Cutter head blockage prevention method suitable for large-dip-angle mountain-down TBM construction

Abstract

The invention discloses a cutterhead blocking prevention method suitable for large-dip-angle downhill TBM construction, and belongs to the technical field of tunneling machine construction.Grooves are evenly milled in the surface of a cutterhead, high-precision temperature sensors are embedded, a distributed temperature monitoring network is formed, a traditional nozzle is replaced with a dual-mode high-abrasion-resistant nozzle, the spraying angle is optimized, the cutterhead face is perpendicular to + / -15 degrees, and the cutterhead face is vertical to + / -15 degrees. The method comprises the following steps: adding a proportion self-adaptive mixer into an original foam generation device, introducing geological preset parameters into an intelligent regulation and control system of a TBM (Tunnel Boring Machine) through dual-mode injection including conventional injection and pulse injection, configuring a temperature threshold model, training an LSTM (Long Short Term Memory) neural network model based on historical construction data, and predicting the blockage risk of a cutterhead. Full-coverage monitoring of the temperature of the cutterhead is achieved through the high-density sensor network, the water spraying angle is optimized in combination with the cutterhead structure and the muck flowing direction, the water spraying angle is adjusted for large-dip-angle downhill construction, water flow can impact in the muck gliding direction, and the probability of blockage of the cutterhead is reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel boring machine construction, and particularly relates to a method for preventing cutterhead blockage in high-angle downhill TBM construction. Background Art

[0002] During steep-angle downhill TBM construction, complex geological conditions, significant slopes, and other factors interact to repeatedly cause cutterhead blockage. This situation severely reduces construction efficiency and significantly slows the overall progress of the project. Conventional methods for preventing blockage are currently inadequate and ineffective in addressing the unique conditions of steep-angle downhill construction. Furthermore, fixed-ratio foaming agents and water spraying strategies are difficult to adapt to the dynamic construction environment. Summary of the Invention

[0003] The object of the present invention is to provide a method for preventing cutterhead clogging in high-angle downhill TBM construction, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned object, the present invention provides the following technical solution: a method for preventing cutterhead blockage in high-angle downhill TBM construction, the method for preventing cutterhead blockage is as follows:

[0005] S1. Mill grooves uniformly on the surface of the cutter head and embed high-precision temperature sensors to form a distributed temperature monitoring network;

[0006] S2. Replace the traditional nozzle with a dual-mode high-wear-resistant nozzle, optimize the spray angle, and add a proportional adaptive mixer to the original foam generating device. The dual-mode spray is conventional spray and pulse spray.

[0007] S3: Importing geological preset parameters into the TBM's intelligent control system, configuring a temperature threshold model, and training an LSTM neural network model based on historical construction data to predict cutterhead blockage risks.

[0008] S4, a high-precision temperature sensor collects temperature data every 0.1 seconds and transmits it to the edge computing device via the CAN bus or LoRa wireless;

[0009] S5. The edge computing device can process and analyze the collected temperature data according to the data processing algorithm;

[0010] S6. Then, according to the analysis results and construction conditions, the strategy for adding foaming agent in the water spraying system and the parameters of the water spraying system are formulated.

