Shield construction stratum loss rate estimation system

Abstract

The invention discloses a shield construction stratum loss rate estimation system, and belongs to the technical field of shield construction stratum loss rate estimation. The shield construction stratum loss rate estimation system comprises a muck conveying belt and is characterized in that a fixed supporting device is arranged on the muck conveying belt, a device box is installed on the fixed supporting device, and an acquisition and analysis system is arranged in the device box and connected with a ground shield digital center through a wireless network. The problems that in the prior art, data collection is not comprehensive, transmission is not timely, the processing and analysis capacity is limited, efficiency is low, and accuracy is difficult to guarantee are solved, a whole-process closed loop of data from collection, processing to display is achieved, data of an underground shield site can be transmitted to a ground shield digital center in time, and the data transmission efficiency is improved. And rapid processing and analysis are carried out, so that the timeliness and accuracy of data processing are guaranteed, and constructors can know the construction condition in time and make corresponding decisions.

Description

Technical Field

[0001] This invention relates to the field of ground loss rate estimation technology in shield tunneling construction, specifically to a ground loss rate estimation system for shield tunneling construction. Background Technology

[0002] During large-scale tunnel boring machine (TBM) construction, the complex geological conditions and unstable construction techniques present significant uncertainties. Therefore, a method for real-time and accurate estimation of the excavated volume during the muck removal process is crucial for providing vital construction information to assist workers in adjusting their decisions and ensuring the safe and stable excavation of the tunnel. This necessitates real-time monitoring of the excavated muck discharged from the conveyor belt.

[0003] Chinese Patent Application No. CN201810267627 discloses a real-time weighing system for excavated soil in shield tunnel engineering and its usage method. The real-time weighing system includes a weighing device, a data acquisition device, a wireless transmission device, and a power supply device installed below the hopper of a dump truck, as well as a data processing and display device installed in the shield machine's operating room. The weighing device weighs the excavated soil; the data acquisition device acquires the weighing signal from the weighing device and transmits the data to the data processing and display device via the wireless transmission device; the data processing and display device processes the data and displays it in real time; the power supply device provides power to the weighing device, data acquisition device, and wireless transmission device. The usage method involves loading the excavated soil into the hopper; triggering the weighing device to automatically weigh the soil; the data acquisition device acquiring the weighing signal from the weighing device and transmitting it to the data processing and display device; and the data processing and display device receiving the weighing signal, converting the received weighing signal into digital data, and displaying it on the display device.

[0004] The aforementioned patent determines the amount of excavated soil by weighing during actual use. However, the excavated soil in the dump truck is not completely compacted, and there are a large number of voids. Furthermore, it is difficult to accurately determine the density of the excavated soil. The data collection is incomplete, the transmission is not timely, and the processing and analysis capabilities are limited, resulting in low efficiency and difficulty in guaranteeing accuracy. Therefore, it does not meet the existing needs. In response, we have proposed a ground loss rate estimation system for shield tunneling construction. Summary of the Invention

[0005] The purpose of this invention is to provide a ground loss rate estimation system for tunnel boring machine (TBM) construction, realizing a closed-loop process from data acquisition and processing to display. It can promptly transmit data from the underground TBM site to the ground-based TBM digital center for rapid processing and analysis, ensuring the timeliness and accuracy of data processing. This allows construction personnel to understand the construction situation in a timely manner and make corresponding decisions. The system stores the volumetric data of excavated soil in a database for subsequent querying, statistics, and analysis, providing strong support for construction management and solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a shield tunneling ground loss rate estimation system, comprising a spoil conveyor belt, characterized in that a fixed support device is provided on the spoil conveyor belt, a device box is installed on the fixed support device, and a data acquisition and analysis system is provided inside the device box. The data acquisition and analysis system is connected to the ground shield tunneling digital center via a wireless network, and the acquisition module is connected to the power supply system. The data acquisition and analysis system consists of an acquisition module, a control module, and a communication module. The acquisition module is used to acquire data, the control module is used to process and control the data, and the communication module is used to transmit the data.

[0007] Preferably, the acquisition module specifically includes:

[0008] A depth camera is used to record video of the construction waste in real time using multimodal data acquisition capabilities, and then images the recorded video onto the image sensor of the depth camera.

[0009] The tunnel boring machine PLC is used to set and debug the acquisition parameters of depth camera 5 via Modbus TCP protocol.

[0010] The local cache SSD is used to encode image data, and the encoded data is then cached locally.

[0011] Preferably, the tunnel boring machine PLC specifically includes:

[0012] Configure the depth camera parameters, and then adjust the depth camera acquisition parameters after configuration.

[0013] Set an appropriate viewing angle range based on the collected parameters to ensure that the camera can completely cover the slag area on the conveyor belt, and calibrate the depth camera according to the viewing angle range.

