A method and system for real-time prediction and intelligent control of cutterhead torque in slurry balance shield tunneling machines.

By using multi-source data acquisition and MPC-PID dual-mode intelligent control, the problems of low torque prediction accuracy and control misjudgment in slurry balance shield tunneling machines have been solved, achieving accurate torque prediction and stable control, thus improving construction efficiency and safety.

CN123109684APending Publication Date: 2026-11-17CHINA RAILWAY 18TH BUREAU GRP CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610971095.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-01
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, slurry balance shield machines neglect the fluid-solid coupling effect of slurry in torque control, resulting in low torque prediction accuracy, high control misjudgment rate, and large torque fluctuations. They also cannot effectively distinguish between the increase in torque caused by hardening of the strata and thickening of the slurry.

Method used

By acquiring multi-source heterogeneous data in real time, a multi-factor coupled cutterhead torque mechanism model is constructed, the coupling weight of mud viscosity and cutting conditions is quantified, and a hierarchical control is performed using an MPC-PID dual-mode intelligent controller to achieve real-time accurate prediction and fluctuation smoothing of torque.

Benefits of technology

It achieves precise torque traceability, reduces torque fluctuation by 35%, improves construction efficiency by 15%, is suitable for construction in complex strata, and avoids the risk of machine jamming and equipment damage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109684A_ABST
    Figure CN123109684A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for real-time prediction and intelligent control of cutterhead torque in slurry-balanced shield tunneling machines, relating to the field of shield tunneling technology. This invention decouples the total cutterhead torque into four independent physical components—cutting, viscosity, pressure difference, and friction—by synchronously collecting slurry physical parameters and propulsion condition parameters in real time. It quantifies the coupling weight between slurry viscosity and cutting conditions, predicts the torque change trajectory based on rolling time-domain optimization, and employs an MPC-PID dual-mode hierarchical control strategy to regulate the cutterhead rotation speed and slurry flow rate in a closed loop. This invention solves the problems of low torque prediction accuracy and high control misjudgment rate caused by neglecting the fluid-structure interaction effect in traditional methods. It can reduce torque fluctuation variance by 35%, increase tunneling efficiency by 15%, effectively avoid the risk of machine jamming, and is suitable for construction in complex strata such as rivers and water-rich sandy layers.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of tunnel boring machine (TBM) construction technology, specifically to a method and system for real-time prediction and intelligent control of cutterhead torque in a slurry balance TBM. Background Technology

[0002] Slurry balance shield tunneling machines are widely used in complex conditions such as crossing rivers, water-rich sand layers, and composite strata due to their excellent face balancing capabilities. The cutterhead torque is one of the most critical mechanical parameters reflecting the shield tunneling status, and its stability directly affects construction safety and efficiency.

[0003] In practical engineering, the cutterhead torque originates not only from the resistance of the cutter cutting the soil but is also significantly influenced by the high-pressure mud in the slurry chamber. Existing torque control methods typically only consider the effects of mechanical propulsion parameters (such as penetration depth and rotational speed), neglecting the "fluid-structure coupling" damping effect of the mud circulation system on the cutterhead torque. In fact, the physical properties of the mud (specific gravity, viscosity, yield stress) change dynamically in real time as the stratum soil is mixed in, and the viscous damping torque generated by high-viscosity mud accounts for a significant proportion of the total torque.

[0004] Without a quantitative analysis of the "mud-propulsion" coupling mechanism, control systems often cannot distinguish whether the increase in cutting torque is due to hardening of the formation or thickening of the mud, leading to misjudgments and incorrect control (such as blindly reducing the speed or misjudging it as machine jamming), resulting in low excavation efficiency or even equipment damage. Therefore, it is urgent to establish a real-time acquisition and synchronization mechanism for multi-source data on mud properties and propulsion conditions, construct a multi-component decoupled coupling torque mechanism model, quantify the coupling effect between mud and propulsion, and propose an intelligent control strategy that takes into account multi-variable constraints and high-frequency interference to achieve real-time and accurate prediction and fluctuation smoothing of cutterhead torque, thereby improving adaptability to complex working conditions. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. Therefore, one objective of this invention is to propose a method and system for real-time prediction and intelligent control of the cutterhead torque of a slurry-balanced tunnel boring machine, solving the problems of low torque prediction accuracy, high control misjudgment rate, and large torque fluctuations caused by neglecting the fluid-structure interaction effect of slurry in the prior art.

