A method for synergistic control of TBM tunneling parameters and wear-resistant agents based on cutter service conditions

By constructing models for predicting cutter wear and tunneling rate, and combining them with the NSGA-II optimization algorithm, the parameters and wear-resistant agents under the service state of the cutter were coordinated and controlled. This solved the problems of accelerated cutter wear and high construction costs in existing technologies, and improved the construction efficiency and safety of TBMs.

CN123064487APending Publication Date: 2026-11-10CHINA UNIV OF MINING & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611356544.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-03
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing TBM cutter wear control technologies fail to achieve coordinated optimization of tunneling parameters and wear-resistant agent injection parameters, leading to increased cutter wear and a lack of dynamic correction, resulting in high construction costs and safety hazards.

Method used

Based on the service status of the cutterhead, wear prediction and tunneling rate prediction models are constructed by acquiring TBM tunneling parameters, cutterhead temperature and vibration information. The NSGA-II multi-objective optimization algorithm is used to obtain the Pareto optimal solution and coordinate the control of cutterhead parameters and wear-resistant agent injection.

Benefits of technology

It achieves coordinated control of minimizing cutter wear, maximizing tunneling efficiency, and minimizing wear-resistant agent usage, thereby improving the construction efficiency and safety of TBMs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123064487A_ABST
    Figure CN123064487A_ABST
Patent Text Reader

Abstract

This invention relates to a method for coordinated control of TBM tunneling parameters and wear-resistant agents based on the service status of cutterheads. The method includes: acquiring TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters; extracting cutterhead temperature and vibration characteristics based on the cutterhead temperature and vibration information, comparing them using preset corresponding discrimination thresholds to determine the cutterhead service status; constructing a cutterhead wear prediction model and a tunneling rate prediction model by combining the TBM tunneling parameters and wear-resistant agent injection parameters to predict cutterhead wear and tunneling rate under different combinations of tunneling parameters; constructing a coordinated optimization model with the optimization objectives of maximizing tunneling rate, minimizing cutterhead wear, and minimizing wear-resistant agent usage to obtain a Pareto optimal solution set; determining the optimal execution plan based on the Pareto optimal solution set and the cutterhead service status, and implementing control using a corresponding control system; acquiring the cutterhead status information after control and revising the control plan for the next control cycle.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for tunnel boring machines (TBMs), and in particular to a method for the coordinated regulation of TBM tunneling parameters and wear-resistant agents based on the service status of the cutter head. Background Technology

[0002] As the core rock-breaking tool of a full-face hard rock tunnel boring machine (TBM), the cutterhead's service condition directly affects the TBM's rock-breaking efficiency, tunneling stability, and continuous construction capability. In high-strength, highly abrasive hard rock formations, the cutterhead wear rate is rapid and the failure modes are complex. Frequent unplanned cutterhead replacements not only drastically increase construction costs but may also induce engineering accidents such as face instability and cutterhead jamming. Therefore, there is an urgent need to develop an efficient TBM cutterhead wear control technology to reduce abnormal cutterhead wear and extend its service life.

[0003] Existing prevention and control measures mainly include two technical approaches: tunneling parameter control and material protection. On the one hand, based on signals such as changes in the strata, abnormal vibration or noise of the cutterhead, or torque fluctuations, parameters such as cutterhead thrust, rotational speed, and penetration are adjusted to reduce cutter wear. On the other hand, by injecting wear-resistant agents into the cutterhead-face interface, the friction coefficient between the cutter ring and rock cuttings is reduced to extend the service life of the cutterhead.