[0011] During the implementation process, grooves are evenly milled on the surface of the TBM cutterhead according to the designed spacing and position. High-precision temperature sensors are accurately embedded in the grooves and fixed and protected to ensure that the temperature sensors can work stably and are not affected by vibration and impact during cutterhead excavation. The data cable of the temperature sensor is connected so that it can communicate normally with subsequent data acquisition and transmission equipment. The original traditional nozzle is removed and a dual-mode high-wear-resistant nozzle is installed. At the same time, a proportional adaptive mixer is installed in the foam generating device, and an electromagnetic flowmeter, a servo motor-driven proportional valve, and related control lines are connected to ensure that the proportional adaptive mixer can work properly and accurately adjust the input ratio of the raw liquid, water, and air according to the construction conditions. Then, the geological preset parameters are accurately imported into the TBM's intelligent control system, and an appropriate temperature threshold model is configured according to the geological conditions. Historical construction data is collected and organized, and an LSTM neural network model is trained using professional machine learning software or a programming environment. The trained model is integrated into the intelligent control system to enable real-time prediction of cutterhead blockage risks. When the TBM is started for construction, the high-precision temperature sensor begins to collect cutterhead temperature data at a frequency of once every 0.1 seconds. Depending on the actual situation at the construction site, the CAN bus or LoRa wireless transmission method is selected to transmit the collected temperature data to the edge computing device. During the transmission process, the accuracy and stability of the data must be ensured to avoid data loss or transmission errors. After receiving the temperature data, the edge computing device processes and analyzes it according to the preset data processing algorithm. When it detects that the temperature of a certain area of the cutterhead exceeds the set threshold, the intelligent control system automatically increases the foam injection volume and the flow rate of each flushing pipeline in the corresponding area, adjusts the foam mixing ratio by controlling the proportional adaptive mixer, and starts the pulse injection mode to prevent cutterhead blockage from multiple angles.

[0012] In a specific embodiment, a dredging component is provided inside the nozzle, and the dredging component is driven to move by the size of the water flow in different modes.

[0013] In the above implementation process, during the TBM excavation process, when the nozzle is working, the water flow in the conventional jet mode and the pulse jet mode is different, and the water flow in the pulse jet mode can push the dredging component to move, so that it can clean the water spray hole and reduce the probability of water spray hole blockage.

[0014] In a specific implementation scheme, the proportional adaptive mixer integrates an electromagnetic flowmeter and a proportional valve driven by a servo motor, and has a built-in multi-mode formula library. According to different construction conditions, the input ratio of the raw liquid, water and air can be switched with one click to adapt to different working conditions.

[0015] During the above implementation process, the optimal amount of foaming agent added is automatically or manually selected according to different geological conditions such as soft rock, hard rock, clay and construction conditions such as slope, temperature, and humidity to achieve accurate prevention of cutterhead blockage.

[0016] In a specific embodiment, the dredging assembly includes a mesh plate and a dredging rod. The mesh plate is placed inside the nozzle. The outer wall of one side of the mesh plate is installed with evenly arranged dredging rods, and one end of the dredging rod extends into the water spray hole of the nozzle. The diameter of the dredging rod is smaller than the diameter of the water spray hole.

[0017] During this process, if the nozzle becomes clogged, the larger water flow will push the screen, causing the dredging rod to move within the nozzle, clearing the blockage and ensuring normal nozzle spraying. Regularly check the working condition of the nozzle and dredging components and replace them if damaged.

[0018] In a specific embodiment, S3 further includes: when the temperature of a region exceeds a set threshold, increasing the foam injection amount of the corresponding region, dynamically adjusting the foam mixing ratio, and starting a pulse injection mode.

[0019] In the above implementation process, when the regional temperature exceeds the set threshold, it indicates that the friction between the slag and the cutter disc in the area has intensified, or the accumulation of slag has led to poor heat dissipation, and there is a high risk of blockage. Increasing the foam spraying volume in the corresponding area can increase the coverage of the foam in the area. More foam can better wrap the slag particles, reduce the friction between the particles, make the slag easier to be stirred and discharged by the cutter disc, and reduce accumulation. Under different working conditions, the slag characteristics and the working state of the cutter disc are different. The appropriate foam mixing ratio can optimize the foam performance. When the temperature is abnormal, increasing the stock liquid ratio can improve the foam stability and lubricity, enhance the improvement effect on the slag, adapt to the special conditions of high temperature areas, and better prevent the slag from sticking. The pulsed spray mode sprays foam in an intermittent, high-energy manner, which can produce a stronger impact on the accumulated slag, effectively loosen and break up the slag clumps, prevent them from further accumulation and blockage, and avoid problems such as local water accumulation that may be caused by continuous spraying.