[0014] Preferably, the control module includes:

[0015] Preprocessing of construction parameters from depth cameras and industrial control computers;

[0016] After processing, control commands are issued based on the construction parameters from the processed depth camera and industrial control computer.

[0017] The depth camera receives control commands, records video, stores the recorded video, and uploads it to the industrial control computer.

[0018] The industrial control computer receives control commands, processes and analyzes the uploaded video, and transmits it to the communication module after processing and analysis.

[0019] Preferably, after the processing is completed, a control command is issued based on the construction parameters of the processed depth camera, specifically including:

[0020] The slag discharge status is determined by multi-parameter fusion decision. If slag discharge occurs, the depth camera status is further determined. If the camera is not working, a recording command is triggered. If the camera is working, recording is maintained.

[0021] If no debris is produced, check if the depth camera is working. If it is working, send a stop command; otherwise, keep it idle.

[0022] Execute recording control, input start / stop commands, output the actual status of the depth camera, and confirm the execution of commands via OPCUA to ensure control synchronization.

[0023] Preferably, the industrial control computer specifically includes:

[0024] The industrial control computer is equipped with formation loss rate estimation software to ensure compatibility between the software and the depth camera driver. The data processing algorithm is preloaded into the industrial control computer and tested. The calculation results are checked for reasonableness through indoor testing of the depth camera.

[0025] After the test is completed, the collected video data of the construction waste is stored in blocks, the input video stream is output MP4 segments with timestamps;

[0026] Real-time analysis of construction waste is performed. The current video frame is input, and the estimated volume of construction waste is output using the integral method. The process is repeated until the depth camera stops working.

[0027] The ground loss rate is calculated using the estimated volume of excavated soil, and the calculation results are sent to the ground shield digital center for storage and in-depth analysis via a stable wired connection and wireless transmission link.

[0028] Preferably, the ground-based shield tunneling digital center includes:

[0029] The cloud is used to receive and store data files transmitted from the industrial control computers on-site, and to perform further analysis.

[0030] The display is used to visualize the results of the soil volume identification and provides a graphical interface for the industrial control computer and depth camera, which facilitates on-site operation and debugging of the equipment.

[0031] Storage devices are used to store video data of the construction waste collected by depth cameras and volume data of the construction waste identified by industrial control computers.

[0032] Preferably, the fixed support device consists of a bottom connecting steel plate, a longitudinal support bar, and a transverse support rod, used to fix the depth camera and the industrial control computer. The bottom connecting steel plate is welded to the side of the slag conveyor belt. The longitudinal support rod is quickly fixed to the bottom connecting steel plate by M8 hexagon socket bolts. The transverse support rod is a transverse telescopic arm to ensure that the center line of the device box is aligned with the center line of the conveyor belt.

[0033] Preferably, the device box has honeycomb-shaped heat dissipation holes on both sides, the inside of the device box is divided into left and right sections, the right section is used to install a depth camera, and the left section is used to install an industrial control computer. The two sections are separated by a galvanized steel plate with a thickness of mm and a spacing of 20mm ± 0.5mm. The bottom of the device box is open, and the top of the device box is provided with two parallel through holes.

[0034] Preferably, the industrial control computer deploys formation loss rate estimation software to ensure compatibility between the software and the depth camera driver. The industrial control computer pre-loads data processing algorithms and performs test runs. Indoor depth camera tests are used to check the reasonableness of the calculation results. Specifically, this includes:

[0035] During the no-spoil-discharge phase before each ring of shield tunneling, three background depth images are acquired. These three background images are then subjected to mean filtering to generate a baseline background image B(u,v), whose expression is:

[0036]

[0037] Among them B i (u,v) represents the depth value of the background image in the i-th frame at pixel coordinates (u,v).

[0038] During the muck removal phase of each ring of shield tunneling operation, the current depth image D(u,v) is acquired and its difference is calculated with the baseline background image B(u,v). A piecewise function is then used to calculate the difference image ΔD(u,v).

[0039]

[0040] The pixel coordinates (u,v) and depth values ​​d(u,v) are converted to 3D coordinates xc, yc, zc in the depth camera 5 coordinate system using the depth camera 5 intrinsic parameter matrix K. The transformation relationship is as follows:

[0041]

[0042] Where s is the depth scaling factor, and d(u,v)=sΔD(u,v) is the actual depth value;

[0043] The expression for the intrinsic parameter matrix K is:

[0044]

[0045] Pixel coordinates to world coordinates conversion:

[0046]

[0047] Mapping between depth values ​​and actual height:

[0048] h(u,v)=Δ(D(u,v)×s)

[0049] Calculate the area S of all connected regions i Filter area is smaller than S min =500 pixels of noise area, retain the largest connected region as the target area R of the slag and soil;