[0004] However, existing TBM hobbing wear control technologies still have the following shortcomings: 1) Existing intelligent control methods mainly reduce the load on the cutter head by adjusting parameters such as cutterhead thrust, rotation speed and penetration depth, while the abrasion-resistant agent injection still uses a preset flow rate. Mechanical load reduction and material wear reduction are independent of each other, and the synergistic optimization of tunneling parameters and abrasion-resistant agent injection parameters has not yet been achieved. 2) Increased wear of the cutter head may be caused by factors such as parameter mismatch, heat accumulation, or abnormal rotation. If the method of simply reducing the load or adding wear-resistant agent is used, it is easy to cause excessive parameter adjustment, ineffective consumption of wear-resistant agent, and decreased tunneling efficiency. Existing methods lack the ability to identify the service status of the cutter head and to differentiate the parameters. 3) Existing methods lack continuous evaluation of the adjusted cutter condition and the effect of wear-resistant agent, making it difficult to dynamically correct subsequent tunneling parameters and wear-resistant agent application strategies, and a closed-loop control system has not yet been formed. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for the coordinated control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head. This method can coordinately control the TBM tunneling parameters and wear-resistant agent injection parameters according to the service status of the cutter head, thereby reducing cutter head wear and improving TBM tunneling efficiency and continuous tunneling capability.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for synergistic control of TBM tunneling parameters and wear-resistant agents based on cutter service status includes: Obtain TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters; Based on the hob temperature and vibration information, the hob temperature characteristics and vibration characteristics are extracted, and compared with preset corresponding discrimination thresholds to determine the hob service status. Combining the TBM tunneling parameters and wear-resistant agent injection parameters, a cutter wear prediction model and a tunneling rate prediction model are constructed respectively to predict the cutter wear amount and the tunneling rate under different combinations of tunneling parameters; With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, a collaborative optimization model is constructed to obtain the Pareto optimal solution set. Based on the Pareto optimal solution set, the optimal execution plan is determined according to the service status of the hob, and the corresponding control system is used to perform the control. Obtain the status information of the roller cutter after adjustment, and revise the adjustment scheme for the next control cycle.

[0007] Optionally, the TBM tunneling parameters include: total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth; The vibration information includes: vibration acceleration signal; The wear-resistant agent injection parameters include: wear-resistant agent injection flow rate.

[0008] Optionally, obtaining TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters includes: The acquired TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters are synchronized in time, outlier removal, filtered, and normalized. The data is then divided according to a set time window to ensure that the TBM tunneling parameters, cutter status information, and wear-resistant agent injection parameters within the same time window correspond to each other, forming a training dataset.

[0009] Optionally, determining the service status of the hob includes: The hob temperature and vibration characteristics are extracted from the real-time monitored hob temperature and vibration information; wherein, the hob temperature characteristics include: hob temperature and temperature rise rate, and the vibration characteristics include: root mean square value of vibration, peak value and periodicity characteristics; The service status of the cutter is determined by comparing the temperature and vibration characteristics of the cutter with preset corresponding discrimination thresholds. The service status of the cutter refers to the operating status of the cutter during the tunneling process, including normal rolling status, abnormal temperature rise status, abnormal rolling status, and uneven wear status.

[0010] Optionally, constructing the hob wear prediction model includes: Using TBM tunneling parameters and wear-resistant agent injection parameters as input features, the actual wear amount of the cutterhead obtained from the opening inspection as the training label, and the cutterhead wear evolution law as a physical constraint, a physical information neural network is trained to construct a cutterhead wear prediction model and output the predicted cutterhead wear amount. The cutterhead wear evolution law includes constraints based on Arcard's wear law, wear monotonicity constraints, and physical boundary constraints. The TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth.

[0011] Optionally, constructing the tunneling rate prediction model includes: The total thrust, cutterhead rotation speed, cutterhead torque, and penetration depth are used as model inputs, and the actual tunneling rate under the corresponding working conditions is used as the training label to train a BP neural network and construct a tunneling rate prediction model, which is used to output the predicted tunneling rate under different combinations of tunneling parameters.

[0012] Optionally, obtaining the Pareto optimal solution set includes: Using the TBM tunneling parameters and wear-resistant agent injection parameters as decision variables, and the cutter wear prediction model and tunneling rate prediction model as evaluation models, the cutter wear amount and tunneling rate corresponding to different parameter combinations are calculated. With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, the NSGA-II multi-objective optimization algorithm was used to solve the Pareto optimal solution set composed of different total thrust, cutterhead rotation speed, and wear-resistant agent injection flow rate.