[0006] To achieve the above objectives, this invention provides a method for real-time prediction and intelligent control of the cutterhead torque of a slurry balance shield tunneling machine, comprising the following steps: S1. Real-time acquisition and synchronization of multi-source heterogeneous data: Construct a data acquisition system to collect the mud physical property parameter set and propulsion condition parameter set of the slurry balance shield machine in real time and synchronously. S2. Construct a multi-factor coupled cutter head torque mechanism model: Based on the principle of mechanical decomposition, the total cutter head torque is decoupled into multiple independent physical components, and a nonlinear mapping relationship is established between each physical component and the parameters collected in step S1. S3. Quantify the coupling weight between mud viscosity and cutting conditions: For the key coupling terms in the cutterhead torque mechanism model, establish corresponding analytical formulas to clarify the contribution of mud physical properties and propulsion conditions to each torque component. S4. Real-time torque prediction based on rolling time domain optimization: Using the state quantity of k at the current time, combined with the tool disc torque mechanism model constructed in step S2 and the coupling weight quantized in step S3, the trajectory of tool disc torque change in the next N sampling periods is predicted. S5. Construct an MPC-PID dual-mode intelligent controller and generate control commands: A hierarchical control strategy is adopted. The MPC main controller solves for the optimal control sequence based on the torque change trajectory predicted in step S4, with the objective function being to minimize the deviation between the predicted torque and the set safety threshold. The PID secondary controller fine-tunes the control quantity output by the MPC main controller to compensate for model errors and high-frequency interference. S6. Execution Control and Feedback Correction: The optimal control command generated in step S5 is sent to the PLC actuator to adjust the corresponding operating parameters of the tunnel boring machine, and the actual torque response is fed back to the system to form a closed-loop control.

[0007] Preferably, the set of mud physical property parameters in step S1 includes mud specific gravity, plastic viscosity, yield stress, and slurry inlet and outlet flow rates; the set of propulsion condition parameters includes cutterhead rotation speed, propulsion speed v, penetration depth P, and total thrust F; and the total cutterhead torque in step S2 is decoupled to the torque of the cutter cutting the soil. Mud viscosity damping torque Static pressure difference damping torque and mechanical friction torque Four independent physical components.

[0008] Preferably, the key coupling term in step S3 includes the cutting torque of the tool cutting the soil. Mud viscosity damping torque ; For cutting torque Introducing the formation strength coefficient The model is represented as: In the formula, D is the diameter of the cutter head, and P is the penetration depth. The softening coefficient; For mud viscosity damping torque Based on the rotational rheological theory of non-Newtonian fluids, the model is expressed as: In the formula, Where r is the fluid drag coefficient and r is the radial coordinate of the cutter head. For mud yield stress, The plastic viscosity of the mud. This is the equivalent clearance between the cutterhead and the earth chamber partition.

[0009] Preferably, the rolling time-domain optimization in step S4 specifically involves: starting from the current time k, updating the state variable once in each sampling period, and re-predicting the torque change trajectory for the next N sampling periods, where N is the preset prediction time-domain length, with a value range of 5 to 20; the constraints of the MPC main controller in step S5 include the upper and lower limits of the cutterhead speed, the upper and lower limits of the mud flow rate, and the upper and lower limits of the propulsion speed, and the optimal control sequence includes the optimal cutterhead speed and the optimal mud flow rate.