[0020] In a specific implementation scheme, in S3, when the LSTM neural network model predicts that there is a risk of cutterhead blockage, the TBM control system will monitor the flow and boundary pressure data of each flushing pipeline in the mud pipeline in real time, and obtain the flow and node pressure distribution of each pipeline through simulation calculation, and dynamically adjust the flow distribution.

[0021] In the implementation described above, the LSTM neural network model, trained on a large amount of historical construction data, establishes a complex mapping relationship between cutterhead blockage risk and various factors. When the model detects that the current input data meets the blockage risk characteristics, it predicts the risk of cutterhead blockage. Upon receiving the blockage risk warning from the LSTM neural network model, the TBM control system rapidly initiates real-time monitoring of the flow and boundary pressure data of each flushing pipeline in the mud pipeline. Based on the monitoring results, the control system dynamically adjusts flow distribution by adjusting relevant valve openings and pump output power. Leveraging the predictive capabilities of the LSTM neural network model, the risk of cutterhead blockage can be detected before it actually occurs, buying time to implement measures, preventing the blockage from worsening, reducing equipment downtime and cleanup time, and improving construction efficiency. Through real-time monitoring and simulation calculations, precise dynamic adjustment of mud pipeline flow distribution is achieved, optimized according to actual working conditions and pipeline system status, improving the effectiveness of mud flushing the cutterhead, and specifically addressing potential blockage risks, ensuring the stable operation of the TBM system.

[0022] In a specific implementation scheme, in S5, the edge computing device is also used to collect various operating parameters of the TBM.

[0023] In the above implementation process, by collecting various operating parameters of the TBM, it is convenient to analyze the influence of parameters such as foaming agent addition ratio, temperature change, foam injection volume, mud pipeline flow rate and boundary pressure on the blockage of the shield machine cutter head.

[0024] In a specific embodiment, in S5, data processing and analysis include data storage, data analysis, data comparison, and data evaluation;

[0025] Data storage: including local storage and cloud storage, which can store the collected data for a long time and use it as training data for the LSTM neural network model;

[0026] Data analysis: including regular analysis of the collected data and records of foam additive addition;

[0027] Data comparison: compare the collected data with historical data;

[0028] Data evaluation: Evaluate the effectiveness of prevention strategies and identify potential problems and areas for improvement.

[0029] In the above implementation process, in terms of data storage, data is stored in local storage devices and cloud storage platforms at the same time, a complete data backup mechanism is established, data analysis is performed on the collected data and the addition records of foam additives on a regular basis, and data analysis tools are used to explore the potential patterns and associations in the data. The collected data is compared with historical data at certain time intervals to promptly detect data anomalies. By comparing the data before and after the implementation of the prevention strategy, the effectiveness of the prevention strategy is evaluated, and existing problems and improvement directions are recorded.

[0030] In a specific embodiment, in S6, the foaming agent addition strategy includes using Python to regularly analyze temperature trends and foaming agent effects and generate an optimization report.

[0031] In the above implementation process, the powerful data processing and analysis capabilities of the Python language are used to regularly analyze the collected temperature data, draw temperature trend curves, and observe how the temperature changes with time, construction progress, and other factors. At the same time, combined with the foaming agent addition records such as the amount, time, and location of addition, as well as the cutterhead working status feedback such as whether blockage occurs and the effect of soil improvement, the actual effect of the foaming agent is comprehensively evaluated. For example, regression analysis, correlation analysis, and other methods are used to find the potential relationship between temperature changes and the effect of the foaming agent. An optimization report is generated based on the analysis results, summarizing the advantages and disadvantages of the current foaming agent addition strategy, and proposing targeted improvement suggestions to improve the efficiency and effect of the foaming agent, better prevent cutterhead blockage, and ensure the smooth progress of TBM construction.

[0032] In a specific embodiment, the parameter adjustment of the water spraying system includes water spraying pressure, water spraying volume and water spraying time, and the water spraying system is equipped with an intelligent control switch.