[0013] Optionally, based on the Pareto optimal solution set, the optimal execution plan is determined according to the hob's service status, including: Based on the current service status of the hob, select the corresponding execution plan from the Pareto optimal solution set; When the cutter head is in normal rolling condition, the scheme with predicted tunneling rate higher than preset rate threshold and predicted cutter head wear meeting the constraint conditions is selected; when the cutter head experiences abnormal temperature rise, the scheme with predicted cutter head wear lower than preset wear threshold and wear-resistant agent injection flow rate higher than normal working condition setting value is selected; when the cutter head experiences abnormal rolling or uneven wear, the scheme with total thrust and cutterhead speed lower than current working condition setting value and wear-resistant agent injection flow rate higher than normal working condition setting value is selected. After determining the execution plan, adjust the TBM total thrust, cutterhead speed and wear-resistant agent injection flow rate according to the execution plan.

[0014] Optionally, obtaining the adjusted cutter state information and revising the adjustment scheme for the next control cycle includes: Obtain the cutterhead status information after adjustment; compare the service status before and after adjustment; when the cutterhead status improves, maintain the current adjustment parameters or enter the next control cycle; when the cutterhead status does not improve or deteriorates further, re-perform cutterhead wear prediction, tunneling rate prediction and multi-objective optimization with the updated cutterhead status to obtain the total thrust, cutterhead speed and wear-resistant agent injection flow rate for the next control cycle.

[0015] A device for coordinated control of TBM tunneling parameters and wear-resistant agent based on cutter service status includes: The data acquisition module is used to acquire TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters, and to perform data preprocessing to form a training dataset; wherein, the TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate and penetration depth, and the wear-resistant agent injection parameters include wear-resistant agent injection flow rate; The hob condition discrimination module is used to extract hob temperature features and vibration features based on the hob temperature and vibration information, and compare them with preset discrimination thresholds to determine the hob service status. The parameter prediction module is used to combine the TBM tunneling parameters and wear-resistant agent injection parameters, and to construct a cutter wear prediction model and a tunneling rate prediction model using a physical information neural network and a BP neural network, respectively, to predict the cutter wear amount and the tunneling rate under different combinations of tunneling parameters; The collaborative optimization module uses total thrust, cutterhead speed and wear-resistant agent injection flow rate as decision variables, and takes the maximum tunneling rate, minimum cutter wear and minimum wear-resistant agent usage as optimization objectives. It uses the NSGA-II multi-objective optimization algorithm to solve the problem and obtain a Pareto optimal solution set composed of different total thrust, cutterhead speed and wear-resistant agent injection flow rate. The control and decision module is used to determine the optimal execution plan based on the Pareto optimal solution set and the service status of the hob, that is, to determine the total thrust, cutter head speed and wear-resistant agent injection flow rate, and to execute the control using the corresponding control system. The feedback correction module is used to obtain the status information of the hobbing cutter after adjustment and correct the adjustment scheme for the next control cycle.

[0016] The beneficial effects of this invention are as follows: 1) This invention uses hob temperature and vibration information to determine the service status of the hob, which can identify the hob's normal rolling, abnormal rolling and uneven wear, and use the hob status as the basis for subsequent parameter adjustment, so that the TBM control process can take into account the actual working status of the hob. 2) This invention establishes a cutter wear prediction model and a tunneling rate prediction model respectively, and predicts the cutter wear and tunneling rate under different combinations of tunneling parameters and wear-resistant agent injection parameters, providing an evaluation basis for subsequent parameter optimization; 3) This invention incorporates the total thrust of the TBM, the cutterhead rotation speed, and the abrasion-resistant agent injection flow rate into the same optimization process. The Pareto optimal solution is obtained with the goal of maximizing the tunneling rate, minimizing the cutter wear, and minimizing the abrasion-resistant agent consumption. The execution plan is determined based on the cutter service status, thereby achieving coordinated control of TBM tunneling parameters and abrasion-resistant agent injection parameters. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a method for synergistic control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of a TBM tunneling parameter and wear-resistant agent co-regulation device based on the service status of the cutter head, according to an embodiment of the present invention. Detailed Implementation

[0019] 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.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1 like Figure 1 As shown, this embodiment proposes a method for synergistic control of TBM tunneling parameters and wear-resistant agents based on the service status of the cutter head, including: S1. Obtain TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters; S2. Based on the hob temperature and vibration information, extract the hob temperature characteristics and vibration characteristics, and compare them with the preset corresponding discrimination thresholds to determine the hob service status; S3. Combining TBM tunneling parameters and wear-resistant agent injection parameters, construct a cutter wear prediction model and a tunneling rate prediction model respectively to predict cutter wear and tunneling rate under different combinations of tunneling parameters; S4. With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, a collaborative optimization model is constructed to obtain the Pareto optimal solution set. S5. Based on the Pareto optimal solution set, determine the optimal execution plan according to the service status of the hob, and use the corresponding control system to perform the control. S6. Obtain the status information of the roller cutter after adjustment, and correct the adjustment scheme for the next control cycle.