[0010] Preferably, in step S5, the PID secondary controller takes the deviation between the actual torque and the predicted torque as input and outputs the finely adjusted control quantity, and its adjustment cycle is less than the optimization cycle of the MPC main controller; in step S6, adjusting the corresponding operating parameters of the tunnel boring machine specifically involves adjusting the frequency of the variable frequency motor to change the cutterhead speed and adjusting the speed of the mud pump to change the slurry flow rate. When the predicted torque will exceed the set safety threshold, a coordinated control strategy combining a smooth reduction of the cutterhead speed and an increase in the slurry pump flow rate is preferred.

[0011] A real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine includes: The multi-source heterogeneous data acquisition module is used to synchronously acquire the mud physical property parameter set and propulsion condition parameter set of the slurry balance shield machine in real time; The multi-component torque mechanism model construction module is connected to the multi-source heterogeneous data acquisition module. It is used to decouple the total torque of the cutter head into multiple independent physical components based on the principle of mechanical decomposition, and to establish a nonlinear mapping relationship between each physical component and the acquired parameters. The coupling weight quantization module is connected to the multi-component torque mechanism model construction module. It is used to establish corresponding analytical formulas for key coupling terms in the cutterhead torque mechanism model, and to clarify the contribution of mud physical properties and propulsion conditions to each torque component. The rolling time-domain prediction module is connected to the multi-source heterogeneous data acquisition module, the multi-component torque mechanism model construction module, and the coupling weight quantization module, respectively. It is used to predict the trajectory of the tool disk torque change over the next N sampling periods by using the state quantity at the current time k, combined with the tool disk torque mechanism model and the quantized coupling weight. The MPC-PID dual-mode intelligent controller, connected to the rolling time-domain prediction module, employs a hierarchical control strategy. The MPC main controller solves for the optimal control sequence based on the predicted torque change trajectory, with the objective function being to minimize the deviation between the predicted torque and the set safety threshold. The PID secondary controller fine-tunes the control output of the MPC main controller to compensate for model errors and high-frequency interference. The execution and feedback correction module, connected to the MPC-PID dual-mode intelligent controller, is used to send the generated optimal control commands to the PLC actuator, adjust the corresponding operating parameters of the tunnel boring machine, and feed back the actual torque response to the system to form a closed-loop control.

[0012] Preferably, the multi-source heterogeneous data acquisition module includes: An online viscometer, installed in the mud circulation feed pipeline, is used to measure the plastic viscosity and yield stress of the mud in real time. Coriolis mass flow meters are installed in the slurry inlet and outlet pipelines respectively to measure the specific gravity and inlet / outlet flow rate of the slurry in real time. The tunnel boring machine's PLC system is used to collect data in real time, including cutterhead rotation speed, propulsion speed v, penetration depth P, and total thrust F.

[0013] Preferably, the multi-component torque mechanism model construction module is specifically used to decouple the total torque of the cutterhead into the torque of the cutter cutting the soil. Mud viscosity damping torque Static pressure difference damping torque and mechanical friction torque Four independent physical components; the coupled weighted quantization module is specifically used to measure the torque of the tool cutting the soil. A model was established by introducing formation strength coefficient and softening coefficient, and a model was developed to address the viscous damping torque of mud. A model is established based on the rotational rheological theory of non-Newtonian fluids.

[0014] Preferably, the MPC-PID dual-mode intelligent controller includes: The MPC main controller is used to handle problems with large time delay and multiple variables. It takes minimizing the deviation between the predicted torque and the set safety threshold as the objective function and solves the optimal control sequence that includes the optimal cutterhead speed and the optimal mud flow rate. The PID secondary controller, connected to the MPC main controller, is used to handle high-frequency interference, fine-tune the control output of the MPC main controller, and quickly compensate for model errors.

[0015] Preferably, the execution and feedback correction module includes: The PLC actuator is used to receive control commands and adjust the frequency of the variable frequency motor and the speed of the mud pump. A torque sensor, mounted on the cutter head drive shaft, is used to measure the actual cutter head torque value in real time; The feedback loop is used to feed back the actual measured cutter head torque value to the rolling time domain prediction module and the MPC-PID dual-mode intelligent controller to achieve closed-loop control.