[0050] Hierarchical integral model:

[0051]

[0052] Where Smin is the actual area corresponding to a single pixel, which is determined by the depth camera resolution and the object distance;

[0053] Overall volume integration:

[0054]

[0055] Among them, V total Total volume;

[0056] Correction of waste volume:

[0057] V diff =V total ×k compaction

[0058] Among them, V diff For the correction of the volume of slag and soil, k compaction Let k be the compaction coefficient of the slag and soil. compaction =0.85.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] This invention uses computer vision technology to process depth images, automatically identify the shape of excavated soil, and calculates its volume through intelligent algorithms, thereby improving measurement efficiency and automation. Through depth cameras, industrial control computers, and wireless transmission, it achieves fully automated data collection, processing, storage, and transmission, reducing manual intervention and improving monitoring efficiency. Data can be transmitted to the ground shield tunneling digital center in real time, allowing construction personnel to monitor the excavation status of the shield tunneling at any time. The sampling frequency can be automatically adjusted according to the shield tunneling speed to ensure the continuity and accuracy of data collection. Excavated soil volume data is stored in a database for subsequent querying, statistics, and analysis, providing strong support for construction management.

[0061] This invention requires only the installation of a depth camera and an industrial control computer. It is easy to install, has lower costs, and is highly applicable. It does not require direct contact with the construction waste and has no impact on the construction process. The position of the depth camera can be precisely adjusted through a fixed support device, and the device box can stably adjust the shooting angle and provide dustproof and heat dissipation protection. Both reduce environmental interference, ensure that the depth camera accurately acquires images, and ensure that the monitoring data is accurate and reliable. Attached Figure Description

[0062] Figure 1 This is a structural diagram of the shield tunneling ground loss rate estimation system of the present invention;

[0063] Figure 2 This is a flowchart of the shield tunneling ground loss rate estimation system of the present invention;

[0064] Figure 3 This is a diagram of the shield tunneling ground loss rate estimation system architecture of the present invention.

[0065] In the diagram: 1. Slag conveyor belt; 2. Equipment box; 3. Power supply system; 4. Industrial control computer; 5. Depth camera; 6. Fixed support device. Detailed Implementation

[0066] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] To address the problems of existing technologies that determine the volume of construction waste by weighing, which result in incomplete compaction of the waste in dump trucks (excessive voids), difficulty in accurately determining waste density, incomplete data collection, untimely transmission, and limited processing and analysis capabilities, leading to inefficiency and unreliable accuracy, please refer to [the relevant documentation / reference needed]. Figures 1-3 This embodiment provides the following technical solution:

[0068] A shield tunneling ground loss rate estimation system includes a spoil conveyor belt 1. The system is characterized by a fixed support device 6 mounted on the spoil conveyor belt 1, with a device box 2 installed on the fixed support device 6. The device box 2 contains a data acquisition and analysis system, which is connected to the ground-based shield tunneling digital center via a wireless network. The acquisition module is connected to a power supply system 3, which powers a depth camera 5, an industrial control computer 6, and other field equipment. The data acquisition and analysis system consists of an acquisition module, a control module, and a communication module. The acquisition module acquires data, the control module processes and controls the data, and the communication module transmits the data, sending the acquired data to the control module and transmitting the processed data to a remote module. This provides network services to the depth camera 5 and the industrial control computer 6, enabling remote debugging of field equipment and transmission of result data. These layers collaborate to form a complete system, achieving comprehensive monitoring and management of the shield tunneling machine's construction process.

[0069] By employing multi-process isolation technology, each video processing task runs independently, improving system stability, resource utilization, and data processing efficiency. This prevents a single task failure from affecting the entire system and facilitates system maintenance and expansion. The use of chunked upload and breakpoint resume technologies, combined with CRC-32 checksums, reduces the impact of network fluctuations on data transmission, ensuring complete and accurate transmission of image data to the ground layer and guaranteeing the reliability of the transmission strategy. Real-time data transmission enables the monitoring platform to obtain video status information promptly, providing real-time decision support for management personnel and achieving precise monitoring and management of the tunnel boring machine construction process. The use of open standard protocols such as Modbus TCP allows for compatibility with different manufacturers. The equipment is interoperable, facilitating system integration and expansion to meet future business development needs. Multi-process isolation and data verification mechanisms ensure data security and integrity, preventing data leakage and malicious attacks, and providing reliable data support for tunnel boring machine construction. The depth camera 5 captures video data of the excavated soil from the conveyor belt in real time. After data processing and depth analysis, the processing results are transmitted wirelessly to the ground tunnel boring machine digital center in real time, realizing the visualization of ground loss rate data. It can realize the real-time shooting, collection, transmission and processing of excavated soil data from the conveyor belt, effectively solving the problem that it is difficult to accurately estimate the amount of soil excavation during tunnel boring machine excavation on the engineering site and that it is impossible to accurately obtain ground loss information.