[0033] In the above implementation process, appropriate water spray pressure can provide the water flow with sufficient energy to effectively flush the cutterhead surface and the debris. Higher pressure can break up and remove debris clumps adhering to the cutterhead, reducing the risk of blockage. Lower pressure can be used to moisten and soften the debris, helping the foaming agent to better perform its ameliorative role. Adjustment is based on the hardness of the debris and the extent of cutterhead blockage. When the debris is hard or the cutterhead is severely blocked, the pressure needs to be increased to enhance the flushing force. When the debris is soft or the risk of cutterhead blockage is low, the pressure can be reduced to avoid excessive flushing and damage to the cutterhead. The amount of water sprayed affects the moisture and fluidity of the debris. An appropriate amount of water spray can maintain the debris at an appropriate moisture level, facilitate discharge, and prevent it from drying out or becoming too sticky. Adjustment is based on the initial moisture level of the debris and the temperature of the cutterhead. When the debris is dry, the water spray volume is increased to reduce the cutterhead temperature and improve the debris fluidity. When the debris is wet, the water spray volume is reduced to prevent the debris from clumping due to excessive moisture. The water spraying time determines the duration and frequency of water spraying, affecting the effect of soil improvement and cutterhead cleaning. During the actual construction process, the construction progress and the risk of cutterhead blockage can be adjusted to extend the single water spraying time or shorten the water spraying interval.

[0034] The intelligent control switch is precisely controlled by the intelligent control system, which can intelligently adjust the flow of foam additives according to the real-time monitored temperature data. Through instant feedback of temperature changes, the intelligent control system automatically adjusts the opening of the valve to control the flow of the foam agent and ensure accurate cooling during the construction process.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The present invention achieves full-coverage monitoring of the cutterhead temperature through a high-density sensor network, and optimizes the water spray angle in combination with the cutterhead structure and the flow direction of the slag. For downhill construction at a large inclination angle, the water spray angle is adjusted so that the water flow can impact along the downward direction of the slag, thereby enhancing the scouring effect on the slag, improving the cutterhead cleaning efficiency, and reducing the probability of cutterhead blockage. This can not only effectively prevent failures and damage caused by excessive cutterhead temperature or blockage, reduce safety risks during construction, but also avoid sudden accidents caused by this, ensuring the safety of construction workers. At the same time, it also reduces downtime and maintenance costs caused by cutterhead blockage, improves the utilization efficiency of TBM equipment, maintains a continuous and efficient excavation speed, and accelerates project progress.

[0037] 2. In the pulse jet mode of the present invention, the instantaneous large flow of water will generate a large thrust. This thrust acts on the dredging component, causing it to move inside the nozzle. The movement of the dredging component can clean the internal channel of the nozzle, preventing impurities such as rock chips and mud from accumulating and clogging in the nozzle. It can reduce the problem of jet function failure caused by nozzle blockage, ensure the continuous and stable cooling, lubrication and cleaning of the cutterhead during TBM construction, and at the same time reduce the frequency of manual nozzle cleaning, save maintenance time and labor costs, and help improve construction efficiency.

[0038] 3. By considering the relationship between multiple factors and the causes of cutterhead blockage, the present invention can flexibly adapt to the complex geological conditions and slope changes in high-angle downhill construction, adjust temperature control and foam additive supply strategies, and enhance the adaptability and reliability of TBM construction in complex environments. To a certain extent, it breaks through the limitations of traditional preventive measures, promotes the development and innovation of TBM construction technology, and provides valuable reference and reference for solving similar engineering problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the present invention;

[0040] Figure 2 This is a planar layout diagram of the high-precision temperature sensor of the present invention;

[0041] Figure 3 This is a schematic diagram of the planar assembly structure of the dredging component of the present invention;

[0042] Figure 4 Schematic diagram of the operation of the intelligent control switch of the present invention.