[0016] The advantages of this invention compared to the prior art are: (1) Quantification of coupling mechanism to achieve accurate torque source tracing: By modeling the multi-component mechanism, the contribution of physical parameters such as mud viscosity to torque is clarified, solving the problem that the traditional "black box" model cannot explain the torque composition. It can accurately distinguish the increase in cutting torque caused by formation hardening and the increase in damping torque caused by mud thickening, avoiding misjudgment of control.

[0017] (2) Dual-mode control strategy, balancing accuracy and response speed: The MPC main controller is good at handling multivariable coupling and large lag problems, and can predict torque trends in advance and perform forward-looking regulation; the PID secondary controller quickly responds to high-frequency interference and performs real-time fine-tuning of the control quantity. The dual collaboration reduces the torque fluctuation variance by 35% and makes the regulation more stable.

[0018] (3) Adaptable to complex working conditions and improves construction efficiency: It is suitable for complex strata such as rivers and water-rich sand layers, which can effectively avoid the risk of machine jamming and equipment damage, and improve tunneling efficiency by 15%. Moreover, it does not require large-scale modification of existing shield machines, but only requires the addition of a small number of sensors and control software, which is convenient to integrate and has good engineering application value. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is the overall control logic flowchart of the present invention; Figure 2 This is a schematic diagram of the coupling mechanism of multiple physical components of the cutter head torque in this invention; Figure 3 This is a comparison chart of real-time torque prediction value and MPC control effect in one embodiment of the present invention. Detailed Implementation

[0021] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0022] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0023] The present invention will now be described in further detail with reference to the accompanying drawings.

[0024] like Figure 1 As shown, this invention discloses a method and system for real-time prediction and intelligent control of the cutterhead torque of a slurry balance shield tunneling machine, wherein the method specifically includes the following steps: S1. Real-time acquisition and synchronization of multi-source heterogeneous data A data acquisition system was constructed, with an online viscometer installed in the mud circulation inlet pipeline to measure the plastic viscosity and yield stress of the mud in real time; Coriolis mass flow meters were installed in both the inlet and outlet pipelines to measure the specific gravity and inlet / outlet flow rates of the mud in real time. The tunnel boring machine's built-in PLC system collects real-time data on cutterhead rotation speed, propulsion speed v, penetration depth P, and total thrust F. All data is synchronized using a unified timestamp to ensure time consistency.

[0025] S2. Construct a multi-factor coupled cutter head torque mechanism model. Based on the principle of mechanical decomposition, the total torque of the cutter head is... The decoupling is divided into four independent physical components, namely: in, The torque for the cutting tool to cut the soil. For mud viscosity damping torque, The static pressure difference damping torque, This refers to the mechanical friction torque.

[0026] A nonlinear mapping relationship was established between each physical component and the collected mud physical properties and propulsion condition parameters. Among these, the static pressure difference damping torque... It is mainly related to the pressure difference inside and outside the mud and water chamber, and the mechanical friction torque. It is mainly related to the mechanical characteristics of the tunnel boring machine itself, and can be obtained through on-site calibration.

[0027] S3, Quantifying the coupling weight between mud viscosity and cutting conditions For the cutting torque of the tool on the soil, which has the greatest impact on the total torque and the most significant coupling effect, Mud viscosity damping torque Establish specific analytical formulas: (1) Torque of the cutting tool in soil Introducing the formation strength coefficient and softening coefficient Considering the softening effect of mud on soil, the model is represented as follows: In the formula, D is the diameter of the cutter head, and P is the penetration depth. The cutterhead rotation speed. Formation strength coefficient. The softening coefficient can be obtained from geological survey data. It can be determined through on-site testing.

[0028] (2) For mud viscous damping torque Based on the rotational rheology theory of non-Newtonian fluids, considering the shear flow between the cutterhead and the soil chamber baffle, the model is expressed as follows: In the formula, Where r is the fluid drag coefficient and r is the radial coordinate of the cutter head. The yield stress of the mud is... The plastic viscosity of the mud. This represents the equivalent clearance between the cutterhead and the soil chamber partition. Fluid drag coefficient. The equivalent gap h can be obtained through on-site calibration.