[0070] The data acquisition module specifically includes:

[0071] Depth camera 5 is used to record videos of the excavated soil in real time using multimodal data acquisition capabilities, providing accurate raw data support for the calculation of formation loss rate, and the recorded video is imaged onto the image sensor of depth camera 5.

[0072] The tunnel boring machine's PLC is used to set and debug the acquisition parameters of depth camera 5 via the Modbus TCP protocol. This includes adjusting the acquisition parameters, setting an appropriate viewing angle range, and ensuring the camera can completely cover the excavated soil area on the conveyor belt. Calibration is also performed to reduce distortion and improve measurement accuracy.

[0073] The local cache SSD is used to encode image data. After encoding, the data is cached locally to ensure data integrity when transmission is unstable.

[0074] The tunnel boring machine PLC specifically includes:

[0075] Configure the parameters of depth camera 5, and then adjust the acquisition parameters of depth camera 5 after configuration.

[0076] Set an appropriate viewing angle range based on the collected parameters to ensure that the camera can completely cover the slag area on the conveyor belt. Then, calibrate and set the depth camera 5 according to the viewing angle range.

[0077] The control module includes:

[0078] The construction parameters of depth camera 5 and industrial control computer 6 are preprocessed;

[0079] After processing, control commands are issued based on the construction parameters of the processed depth camera 5 and industrial control computer 6.

[0080] The depth camera 5 receives control commands, records video, stores the recorded video, and uploads it to the industrial control computer 6.

[0081] The industrial control computer 6 receives control commands, processes and analyzes the uploaded video, and transmits it to the communication module and pushes metadata after processing and analysis are completed.

[0082] After processing, control commands are issued based on the processed construction parameters of depth camera 5, specifically including:

[0083] The slag discharge status is determined by multi-parameter fusion decision. If slag discharge occurs, the status of depth camera 5 is further determined. If it is not working, a recording command is triggered and needs to be confirmed for 3 consecutive cycles. If it is working, recording is maintained.

[0084] If no slag is produced, determine whether depth camera 5 is working. If it is working, send a stop command and confirm for 5 consecutive cycles; otherwise, keep it idle.

[0085] Execute recording control, input start / stop commands, output the actual status of depth camera 5, and confirm the execution of commands via OPCUA to ensure control synchronization.

[0086] Industrial control computer 6, specifically including:

[0087] The industrial control computer 6 is equipped with formation loss rate estimation software to ensure software compatibility with depth camera 5 driver. The industrial control computer 6 is preloaded with data processing algorithms and tested. The calculation results are checked for reasonableness through indoor testing of depth camera 5.

[0088] After the test is completed, the collected video data of the construction waste is stored in blocks. The input video stream is saved every 30 seconds or when it stops, and the output MP4 segments with timestamps are saved.

[0089] Real-time analysis of construction waste is performed. The current video frame is input, and the estimated volume of construction waste is output using the integral method. The process is repeated until the depth camera 5 stops working.

[0090] The ground loss rate is calculated using the estimated volume of excavated soil, and the calculation results are sent to the ground shield digital center for storage and in-depth analysis via a stable wired connection and wireless transmission link.

[0091] Remote module: The data is stored in the cloud at the digital center, and the monitoring platform visualizes the data, making it easy for managers to keep track of the construction progress in real time.

[0092] The ground-based shield tunneling digital center includes:

[0093] The cloud is used to receive and store data files transmitted from the field industrial control computer 6, and to perform further analysis;

[0094] The display is used to visualize the results of the soil volume identification and provides a graphical interface for the industrial control computer 6 and the depth camera 5, which facilitates on-site operation and debugging of the equipment.

[0095] Storage device for storing video data of the excavated soil collected by depth camera 5 and volume data of the excavated soil identified by industrial control computer 6.

[0096] The fixed support device 6 consists of a bottom connecting steel plate, longitudinal support bars, and transverse support rods. It is used to fix the depth camera 5 and the industrial control computer 6. The longitudinal support rod adopts a two-section telescopic design with a height adjustment range of 0.6–1.2 m. Each section is fixed by bolts and can be manually adjusted by sliding after unlocking, with an accuracy of ±1 cm. The transverse support rod also adopts a two-section telescopic design with a transverse extension length adjustment range of 0.4–0.8 m. Each section is fixed by bolts and can be manually adjusted by sliding after unlocking, with an accuracy of ±1 cm. The bottom connecting steel plate, approximately 1 cm thick, has pre-drilled bolt holes and is fixed to the support rod base using M8 hex bolts. This design allows for easy installation on-site by simply welding the bottom connecting steel plate to the installation area, and easy disassembly by simply loosening the bolts, separating the support rod from the steel plate without damage (disassembly time <30 seconds). Furthermore, considering the impact of vibration on the stability of the device during tunnel boring machine (TBM) construction, polyurethane damping pads are used at the bolt connections, combined with rubber damping rings, to reduce the transmission rate of construction vibrations.

