all muon detection data is collected. Step 5: The data input interface (41) and data output interface (27) of the tunnel muon imaging server (4) are connected. The server-side memory (42) collects all the muon detection data stored in the vehicle-mounted memory (25). The computer (43) outputs the tunnel three-dimensional reconstruction image based on compressed sensing processing and three-dimensional imaging technology.

10. The tunneling muon imaging method based on compressed sensing according to claim 9, characterized in that, If the result of step 3 indicates the presence of a break or abnormal deflection, then the mobile sampling step is executed: The track vehicle controller (26) uses the position closed-loop control of the positioner (15) and the position receiver (23) to park the track vehicle (22) directly below the corresponding fixed muon detector array (1); The mobile muon detector array is activated to replace the corresponding fixed muon detector array (1) to collect muon data; the supplementary collected data is written into the vehicle memory (25), and at the same time, the data analysis chip (32) repeats step 3 to preprocess the supplementary data to confirm that there is no breakage or abnormal deflection. The track vehicle (22) continues to move and continues to execute steps 4 and 5.

Images