[0029] S4. Real-time torque prediction based on rolling time-domain optimization Starting from the current time k, using all state variables collected at the current time, combined with the toolhead torque mechanism model constructed in step S2 and the coupling weights quantized in step S3, the toolhead torque change trajectory for the next N sampling periods is predicted. N is the preset prediction time domain length, ranging from 5 to 20. The state variables are updated once in each sampling period, and the prediction is performed again to achieve rolling real-time prediction of torque.

[0030] S5. Construct an MPC-PID dual-mode intelligent controller and generate control commands. A hierarchical control strategy is adopted to construct an MPC-PID dual-mode intelligent controller: (1) MPC main controller: handles large time delay and multivariate constraint problems. The objective function is to minimize the deviation between the predicted torque and the set safety threshold. At the same time, the upper and lower limits of the cutterhead speed, the upper and lower limits of the mud flow rate and the upper and lower limits of the propulsion speed are considered. The optimal control sequence, including the optimal cutterhead speed and the optimal mud flow rate, is solved by rolling optimization.

[0031] (2) PID secondary controller: handles high-frequency interference. Taking the deviation between the actual torque and the predicted torque as input, it outputs a finely adjusted control quantity to finely adjust the control quantity output by the MPC main controller in real time, quickly compensating for model errors and high-frequency interference. The adjustment cycle of the PID secondary controller is shorter than the optimization cycle of the MPC main controller, ensuring the system's fast response.

[0032] S6. Execution Control and Feedback Correction The optimal control commands generated by the MPC-PID dual-mode intelligent controller are sent to the PLC actuator to adjust the frequency of the variable frequency motor to change the cutterhead speed and the speed of the mud pump to change the slurry inlet and outlet flow rates. Simultaneously, a torque sensor installed on the cutterhead drive shaft measures the actual cutterhead torque value in real time and feeds the actual torque response back to the rolling time-domain prediction module and the MPC-PID dual-mode intelligent controller, forming a closed-loop control that continuously corrects model parameters and control commands, thereby improving control accuracy.

[0033] When the predicted torque is about to exceed the set safety threshold, a coordinated control strategy combining a smooth reduction in cutterhead speed with an increase in slurry pump flow rate is prioritized. Smoothly reducing the cutterhead speed can simultaneously reduce the viscous damping induced by cutting torque and shear rate; increasing the slurry pump flow rate can accelerate the discharge of high-viscosity slurry, reduce the slurry density in the chamber, thereby effectively reducing the total torque, achieving "soft landing" control, and avoiding triggering an emergency shutdown.

[0034] The following example, using the construction of a large-diameter slurry balance shield tunnel in a cross-river tunnel project, further illustrates the specific implementation of this invention: The project utilizes a 14.5m diameter slurry shield tunneling machine, traversing strata primarily composed of silty clay. The safety alarm threshold for cutterhead torque is set at 4500 kNm.

[0035] S1, Data Acquisition An online viscometer was installed in the mud circulation inlet pipeline, and the viscosity of the mud entering the storage chamber was measured in real time to be 35 Pas, and the yield stress was 12 Pa. Coriolis mass flow meters were installed in both the inlet and outlet pipelines, and the specific gravity of the mud was measured in real time to be 1.25 g / cm³. 3 The feed flow rate is 800 m³ / s. 3 / h, slurry discharge flow rate is 820m³ / h 3 / h; Data transmitted back through the tunnel boring machine's PLC system: cutterhead rotation speed is 1.5 rpm, penetration depth is 20 mm / r, propulsion speed is 30 mm / min, and total thrust is 120000 kN.

[0036] S2 & S3, Model Calculation The system calculates the torque components according to the above formula: (1) Cutting torque: The formation is silty clay, and the formation strength coefficient is set. =0.8MPa, softening coefficient =0.7, substituting into the formula, we get =2800 kNm; (2) Viscous damping torque: Substituting into the integral formula, we get =1200kNm; (3) Static pressure difference damping torque =300kNm, mechanical friction torque =200kNm, all of which are field calibration values; Total torque =2800+1200+300+200=4500kNm, which has reached the alarm threshold.