[0097] The device box 2 has honeycomb-shaped heat dissipation holes with a diameter of 1cm and an opening rate of 30% on both sides, which can dissipate the heat generated by the industrial control computer 6 and depth camera 5 during operation. The heat is also conducted out in conjunction with the heat-conducting plate structure of the industrial control computer 6 shell, and the operating temperature is <60℃. A detachable dust cover made of PET material with a light transmittance of >90% is installed at the lens opening of the depth camera 5 to prevent soil splashing or construction dust from contaminating the lens. The device box 2 adopts a partitioned layout structure, with the inside of the device box 2 divided into left and right sections. The depth camera 5 is installed in the right section and the industrial control computer 6 is installed in the left section. The two sections are separated by a 1mm thick galvanized steel plate with a spacing of 20mm±0.5mm. The bottom opening of the device box 2 and the galvanized steel plate also serve to fix the depth camera 5, which is convenient for capturing images of soil. The top of the device box 2 has two parallel through holes. By tightening or loosening the bolts at both ends, the pitch angle can be adjusted within a range of ±15° by changing the angle between the device box 2 and the fixed support device 6. Angle scale lines with an accuracy of ±1° are engraved on the side of the device box 2. Combined with the pointer marks on the fixed support device 6, this allows for visual adjustment of the pitch angle. Simultaneously, considering the impact of vibration during tunnel boring machine (TBM) construction on the stability of the device box, shims are added between the bolts and the box body to prevent bolt loosening due to vibration and ensure angle locking stability. The quick adjustment and locking process is as follows: Loosen the bolts until the device box can rotate freely; when adjusting the pitch angle, push the device box to the target angle according to the scale indication; then tighten the bolts on both sides clockwise, using the shims to press the box body into place, thus fixing the angle.

[0098] The bottom connecting steel plate is a Q235 carbon steel plate with pre-threaded construction. It is welded to the side of the slag conveyor belt 1, with a welding strength ≥200MPa. The surface is coated with an epoxy anti-rust coating. The longitudinal support rod is a telescopic support rod, and the transverse support rod is an adjustable support rod. The longitudinal support rod is quickly fixed to the bottom connecting steel plate using M8 hex bolts. The bolt heads are embedded in the grooves of the bottom connecting steel plate to prevent damage during construction. The transverse support rod has a transverse telescopic arm with a rack and pinion structure, adjustable from 0.4 to 0.8m, ensuring that the alignment deviation between the centerline of the device box and the centerline of the conveyor belt is ≤1cm.

[0099] During installation, loosen the wing nuts on both sides of the rod hole at the top of the device box 2, push the box to the target angle scale with an accuracy of ±1°, tighten the nuts, and implement environmental protection measures. The device box 2 has an IP54 protection rating. A removable PET dust cover with a light transmittance of >90% is installed at the lens opening. The box has honeycomb-shaped heat dissipation holes with a diameter of 3mm on both sides, and a built-in silent fan with a wind speed of 0.8m / s to ensure that the working temperature is ≤50℃.

[0100] The equipment wiring and power supply system connects the depth camera 5 and the industrial control computer 6 via a wired connection, ensuring stability and reliability while avoiding signal transmission interference. Wired transmission speed tests are conducted to ensure data can be successfully uploaded to the ground-based shield tunneling digital center. Environmental adaptability testing involves trial operation to monitor the equipment's stability under complex environments such as shield tunneling vibration, dust, and moisture. The angle, focal length, and exposure time of the depth camera 5 are adjusted appropriately to optimize image quality.

[0101] Industrial control computer 6 is equipped with formation loss rate estimation software to ensure compatibility with depth camera 5 driver. Data processing algorithms are pre-loaded onto industrial control computer 6 and tested. Indoor testing with depth camera 5 is used to verify the reasonableness of the calculation results, specifically including:

[0102] During the no-spoil-discharge phase before each ring of shield tunneling, three background depth images are acquired. These three background images are then subjected to mean filtering to generate a baseline background image B(u,v), whose expression is:

[0103]

[0104] Among them B i (u,v) represents the depth value of the background image in the i-th frame at pixel coordinates (u,v).

[0105] During the muck removal phase of each ring of shield tunneling operation, the current depth image D(u,v) is acquired and its difference is calculated with the baseline background image B(u,v). A piecewise function is then used to calculate the difference image ΔD(u,v).