[0043] In the figure: 1. High-precision temperature sensor; 2. Screen; 3. Dredging rod. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] See also Figures 1-4 The present invention provides a method for preventing cutterhead blockage in high-angle downhill TBM construction. The method for preventing cutterhead blockage is as follows:

[0046] S1. Mill grooves uniformly on the surface of the cutter head and embed high-precision temperature sensors 1 to form a distributed temperature monitoring network;

[0047] S2. Replace the traditional nozzle with a dual-mode high-wear-resistant nozzle, optimize the spray angle, and add a proportional adaptive mixer to the original foam generating device. The dual-mode spray is conventional spray and pulse spray.

[0048] S3: Importing geological preset parameters into the TBM's intelligent control system, configuring a temperature threshold model, and training an LSTM neural network model based on historical construction data to predict cutterhead blockage risks.

[0049] S4, high-precision temperature sensor 1 collects temperature data every 0.1 seconds and transmits it to the edge computing device via CAN bus or LoRa wireless;

[0050] S5. The edge computing device can process and analyze the collected temperature data according to the data processing algorithm;

[0051] In S5, data processing and analysis include data storage, data analysis, data comparison, and data evaluation;

[0052] Data storage: including local storage and cloud storage, which can store the collected data for a long time and use it as training data for the LSTM neural network model;

[0053] Data analysis: including regular analysis of the collected data and records of foam additive addition;

[0054] Data comparison: compare the collected data with historical data;

[0055] Data evaluation: Evaluate the effectiveness of prevention strategies and identify potential problems and areas for improvement.

[0056] S6. Then, based on the analysis results and construction conditions, a strategy for adding foaming agents to the water spraying system and water spraying system parameters are formulated. The strategy for adding foaming agents includes using Python to regularly analyze temperature trends and foaming agent effects and generate optimization reports. The water spraying system parameter adjustments include water spraying pressure, water spraying volume, and water spraying time. The water spraying system is equipped with an intelligent control switch.

[0057] Furthermore, in S1: a distributed temperature monitoring network is constructed. Grooves are evenly milled on the cutterhead surface, and high-precision temperature sensors 1 are embedded therein. These grooves are designed to ensure stable installation of the temperature sensors without significantly affecting the cutterhead's structural strength. High-precision temperature sensors 1 are evenly distributed across the cutterhead surface at regular intervals, forming a distributed temperature monitoring network. This network enables real-time and accurate monitoring of temperature changes across different parts of the cutterhead, providing crucial data for subsequent assessment of whether the cutterhead is at risk of clogging.

[0058] In S2: The traditional nozzle is replaced with a dual-mode high-wear-resistant nozzle, and the optimized spray angle is ±15° perpendicular to the cutter head surface. This dual-mode high-wear-resistant nozzle has two modes: conventional spray and pulse spray. The conventional spray mode can continuously spray foam or water onto the cutter head surface to cool and lubricate the cutter head and dilute the rock cuttings. The pulse spray mode can use instantaneous high-pressure water flow or foam flow to impact and clean any deposits that may appear on the cutter head surface. The optimized spray angle can ensure that the spray medium covers the cutter head surface more comprehensively and effectively, thereby improving the spray effect. At the same time, a proportional adaptive mixer is added to the original foam generating device. The electromagnetic flowmeter can accurately measure the flow of raw liquid, water and air. The proportional valve driven by the servo motor can accurately adjust the input ratio of the three according to different construction conditions. Through the built-in multi-mode formula library, the operator can switch the mixing ratio of raw liquid, water and air with one click according to the actual geological conditions, excavation speed and other construction parameters to adapt to the requirements of foam performance under different working conditions. For example, when the rock hardness is high, the viscosity and lubricity of the foam are increased to better protect the cutterhead tools. In areas with abundant groundwater, the dilution of the foam is adjusted to enhance its dispersion and carrying capacity of mud and water.