[0037] S4, Trend Forecasting The predictive model shows that if the current advance speed is maintained, as the machine enters the clay layer, the mud viscosity will further increase to 45 Pas. It is estimated that the total torque will exceed the alarm threshold of 4500 kNm in 30 seconds, reaching 5200 kNm, which may trigger an emergency shutdown.

[0038] S5, MPC Intelligent Decision Making The controller constructs the objective function: The constraints include: mud flow rate .

[0039] After rolling optimization, the system outputs the optimal control strategy: (1) Active speed reduction: The cutting head speed is smoothly reduced from 1.5 rpm to 1.2 rpm to reduce the viscous damping guided by cutting torque and shear rate; (2) Coordinated slurry cleaning: Instructions to increase the flow rate of the slurry pump by 10%, from 820m³ / h. 3 / h increased to 900m 3 / h, accelerates the discharge of high-viscosity mud and reduces the density of mud in the bin.

[0040] S6. Implementation Results After the command was issued, the measured torque curve was as follows: Figure 3As shown, after rising to 4600 kNm, the torque smoothly dropped back to below 4000 kNm without triggering an emergency shutdown, achieving a "soft landing" control. Compared to traditional manual operation, the torque fluctuation variance was reduced by 35%, and the tunneling efficiency was improved by 15%.

[0041] This invention also discloses a real-time prediction and intelligent control system for the cutterhead torque of a slurry-propulsion shield tunneling machine that considers the coupling of slurry and propulsion conditions. This system is used to implement the aforementioned method and includes a multi-source heterogeneous data acquisition module, a multi-component torque mechanism model construction module, a coupling weight quantization module, a rolling time-domain prediction module, an MPC-PID dual-mode intelligent controller, and an execution and feedback correction module. The functions and connections of each module have been described in detail in the invention description section and will not be repeated here.

[0042] Finally, any aspects not fully described in this invention utilize existing mature products and technologies.

[0043] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for real-time prediction and intelligent control of cutterhead torque in a slurry-balanced shield tunneling machine, characterized in that, Includes the following steps: S1. Real-time acquisition and synchronization of multi-source heterogeneous data: Construct a data acquisition system to collect the mud physical property parameter set and propulsion condition parameter set of the slurry balance shield machine in real time and synchronously. S2. Construct a multi-factor coupled cutter head torque mechanism model: Based on the principle of mechanical decomposition, the total cutter head torque is decoupled into multiple independent physical components, and a nonlinear mapping relationship is established between each physical component and the parameters collected in step S1. S3. Quantify the coupling weight between mud viscosity and cutting conditions: For the key coupling terms in the cutterhead torque mechanism model, establish corresponding analytical formulas to clarify the contribution of mud physical properties and propulsion conditions to each torque component. S4. Real-time torque prediction based on rolling time domain optimization: Using the state quantity of k at the current time, combined with the tool disc torque mechanism model constructed in step S2 and the coupling weight quantized in step S3, the trajectory of tool disc torque change in the next N sampling periods is predicted. S5. Construct an MPC-PID dual-mode intelligent controller and generate control commands: A hierarchical control strategy is adopted. The MPC main controller solves for the optimal control sequence based on the torque change trajectory predicted in step S4, with the objective function being to minimize the deviation between the predicted torque and the set safety threshold. The PID secondary controller fine-tunes the control quantity output by the MPC main controller to compensate for model errors and high-frequency interference. S6. Execution Control and Feedback Correction: The optimal control command generated in step S5 is sent to the PLC actuator to adjust the corresponding operating parameters of the tunnel boring machine, and the actual torque response is fed back to the system to form a closed-loop control.