[0022] Furthermore, in S1, the TBM tunneling parameters include: total thrust. F n (kN), cutter head speed RPM (rpm), cutter head torque T q (kN·m), tunneling rate v (mm / min) and penetration h (mm / r); where total thrust and cutterhead speed are candidate control parameters for subsequent collaborative optimization; TBM tunneling parameters are acquired in real time by the TBM main control system or programmable logic controller; Furthermore, hobbing temperature T ( t Vibration information is acquired through temperature and vibration sensors located at the cutter shaft or cutter holder; the vibration information includes vibration acceleration signals. a ( t The system extracts the root mean square value, peak value, and frequency domain features of the vibration acceleration signal to distinguish between normal rolling, abnormal rolling, and uneven wear of the hob. The root mean square value of vibration is used to characterize the vibration intensity of the hob, and its calculation formula is as follows: In the formula, a i For the first time window i One vibration acceleration sample value, N This represents the number of sampling points within this time window.

[0023] Furthermore, the wear-resistant agent injection parameters include at least the wear-resistant agent injection flow rate. Q It may include injection pressure; the wear-resistant agent ratio is predetermined in the current construction section, the injection flow rate is used as a candidate control parameter for the collaborative optimization model, and the injection pressure is used to monitor the working status of the wear-resistant agent injection system.

[0024] Specifically, the wear-resistant agent is delivered to the injection position in front of the cutter head via a supply line, and a flow meter and a pressure sensor are installed on the supply line to collect the real-time injection flow rate of the wear-resistant agent. Q ( t ) and injection pressure P ( tThe ratio of the wear-resistant agent is predetermined based on the rock abrasiveness and wear resistance test results of the current construction section. During online control, the main adjustment is the injection flow rate of the wear-resistant agent. The injection flow rate of the wear-resistant agent is adjusted through a flow regulating valve and fed back to the control system so that the actual injection flow rate follows the target injection flow rate.

[0025] Furthermore, the acquired TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters are subjected to time synchronization, outlier removal, filtering, and normalization. The data is then divided according to a set time window to ensure that the TBM tunneling parameters, cutter status information, and wear-resistant agent injection parameters within the same time window correspond to each other, forming a training dataset.

[0026] Specifically, the preprocessing steps include: Using the TBM main control system timestamp as a unified benchmark, the total thrust, cutterhead speed, cutterhead torque, tunneling rate, penetration depth, cutterhead temperature, vibration information, and wear-resistant agent injection flow rate are correlated according to the time of acquisition; for data with different sampling frequencies, resampling is performed according to a unified sampling interval. Outlier filtering is performed on the synchronized data to remove data that exceeds the normal operating range of the equipment; for data with sudden changes, the 3σ criterion can be used for judgment. Smoothing and filtering are performed on TBM tunneling parameters and cutter temperature signals, and noise reduction and filtering are performed on vibration acceleration signals; Normalize the different dimensional parameters used for neural network training: in, x i This is the original data. For the normalized data, x min and x max These are the minimum and maximum values ​​of the corresponding parameters, respectively. Specifically, the matched data is used as training samples for the model to form training datasets for the cutter wear prediction model and the tunneling rate prediction model.

[0027] Further, determining the service status of the hob includes: The temperature and vibration characteristics of the hob are extracted from the real-time monitored hob temperature and vibration information. The temperature characteristics include hob temperature and temperature rise rate, and the vibration characteristics include root mean square value, peak value and periodicity. The service status of the hob is determined by comparing its temperature and vibration characteristics with preset threshold values.