[0059] In S3, geological preset parameters, including rock hardness, groundwater content, and formation stability, are imported into the TBM's intelligent control system. A temperature threshold model is configured based on these parameters, with different geological conditions corresponding to different temperature threshold ranges. Simultaneously, an LSTM neural network model is trained based on historical construction data, which contains rich information such as the cutterhead's operating status, temperature changes, and blockage conditions under different geological conditions and tunneling parameters. By learning from this data, the LSTM neural network model can establish a complex mapping relationship between geological conditions, cutterhead operating parameters, and cutterhead blockage risk, thereby predicting the risk of cutterhead blockage. When the temperature in a region exceeds the set threshold, the system automatically increases the foam injection volume in the corresponding area to enhance the cooling and lubrication effect. The system also dynamically adjusts the foam mixing ratio to improve the foam's ability to absorb and carry rock cuttings. The system also activates pulsed injection mode, using the impact force of the pulsed injection to clear deposits. When the LSTM neural network model predicts that the cutterhead is at risk of blockage, the TBM's control system will monitor and obtain the flow and boundary pressure data of each flushing pipeline in the mud pipeline in real time, and obtain the flow and node pressure distribution of each pipeline through simulation calculation, and dynamically adjust the flow distribution. If the flushing pipeline flow in a certain area is found to be insufficient, the system will automatically increase the flow of that pipeline to ensure that all parts of the cutterhead are adequately flushed and cleaned, preventing further blockage.

[0060] In S4: High-precision temperature sensor 1 collects temperature data every 0.1 seconds. This high-frequency data collection can capture subtle changes in the cutter head temperature in a timely manner. The collected temperature data is transmitted to the edge computing device via the CAN bus or LoRa wireless. The CAN bus has the characteristics of reliable communication and strong real-time performance. It is suitable for data transmission in a relatively stable construction environment and convenient wiring. LoRa wireless transmission has the advantages of low power consumption and long-distance transmission. It can play a good role in areas with complex construction environments and difficult wiring. It ensures that temperature data can be transmitted to the edge computing device stably and quickly, providing timely data support for subsequent data processing and analysis. Whether it can be reasonably arranged according to the actual construction situation.

[0061] In S5: Edge computing devices can process and analyze collected temperature data based on data processing algorithms. Data storage includes local and cloud storage. Local storage can use large-capacity solid-state drives or disk arrays to save multiple collected data in real time, ensuring local security and accessibility. Cloud storage uses the cloud computing platform's storage services to back up data to the cloud to prevent data loss due to local storage device failures. This long-term data not only serves as a record of the construction process but also as training data for the LSTM neural network model, continuously optimizing the model's predictive performance and improving the accuracy of cutterhead blockage risk predictions. Data analysis includes regular analysis of multiple collected data points and foam additive addition records. Through data analysis, potential relationships between data can be discovered, such as the correlation between temperature changes and the amount of foam additive added, and the changing patterns of cutterhead temperature under different geological conditions. This helps to better understand the operating status of the cutterhead and provides a basis for developing more reasonable prevention strategies. Data comparison involves comparing collected data with historical data. This comparison allows us to determine whether the current operating status of the cutterhead is normal, allowing us to promptly detect abnormalities. For example, if the temperature of a certain area of the cutterhead is significantly higher than historical data and persists for a long period of time, this indicates a blockage risk in that area, requiring timely action. By comparing and analyzing the cutterhead operating status data before and after the implementation of preventive strategies, we can evaluate the effectiveness of various preventive measures. For example, we can observe whether the cutterhead temperature is effectively controlled and whether blockages are reduced after adjusting the foam injection volume and mixing ratio. If certain preventive strategies are found to be ineffective, we can identify the problem based on the evaluation results and optimize and improve the preventive strategies.

[0062] In S6, the foaming agent addition strategy uses Python to regularly analyze temperature trends and foaming agent effects, generating optimization reports. Python can perform in-depth analysis of temperature data and foaming agent addition records, such as plotting temperature changes over time. It then analyzes the impact of foaming agent on cutterhead temperature by combining the amount and time of addition. Based on these results, an optimization report is generated, providing reasonable foaming agent addition recommendations, such as increasing the amount of foaming agent during periods of rapid temperature rise and appropriately reducing it when the temperature stabilizes, to achieve optimal anti-clogging effects while also reducing foaming agent waste and saving costs. The adjustment of water spray system parameters includes water spray pressure, water spray volume and water spray time. According to the operating status of the cutter disc, geological conditions and temperature changes, the water spray pressure, water spray volume and water spray time should be reasonably adjusted. In areas where the cutter disc temperature is high and there are more rock chips, the water spray pressure and water spray volume should be appropriately increased, and the water spray time should be extended to enhance the cooling and cleaning effects. In areas where the cutter disc operation is relatively stable and the temperature is normal, the water spray pressure and water spray volume should be reduced, the water spray time should be shortened, and water resources should be saved. At the same time, adverse effects on the construction environment caused by excessive water spraying should be avoided.