[0106]

[0107] This process retains only the pixels with increased depth and eliminates the negative difference noise caused by the jitter of the depth camera (5);

[0108] The transformation from pixel coordinates to depth camera (5) coordinates: In the image coordinate system, u and v in M(u,v) represent the column coordinates and row coordinates of the pixel (with the upper left corner of the image as the origin).

[0109] The transformation relationship between pixel coordinates (u,v) and depth value d(u,v) and the three-dimensional coordinates (xc, yc, zc) in the depth camera (5) coordinate system is as follows, using the intrinsic parameter matrix K of the depth camera (5):

[0110]

[0111] Where s is the depth scaling factor, in mm / pixel, obtained from the calibration of the depth camera 5, and d(u,v)=sΔD(u,v) is the actual depth value;

[0112] The expression for the intrinsic parameter matrix K is:

[0113]

[0114] Pixel coordinates to world coordinates conversion:

[0115]

[0116] Mapping between depth values ​​and actual height:

[0117] h(u,v)=Δ(D(u,v)×s)

[0118] Calculate the area S of all connected regions i Filter area is smaller than S min =500 pixels of noise area, retain the largest connected region as the target area R of the slag and soil;

[0119] Hierarchical integral model:

[0120]

[0121] Among them, S min The actual area corresponding to a single pixel is determined by the resolution of the depth camera and the object distance.

[0122] Overall volume integration:

[0123]

[0124] Among them, V total Total volume;

[0125] Correction of waste volume:

[0126] V diff =V total ×k compaction

[0127] Among them, V diff For the correction of the volume of slag and soil, k compaction Let k be the compaction coefficient of the slag and soil. compaction =0.85. Based on the loose characteristics of the excavated soil, the total volume is corrected by the compaction coefficient to obtain the actual reference volume of the project.

[0128] By combining multi-parameter fusion decision-making with slag discharge status judgment, recording is ensured to start / stop at the appropriate time, improving the correlation between video and construction parameters. Through handling of equipment connection anomalies, recording command constraints (continuous periodic confirmation), and breakpoint resume protocol, system stability is enhanced and the risk of data loss is reduced. Parameter preprocessing optimizes data quality, real-time slag analysis provides immediate construction information, block storage and asynchronous uploading improve data processing efficiency, OPCUA confirms command execution, is compatible with industrial communication standards, and the process design is flexible to adapt to different construction scenarios.

[0129] Computer vision technology is used to process depth images, automatically identifying the shape of excavated soil and calculating its volume through intelligent algorithms, improving measurement efficiency and automation. Through a depth camera 5, an industrial control computer 6, and wireless transmission, fully automated data acquisition, processing, storage, and transmission are achieved, reducing manual intervention and improving monitoring efficiency. Data can be transmitted in real time to the ground-based shield tunneling digital center, allowing construction personnel to monitor the excavation status of the shield tunneling at any time. Only the depth camera 5 and the industrial control computer 6 need to be installed; installation is simple, cost-effective, and highly applicable. The sampling frequency can be automatically adjusted according to the shield tunneling speed to ensure the continuity and accuracy of data acquisition. The fixed support device 6 can precisely... The device box 2 is designed to precisely adjust the position of the depth camera 5 and provides dust and heat protection, minimizing environmental interference and ensuring accurate image acquisition by the depth camera 5. This guarantees accurate and reliable monitoring data. Through automated acquisition codes and network transmission, the entire process of data acquisition, transmission, and processing is automated, reducing manual intervention and improving monitoring efficiency. The proposed method for estimating slag volume requires only one depth camera 5 and an industrial control computer 6 for data processing and calculation at the construction site. Compared to other volume measurement devices or laser scanners, this method is simpler to install, lower in cost, and does not require direct contact with the slag, thus having no impact on the construction process. The reasoning speed of the calculation algorithm itself is sufficient to keep up with the image acquisition speed of the depth camera 5. Therefore, this method is real-time and low-cost, enabling timely transmission of data from the underground shield tunneling site to the ground-based shield tunneling digital center for rapid processing and analysis. This allows construction personnel to understand the construction situation promptly, make appropriate decisions, and store slag volume data in a database for subsequent querying, statistics, and analysis, providing strong support for construction management.

[0130] Working principle: When using the shield tunneling ground loss rate estimation system of this invention, based on... Figure 1 , Figure 2 and Figure 3 Specifically, it includes:

[0131] Device connection and parameter reading include:

[0132] Initialize device connection; no input or output PLC and depth camera 5 connection status; if abnormal, retry 3 times and then alarm to ensure connection reliability.

[0133] Read real-time parameters from the PLC, such as the screw conveyor speed and propulsion speed, and input the device connection status.

[0134] The parameters are preprocessed, specifically including:

[0135] The original parameters are filtered by a moving average filter window of 5 and standardized by unit for torque (N·m) and current (A), and the filtered parameter vector is output to improve data quality.