[0028] Specifically, the extracted temperature and vibration characteristics of the cutter are compared with preset corresponding discrimination thresholds, and the service status of the cutter is determined based on the judgment results of each characteristic parameter. The service status of the cutter refers to the operating status of the cutter during the tunneling process, including normal rolling state, abnormal temperature rise state, abnormal rolling state, and uneven wear state, but does not include the cutter wear amount. The cutter wear amount or wear rate is predicted by an independent cutter wear prediction model. Specifically, the hob temperature and temperature rise rate are used to determine the heat accumulation during hob operation; the root mean square value of vibration is used to characterize the overall vibration intensity of the hob, the peak value is used to identify transient impact characteristics, and the periodicity characteristic is used to reflect the rotational stability of the hob. These characteristic parameters are compared with their corresponding discrimination thresholds, and the hob service status is determined by combining the judgment results of different characteristic parameters. Specifically, the vibration sensor sampling frequency was set to 10 kHz, with 1 second as an analysis time window and an overlap rate of 50% between adjacent windows. First, vibration data for at least 30 minutes under normal rolling conditions of the hob were selected to establish a normal state benchmark, and the root mean square benchmark value of the vibration was calculated for each. Peak reference value and cyclical indicators P 0.

[0029] Set the hob temperature threshold and vibration characteristic threshold, and determine the hob status by comparing the real-time monitoring value with the corresponding threshold; when the number of abnormal hobs reaches the set value, determine the corresponding control scheme.

[0030] As shown in Table 1, the specific discrimination table is as follows: Table 1 Furthermore, the construction of the hob wear prediction model includes: Using TBM tunneling parameters and wear-resistant agent injection parameters as input features, the actual wear amount of the cutterhead obtained from the opening inspection as the training label, and the cutterhead wear evolution law as a physical constraint, a physical information neural network is trained to construct a cutterhead wear prediction model and output the predicted cutterhead wear amount; wherein, the TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth.

[0031] Specifically, the hob wear evolution law includes: constraints based on Archard's wear law, wear monotonicity constraints, and physical boundary constraints; the total loss function is: In the formula, L data and L phy These are the data loss term and the physical constraint loss term, with corresponding weighting coefficients of . λdata and λ phy .

[0032] Furthermore, the construction of the tunneling rate prediction model includes: The total thrust, cutterhead rotation speed, cutterhead torque, and penetration depth are used as model inputs, and the actual tunneling rate under the corresponding working conditions is used as the training label to train a BP neural network to build a tunneling rate prediction model, which is used to output the predicted tunneling rate under different combinations of tunneling parameters.

[0033] Specifically, the training dataset is divided into a training set, a validation set, and a test set in a 7:2:1 ratio for training, parameter tuning, and prediction performance evaluation of the BP neural network. A BP neural network consists of an input layer, hidden layers, and an output layer. The input layer has four neurons, corresponding to the total thrust, cutterhead rotation speed, cutterhead torque, and penetration depth, respectively. The output layer has one neuron, which outputs the predicted tunneling rate. Two hidden layers are set up, with 16 and 8 neurons respectively. The hidden layers use the ReLU activation function, and the output layer uses the linear activation function. During training, the mean square error between the predicted tunneling rate and the actual tunneling rate is used as the loss function: In the formula, N The number of training samples. v i For the first i The actual tunneling rate of each sample The corresponding predicted tunneling rate is determined; the backpropagation algorithm is used to update the network weights and biases until the validation set loss reaches the set convergence condition, thus obtaining the trained tunneling rate prediction model.

[0034] Furthermore, obtaining the Pareto optimal solution set includes: Using TBM tunneling parameters and wear-resistant agent injection parameters as decision variables, and the cutter wear prediction model and tunneling rate prediction model as evaluation models, the cutter wear amount and tunneling rate corresponding to different parameter combinations are calculated. With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, the NSGA-II multi-objective optimization algorithm was used to solve the problem, obtaining a Pareto optimal solution set composed of different total thrust, cutterhead speed, and wear-resistant agent injection flow rate.