[0063] A dredging component is provided inside the nozzle, and the dredging component is pushed to move by the size of the water flow in different modes. The dredging component includes a mesh plate 2 and a dredging rod 3. The mesh plate 2 is placed inside the nozzle, and the outer wall of one side of the mesh plate 2 is installed with evenly arranged dredging rods 3, and one end of the dredging rod 3 extends into the water spray hole of the nozzle. The diameter of the dredging rod 3 is smaller than the diameter of the water spray hole.

[0064] Furthermore, in the conventional spray mode, only a part of the dredging rod 3 is located inside the water spray hole. When switching to the pulse spray mode, the increased water flow and impact force will push the mesh plate 2 to move, thereby pushing the dredging rod 3 to insert into the outermost side of the water spray hole, thereby pushing out the blocked mud cake in the water spray hole, and cooperating with the water flow impact to make the water spray hole no longer blocked, thereby reducing the probability of the water spray hole being blocked to a certain extent.

[0065] The proportional adaptive mixer integrates an electromagnetic flowmeter and a proportional valve driven by a servo motor. It has a built-in multi-mode formula library and can switch the input ratio of raw liquid, water and air with one click to adapt to different working conditions according to different construction conditions.

[0066] Furthermore, the system automatically or manually selects the optimal strategy based on different geological conditions (soft rock, hard rock, clay), and construction conditions (slope, temperature, and humidity) to precisely prevent cutterhead blockage. Alternatively, it automatically matches the optimal recipe based on real-time monitoring data such as temperature, pressure, and geological parameters. Combining historical data with machine learning models, it continuously updates recipe parameters, allowing engineers to manually add, modify, or lock in specific recipes. Presets based on geological reports and experimental data, such as clay formations requiring a higher proportion of digester to break down sticky materials, can also be used to analyze historical construction data and discover optimal mix patterns. Input features include temperature, lithology, slope, and foamer dosage, and output targets include blockage risk level and construction efficiency score.

[0067] S3 also includes: when the regional temperature exceeds the set threshold, increasing the foam injection volume in the corresponding area, dynamically adjusting the foam mixing ratio, and starting the pulse injection mode. In S3, when the LSTM neural network model predicts that there is a risk of cutterhead blockage, the TBM control system will monitor and obtain the flow and boundary pressure data of each flushing pipeline in the mud pipeline in real time, and obtain the flow and node pressure distribution of each pipeline through simulation calculation, and dynamically adjust the flow distribution.

[0068] Furthermore, the LSTM neural network model establishes a complex mapping relationship between the risk of cutterhead blockage and various factors by learning and training on a large amount of historical construction data, including cutterhead operating parameters, slag characteristics, construction environment and other multi-dimensional data. When the model monitors that the current input data, such as abnormal cutterhead torque and slag temperature changes, meets the characteristics of blockage risk, it will predict that there is a risk of cutterhead blockage. After receiving the blockage risk warning from the LSTM neural network model, the TBM control system quickly initiates real-time monitoring of the flow rate and boundary pressure data of each flushing pipeline in the mud pipeline. These data reflect the flow state and pressure distribution of the mud in the pipeline, and form the basis for subsequent analysis and adjustment. The simulation calculation uses the relevant principles and algorithms of fluid mechanics, and performs simulation calculations based on the flow and pressure data obtained from real-time monitoring. By establishing a mathematical model of the mud pipeline system, the flow process of the mud in the pipeline is simulated, so as to obtain the flow rate of each pipeline and the pressure distribution of the node. It can clearly show the dynamic behavior of the mud in the entire pipeline system, find out the problems such as uneven flow and abnormal pressure in the pipeline system, and identify the pipeline areas with unreasonable flow distribution or abnormal pressure based on the simulation calculation results. The control system dynamically adjusts the flow distribution by adjusting the opening of relevant valves, the output power of the pump, etc. For example, for pipelines with small flow, which may cause the debris to be unable to be effectively washed away, its flow rate is increased. For areas where the pressure is too high, which may cause pipeline damage or affect the normal transportation of mud, the pressure is appropriately reduced to ensure that the mud can evenly and efficiently wash the cutter head and reduce the risk of blockage.