[0136] Recording control decisions specifically include:

[0137] The system determines the slag discharge status through multi-parameter fusion decision-making. If slag discharge occurs, the system further determines the status of depth camera 5. If it is not working, a recording command is triggered, which needs to be confirmed for 3 consecutive cycles. If it is working, the recording is maintained. If no slag discharge occurs, the system determines whether depth camera 5 is working. If it is working, a stop command is sent, which needs to be confirmed for 5 consecutive cycles. Otherwise, it remains idle.

[0138] Execute recording control, input start / stop commands, output the actual status of depth camera 5, and confirm the execution of commands via OPCUA to ensure control synchronization.

[0139] Video processing and analysis include:

[0140] The video is segmented and stored. The input video stream is saved as timestampd MP4 segments, saved every 30 seconds or at the stop. Real-time waste soil analysis is performed: the current video frame is input, and an estimated waste soil volume is output using an integral method. This process is repeated until depth camera 5 stops working.

[0141] Uploading data specifically includes:

[0142] Data is uploaded asynchronously. Input video segments and analysis results, and output the cloud storage path. The HTTP / 2+ interruptible resume protocol is used to ensure the integrity of data transmission.

[0143] In summary, this invention ensures the stable operation of all components through a power supply device and network system. It utilizes a depth camera 5 to capture real-time video data of the excavated soil from the conveyor belt. After data processing and depth analysis, the results are wirelessly transmitted in real-time to the ground-based shield tunneling digital center for display, achieving a visualized presentation of the ground loss rate data. The depth camera 5 and the industrial control computer 6, as core components of the system, work closely together to accurately calculate the ground loss rate. The depth camera 5 possesses multimodal data acquisition capabilities, simultaneously acquiring RGB, infrared, and depth images, providing accurate raw data support for ground loss rate calculation. The industrial control computer 6 is pre-loaded with algorithms to process the acquired image data in real-time, calculating the ground loss rate... The system measures the layer loss rate and transmits the analysis results to the ground shield digital center via stable wired and wireless connections for storage and depth analysis. All devices in the system work collaboratively to achieve a closed-loop data processing and display process. The fixed support device 6 ensures the stable operation of the depth camera 5 and the industrial control computer 6 in complex construction environments. The data processing and storage equipment works efficiently with the depth camera 5 to ensure the timeliness and accuracy of data processing. This system can continuously and stably monitor the shield tunneling process without interfering with normal construction, providing crucial data support for engineers to accurately assess the layer loss status and predict construction risks, significantly improving the intelligent management level and safety control capabilities of shield construction.

[0144] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0145] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A ground loss rate estimation system for shield tunneling construction, including a spoil conveyor belt (1), characterized in that, The slag conveyor belt (1) is equipped with a fixed support device (6), and a device box (2) is installed on the fixed support device (6). The device box (2) is equipped with a data acquisition and analysis system. The data acquisition and analysis system is connected to the ground shield digital center through a wireless network. The data acquisition module is connected to the power supply system (3). The data acquisition and analysis system consists of a data acquisition module, a control module and a communication module. The data acquisition module is used to acquire data, the control module is used to process and control data, and the communication module is used to transmit data.

2. The shield tunneling ground loss rate estimation system according to claim 1, characterized in that, The acquisition module specifically includes: The depth camera (5) is used to record the video of the slag in real time using the multimodal data acquisition capability, and the recorded video is imaged on the image sensor of the depth camera (5). The tunnel boring machine PLC is used to set and debug the acquisition parameters of the depth camera (5) via the Modbus TCP protocol. The local cache SSD is used to encode image data, and the encoded data is then cached locally.

3. The shield tunneling ground loss rate estimation system according to claim 2, characterized in that, The tunnel boring machine PLC specifically includes: Set the parameters of the depth camera (5), and adjust the acquisition parameters of the depth camera (5) after setting. Set an appropriate viewing angle range according to the acquisition parameters to ensure that the camera can completely cover the slag area on the conveyor belt, and calibrate the depth camera (5) according to the viewing angle range.

4. The shield tunneling ground loss rate estimation system according to claim 1, characterized in that, The control module includes: The construction parameters of the depth camera (5) and the industrial control computer (6) are preprocessed; After processing, control commands are issued based on the construction parameters of the processed depth camera (5) and industrial control computer (6); The depth camera (5) receives control commands to record video, stores the recorded video, and uploads it to the industrial control computer (6); The industrial control computer (6) receives control commands, processes and analyzes the uploaded video, and transmits it to the communication module after processing and analysis.