[0035] Specifically, the collaborative optimization process includes: TBM total thrust F n Cutter head speed RPM and wear-resistant agent injection flow rate QAs decision variables in multi-objective optimization, they constitute candidate parameter combinations: The value ranges of each variable are set according to the allowable operating range of the TBM equipment and the operating range of the wear-resistant agent injection device; different combinations of total thrust, cutterhead speed, and wear-resistant agent injection flow rate are generated within the above parameter ranges. For the first... j One candidate solution: The current control cycle's cutterhead torque, penetration depth, and other state parameters are combined with candidate parameters and input into the trained cutter wear prediction model and tunneling rate prediction model. The optimization objectives are to maximize the predicted tunneling rate, minimize the predicted cutter wear, and minimize the wear-resistant agent injection flow rate. Therefore, the following structure is constructed: in, To predict the tunneling rate, To predict wear and tear; In this embodiment, the initial population size is set to 100, the maximum number of iterations is 200, the crossover probability is 0.9, and the mutation probability is set according to the number of decision variables. Three objective function values ​​are calculated for each candidate parameter combination, and fast non-dominated sorting, crowding calculation, selection, crossover and mutation are performed in sequence to generate the next generation of candidate schemes. The algorithm stops iterating when it reaches the maximum number of iterations or when the objective function changes to meet the set convergence conditions, thus obtaining the Pareto optimal solution set.

[0036] Furthermore, based on the Pareto optimal solution set and according to the service status of the hobbing cutter, the optimal execution plan is determined as follows: Based on the current service status of the cutterhead, the corresponding execution scheme is selected from the Pareto optimal scheme set. When the cutterhead is in normal rolling condition, the scheme with a higher predicted tunneling rate and the predicted cutterhead wear meets the constraint conditions is selected first. When the cutterhead exhibits abnormal temperature rise, the scheme with a lower predicted cutterhead wear and a wear-resistant agent injection flow rate higher than the normal operating condition setting value is selected first to reduce the accumulation of frictional heat during the contact process between the cutterhead and the rock. When the cutterhead exhibits abnormal rolling or uneven wear, the scheme with a total thrust and cutterhead speed lower than the current operating condition setting value and a wear-resistant agent injection flow rate higher than the normal operating condition setting value is selected first to reduce the impact load on the cutterhead and improve the wear reduction effect. After determining the execution scheme, the TBM total thrust, cutterhead speed, and wear-resistant agent injection flow rate are adjusted according to the execution scheme.

[0037] For example, in hard rock tunneling conditions, when the TBM total thrust is 18000 kN, the cutterhead speed is 6 rpm, and the wear-resistant agent injection flow rate is 30 L / min, if the cutterhead is in normal rolling condition, a scheme with higher tunneling efficiency is selected from the Pareto optimal scheme set. For example, the total thrust is adjusted to 19000 kN, the cutterhead speed is adjusted to 7 rpm, and the wear-resistant agent injection flow rate is adjusted to 35 L / min. If the cutterhead exhibits abnormal temperature rise, the load level on the cutterhead is reduced. For example, the total thrust is adjusted to 16000 kN, the cutterhead speed is adjusted to 5 rpm, and the wear-resistant agent injection flow rate is increased to 50 L / min. If the cutterhead exhibits abnormal rolling or uneven wear, the mechanical load is further reduced. For example, the total thrust is adjusted to 14000 kN, the cutterhead speed is adjusted to 4 rpm, and the wear-resistant agent injection flow rate is increased to 60 L / min.

[0038] Furthermore, obtaining the adjusted cutter status information and revising the control scheme for the next control cycle includes: After completing the parameter adjustment for the current control cycle, obtain the cutterhead status information after adjustment; compare the service status before and after adjustment; when the cutterhead status improves, maintain the current adjustment parameters or enter the next control cycle; when the cutterhead status does not improve or deteriorates further, re-perform cutterhead wear prediction, tunneling rate prediction and multi-objective optimization with the updated cutterhead status to obtain the total thrust, cutterhead speed and wear-resistant agent injection flow rate for the next control cycle.