[0069] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for preventing cutterhead blockage in high-angle downhill TBM construction, characterized in that: The following are the methods to prevent cutter disc blockage: S1, evenly milling grooves on the surface of the cutter head, embedding high-precision temperature sensors (1) to form a distributed temperature monitoring network; S2. Replace the traditional nozzle with a dual-mode high-wear-resistant nozzle, optimize the spray angle, and add a proportional adaptive mixer to the original foam generating device. The dual-mode spray is conventional spray and pulse spray. S3: Importing geological preset parameters into the TBM's intelligent control system, configuring a temperature threshold model, and training an LSTM neural network model based on historical construction data to predict cutterhead blockage risks. S4, a high-precision temperature sensor (1) collects temperature data every 0.1 seconds and transmits it to the edge computing device via CAN bus or LoRa wireless; S5. The edge computing device can process and analyze the collected temperature data according to the data processing algorithm; S6. Then, according to the analysis results and construction conditions, the strategy for adding foaming agent in the water spraying system and the parameters of the water spraying system are formulated.

2. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1 is characterized in that: A dredging component is provided inside the nozzle, and the dredging component is driven to move by the size of the water flow in different modes.

3. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1 is characterized in that: The proportional adaptive mixer integrates an electromagnetic flowmeter and a proportional valve driven by a servo motor. It has a built-in multi-mode formula library and can switch the input ratio of raw liquid, water and air with one click to adapt to different working conditions according to different construction conditions.

4. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 2, characterized in that: The dredging assembly comprises a mesh plate (2) and a dredging rod (3), wherein the mesh plate (2) is placed inside the nozzle, and the dredging rods (3) are evenly arranged on the outer wall of one side of the mesh plate (2), and one end of the dredging rod (3) extends into the water spray hole of the nozzle, and the diameter of the dredging rod (3) is smaller than the diameter of the water spray hole.

5. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: In S3, it also includes: when the temperature of the area exceeds the set threshold, increasing the foam injection amount of the corresponding area, dynamically adjusting the foam mixing ratio and starting the pulse injection mode.

6. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: In S3, when the LSTM neural network model predicts that the cutterhead is at risk of blockage, the TBM control system will monitor the flow and boundary pressure data of each flushing pipeline in the mud pipeline in real time, and obtain the flow and node pressure distribution of each pipeline through simulation calculation, and dynamically adjust the flow distribution.

7. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: In S5, edge computing devices are also used to collect various operating parameters of the TBM.

8. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: In S5, data processing and analysis include data storage, data analysis, data comparison, and data evaluation; Data storage: including local storage and cloud storage, which can store the collected data for a long time and use it as training data for the LSTM neural network model; Data analysis: including regular analysis of the collected data and records of foam additive addition; Data comparison: compare the collected data with historical data; Data evaluation: Evaluate the effectiveness of prevention strategies and identify potential problems and areas for improvement.

9. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: In S6, the foaming agent addition strategy includes using Python to regularly analyze temperature trends and foaming agent effects and generate optimization reports.

10. The method for preventing cutterhead blockage in high-angle downhill TBM construction according to claim 1, characterized in that: The parameters of the water spraying system include water spraying pressure, water spraying volume and water spraying time. The water spraying system is equipped with an intelligent control switch.

Images