5. The shield tunneling ground loss rate estimation system according to claim 4, characterized in that, After the processing is completed, control commands are issued based on the construction parameters of the processed depth camera (5), specifically including: The slag discharge status is determined by multi-parameter fusion decision. If slag discharge occurs, the status of the depth camera (5) is further determined. If it is not working, the recording command is triggered. If it is working, the recording is maintained. If no slag is discharged, determine whether the depth camera (5) is working. If it is working, send a stop command; otherwise, keep it idle. Execute recording control, input start / stop commands, output the actual status of the depth camera (5), and execute the confirmation command through OPCUA to ensure control synchronization.

6. The shield tunneling ground loss rate estimation system according to claim 4, characterized in that, The industrial control computer (6) specifically includes: The industrial control computer (6) is equipped with formation loss rate estimation software to ensure compatibility between the software and the depth camera (5). The industrial control computer (6) is preloaded with data processing algorithms and tested. The calculation results are checked for reasonableness through indoor testing of the depth camera (5). After the test is completed, the collected video data of the construction waste is stored in blocks, the input video stream is output MP4 segments with timestamps; Real-time analysis of slag and soil is performed. The current video frame is input, and the volume estimate of slag and soil is output using the integral method. The process is repeated until the depth camera (5) stops working. The ground loss rate is calculated using the estimated volume of excavated soil, and the calculation results are sent to the ground shield digital center for storage and in-depth analysis via a stable wired connection and wireless transmission link.

7. The shield tunneling ground loss rate estimation system according to claim 1, characterized in that, The ground-based shield tunneling digital center includes: The cloud is used to receive and store data files transmitted by the on-site industrial control computer (6) and perform further analysis; The display is used to visualize the results of the soil volume identification and to provide a graphical page for the industrial control computer (6) and the depth camera (5) to facilitate on-site operation and debugging of the equipment. Storage device for storing video data of the slag collected by the depth camera (5) and volume data of the slag identified by the industrial control computer (6).

8. The shield tunneling ground loss rate estimation system according to claim 1, characterized in that, The fixed support device (6) consists of a bottom connecting steel plate, a longitudinal support bar and a transverse support rod, used to fix the depth camera (5) and the industrial control computer (6). The bottom connecting steel plate is welded to the side of the slag conveyor belt (1). The longitudinal support rod is quickly fixed to the bottom connecting steel plate by M8 hexagonal bolts. The transverse support rod has a transverse telescopic arm to ensure that the center line of the device box is aligned with the center line of the conveyor belt.

9. The shield tunneling ground loss rate estimation system according to claim 8, characterized in that, The device box (2) has honeycomb-shaped heat dissipation holes on both sides. The inside of the device box (2) is divided into left and right areas. The right area is equipped with a depth camera (5) and the left area is equipped with an industrial control computer (6). The two layers are separated by a 1mm thick galvanized steel plate with a spacing of 20mm±0.5mm. The bottom of the device box (2) is open, and the top of the device box (2) is provided with two parallel through holes.

10. The shield tunneling ground loss rate estimation system according to claim 9, characterized in that, The industrial control computer (6) is equipped with formation loss rate estimation software to ensure compatibility between the software and the depth camera (5) driver. The industrial control computer (6) is preloaded with data processing algorithms and tested. The calculation results are checked for reasonableness through indoor testing of the depth camera (5), specifically including: During the no-spoil-discharge phase before each ring of shield tunneling, three background depth images are acquired. These three background images are then subjected to mean filtering to generate a baseline background image B(u,v), whose expression is: Among them B i (u,v) represents the depth value of the background image in the i-th frame at pixel coordinates (u,v). During the muck removal phase of each ring of shield tunneling operation, the current depth image D(u,v) is acquired and its difference is calculated with the baseline background image B(u,v). A piecewise function is then used to calculate the difference image ΔD(u,v). The transformation relationship between pixel coordinates (u,v) and depth value d(u,v) and the three-dimensional coordinates (xc, yc, zc) in the depth camera (5) coordinate system is as follows, using the intrinsic parameter matrix K of the depth camera (5): Where s is the depth scaling factor, and d(u,v)=sΔD(u,v) is the actual depth value; The expression for the intrinsic parameter matrix K is: Pixel coordinates to world coordinates conversion: Mapping between depth values ​​and actual height: h(u,v)=Δ(D(u,v)×s) Calculate the area S of all connected regions i Filter area is smaller than S min =500 pixels of noise area, retain the largest connected region as the target area R of the slag and soil; Hierarchical integral model: Wherein, Smin is the actual area corresponding to a single pixel, which is determined by the resolution of the depth camera (5) and the object distance; Overall volume integration: Among them, V total Total volume; Correction for the volume of construction waste: V diff =V total ×k compaction Among them, V diff For the correction of the volume of slag and soil, k compaction Let k be the compaction coefficient of the slag and soil. compaction =0.85.

Images