[0039] Example 2 like Figure 2 As shown, this embodiment proposes a device for the coordinated control of TBM tunneling parameters and wear-resistant agents based on the service status of the cutter head, including: The data acquisition module is used to acquire TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters, and to perform data preprocessing to form a training dataset. The TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth, while the wear-resistant agent injection parameters include wear-resistant agent injection flow rate. The hob condition discrimination module is used to extract hob temperature and vibration characteristics based on hob temperature and vibration information, and compare them with preset discrimination thresholds to determine the hob service status. The parameter prediction module is used to combine TBM tunneling parameters and wear-resistant agent injection parameters, and to construct cutter wear prediction models and tunneling rate prediction models using physical information neural networks and BP neural networks, respectively, to predict cutter wear and tunneling rate under different combinations of tunneling parameters. The collaborative optimization module uses total thrust, cutterhead speed and wear-resistant agent injection flow rate as decision variables, and takes the maximum tunneling rate, minimum cutter wear and minimum wear-resistant agent usage as optimization objectives. It uses the NSGA-II multi-objective optimization algorithm to solve the problem and obtain a Pareto optimal solution set composed of different total thrust, cutterhead speed and wear-resistant agent injection flow rate. The control and decision module is used to determine the optimal execution plan based on the Pareto optimal solution set and the service status of the hob, that is, to determine the total thrust, cutter head speed and wear-resistant agent injection flow rate, and to execute the control using the corresponding control system. The feedback correction module is used to obtain the state of the roller cutter after regulation, evaluate the regulation effect, and update the regulation scheme for the next control cycle.

[0040] Example 3 This embodiment also provides a TBM tunneling parameter and wear-resistant agent coordinated control system based on the service status of the cutter head, including: a memory and a processor. The memory stores a computer program that is run by the processor. When the computer program is run by the processor, it executes a TBM tunneling parameter and wear-resistant agent coordinated control method based on the service status of the cutter head.

[0041] Example 4 This embodiment also provides a storage medium storing a computer program, which executes a method for the coordinated control of TBM tunneling parameters and wear-resistant agents based on the service status of the cutter head.

[0042] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for synergistic control of TBM tunneling parameters and wear-resistant agents based on the service status of the cutterhead, characterized in that, include: Obtain TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters; Based on the hob temperature and vibration information, the hob temperature characteristics and vibration characteristics are extracted, and compared with preset corresponding discrimination thresholds to determine the hob service status. Combining the TBM tunneling parameters and wear-resistant agent injection parameters, a cutter wear prediction model and a tunneling rate prediction model are constructed respectively to predict the cutter wear amount and the tunneling rate under different combinations of tunneling parameters; With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, a collaborative optimization model is constructed to obtain the Pareto optimal solution set. Based on the Pareto optimal solution set, the optimal execution plan is determined according to the service status of the hob, and the corresponding control system is used to perform the control. Obtain the status information of the roller cutter after adjustment, and revise the adjustment scheme for the next control cycle.

2. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on cutter service status according to claim 1, characterized in that, The TBM tunneling parameters include: total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth. The vibration information includes: vibration acceleration signal; The wear-resistant agent injection parameters include: wear-resistant agent injection flow rate.

3. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on cutter service status according to claim 1, characterized in that, Obtaining TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters includes: The acquired TBM tunneling parameters, cutter temperature and vibration information, and wear-resistant agent injection parameters are synchronized in time, outlier removal, filtered, and normalized. The data is then divided according to a set time window to ensure that the TBM tunneling parameters, cutter status information, and wear-resistant agent injection parameters within the same time window correspond to each other, forming a training dataset.

4. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head as described in claim 1, characterized in that, Determining the service status of the hob includes: The hob temperature and vibration characteristics are extracted from the real-time monitored hob temperature and vibration information; wherein, the hob temperature characteristics include: hob temperature and temperature rise rate, and the vibration characteristics include: root mean square value of vibration, peak value and periodicity characteristics; The service status of the cutter is determined by comparing the temperature and vibration characteristics of the cutter with preset corresponding discrimination thresholds. The service status of the cutter refers to the operating status of the cutter during the tunneling process, including normal rolling status, abnormal temperature rise status, abnormal rolling status, and uneven wear status.

5. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head as described in claim 1, characterized in that, Constructing the hob wear prediction model includes: Using TBM tunneling parameters and wear-resistant agent injection parameters as input features, the actual wear amount of the cutterhead obtained from the opening inspection as the training label, and the cutterhead wear evolution law as a physical constraint, a physical information neural network is trained to construct a cutterhead wear prediction model and output the predicted cutterhead wear amount. The cutterhead wear evolution law includes constraints based on Arcard's wear law, wear monotonicity constraints, and physical boundary constraints. The TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate, and penetration depth.

6. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head as described in claim 2, characterized in that, Constructing the tunneling rate prediction model includes: The total thrust, cutterhead rotation speed, cutterhead torque, and penetration depth are used as model inputs, and the actual tunneling rate under the corresponding working conditions is used as the training label to train a BP neural network and construct a tunneling rate prediction model, which is used to output the predicted tunneling rate under different combinations of tunneling parameters.

7. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on cutter service status according to claim 1, characterized in that, Obtaining the Pareto optimal solution set includes: Using the TBM tunneling parameters and wear-resistant agent injection parameters as decision variables, and the cutter wear prediction model and tunneling rate prediction model as evaluation models, the cutter wear amount and tunneling rate corresponding to different parameter combinations are calculated. With the optimization objectives of maximizing tunneling rate, minimizing cutter wear, and minimizing wear-resistant agent usage, the NSGA-II multi-objective optimization algorithm was used to solve the Pareto optimal solution set composed of different total thrust, cutterhead rotation speed, and wear-resistant agent injection flow rate.

8. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on cutter service status according to claim 1, characterized in that, Based on the Pareto optimal solution set, and according to the hob's service status, the optimal execution solution is determined as follows: Based on the current service status of the hob, select the corresponding execution plan from the Pareto optimal solution set; When the cutter head is in normal rolling condition, the scheme with predicted tunneling rate higher than preset rate threshold and predicted cutter head wear meeting the constraint conditions is selected; when the cutter head experiences abnormal temperature rise, the scheme with predicted cutter head wear lower than preset wear threshold and wear-resistant agent injection flow rate higher than normal working condition setting value is selected; when the cutter head experiences abnormal rolling or uneven wear, the scheme with total thrust and cutterhead speed lower than current working condition setting value and wear-resistant agent injection flow rate higher than normal working condition setting value is selected. After determining the execution plan, adjust the total thrust of the TBM, the cutter head speed, and the wear-resistant agent injection flow rate according to the execution plan.

9. The method for synergistic control of TBM tunneling parameters and wear-resistant agent based on the service status of the cutter head as described in claim 1, characterized in that, Obtaining the adjusted cutter status information and revising the adjustment scheme for the next control cycle includes: Obtain the cutterhead status information after adjustment; compare the service status before and after adjustment; when the cutterhead status improves, maintain the current adjustment parameters or enter the next control cycle; when the cutterhead status does not improve or deteriorates further, re-perform cutterhead wear prediction, tunneling rate prediction and multi-objective optimization with the updated cutterhead status to obtain the total thrust, cutterhead speed and wear-resistant agent injection flow rate for the next control cycle.

10. A device for coordinated control of TBM tunneling parameters and wear-resistant agent based on cutterhead service status, characterized in that, The apparatus for carrying out the method as described in any one of claims 1-9 comprises: The data acquisition module is used to acquire TBM tunneling parameters, cutterhead temperature and vibration information, and wear-resistant agent injection parameters, and to perform data preprocessing to form a training dataset; wherein, the TBM tunneling parameters include total thrust, cutterhead rotation speed, cutterhead torque, tunneling rate and penetration depth, and the wear-resistant agent injection parameters include wear-resistant agent injection flow rate; The hob condition discrimination module is used to extract hob temperature features and vibration features based on the hob temperature and vibration information, and compare them with preset discrimination thresholds to determine the hob service status. The parameter prediction module is used to combine the TBM tunneling parameters and wear-resistant agent injection parameters, and to construct a cutter wear prediction model and a tunneling rate prediction model using a physical information neural network and a BP neural network, respectively, to predict the cutter wear amount and the tunneling rate under different combinations of tunneling parameters; The collaborative optimization module uses total thrust, cutterhead speed and wear-resistant agent injection flow rate as decision variables, and takes the maximum tunneling rate, minimum cutter wear and minimum wear-resistant agent usage as optimization objectives. It uses the NSGA-II multi-objective optimization algorithm to solve the problem and obtain a Pareto optimal solution set composed of different total thrust, cutterhead speed and wear-resistant agent injection flow rate. The control and decision module is used to determine the optimal execution plan based on the Pareto optimal solution set and the service status of the hob, that is, to determine the total thrust, cutter head speed and wear-resistant agent injection flow rate, and to execute the control using the corresponding control system. The feedback correction module is used to obtain the status information of the hobbing cutter after adjustment and correct the adjustment scheme for the next control cycle.