2. The method for real-time prediction and intelligent control of cutterhead torque in slurry balance shield tunneling machines according to claim 1, characterized in that: The set of mud physical property parameters mentioned in step S1 includes mud specific gravity, plastic viscosity, yield stress, and mud inlet and outlet flow rates; the set of propulsion condition parameters includes cutterhead rotation speed, propulsion speed v, penetration depth P, and total thrust F; the total cutterhead torque mentioned in step S2 is decoupled as the torque of the cutter cutting the soil. Mud viscosity damping torque Static pressure difference damping torque and mechanical friction torque Four independent physical components.

3. The method for real-time prediction and intelligent control of slurry balance shield tunnel cutterhead torque according to claim 2, characterized in that: The key coupling term mentioned in step S3 includes the cutting torque of the tool cutting the soil. Mud viscosity damping torque ; For cutting torque Introducing the formation strength coefficient The model is represented as: In the formula, D is the diameter of the cutter head, and P is the penetration depth. The softening coefficient; For mud viscosity damping torque Based on the rotational rheological theory of non-Newtonian fluids, the model is expressed as: In the formula, Where r is the fluid drag coefficient and r is the radial coordinate of the cutter head. For mud yield stress, The plastic viscosity of the mud. This is the equivalent clearance between the cutterhead and the earth chamber partition.

4. The method for real-time prediction and intelligent control of cutterhead torque in slurry balance shield tunneling machines according to claim 1, characterized in that: The rolling time-domain optimization in step S4 specifically involves updating the state variable once in each sampling period, starting from the current time k, and re-predicting the torque change trajectory for the next N sampling periods, where N is the preset prediction time-domain length, with a value range of 5 to 20. The constraints of the MPC main controller in step S5 include the upper and lower limits of the cutterhead speed, the upper and lower limits of the mud flow rate, and the upper and lower limits of the propulsion speed. The optimal control sequence includes the optimal cutterhead speed and the optimal mud flow rate.

5. The method for real-time prediction and intelligent control of cutterhead torque in slurry balance shield tunneling machines according to claim 1, characterized in that: The PID secondary controller mentioned in step S5 takes the deviation between the actual torque and the predicted torque as input and outputs the fine-tuned control quantity. Its adjustment cycle is less than the optimization cycle of the MPC main controller. The adjustment of the corresponding operating parameters of the tunnel boring machine mentioned in step S6 specifically involves adjusting the frequency of the variable frequency motor to change the cutterhead speed and adjusting the speed of the mud pump to change the slurry flow rate. When the predicted torque will exceed the set safety threshold, a coordinated control strategy combining a smooth reduction of the cutterhead speed and an increase in the slurry pump flow rate is preferred.

6. A real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine, characterized in that, include: The multi-source heterogeneous data acquisition module is used to synchronously acquire the mud physical property parameter set and propulsion condition parameter set of the slurry balance shield machine in real time; A multi-component torque mechanism model construction module is connected to the multi-source heterogeneous data acquisition module. It is used to decouple the total torque of the cutter head into multiple independent physical components based on the principle of mechanical decomposition, and to establish a nonlinear mapping relationship between each physical component and the acquired parameters. The coupling weight quantization module is connected to the multi-component torque mechanism model construction module. It is used to establish corresponding analytical formulas for key coupling terms in the cutterhead torque mechanism model, and to clarify the contribution of mud physical property parameters and propulsion condition parameters to each torque component. The rolling time-domain prediction module is connected to the multi-source heterogeneous data acquisition module, the multi-component torque mechanism model construction module, and the coupling weight quantization module, respectively. It is used to predict the trajectory of the tool disk torque change over the next N sampling periods by using the state quantity at the current time k, combined with the tool disk torque mechanism model and the quantized coupling weight. The MPC-PID dual-mode intelligent controller, connected to the rolling time-domain prediction module, is used to employ a hierarchical control strategy. The MPC main controller solves for the optimal control sequence based on the predicted torque change trajectory, with the objective function being to minimize the deviation between the predicted torque and the set safety threshold. The PID secondary controller fine-tunes the control output of the MPC main controller to compensate for model errors and high-frequency interference. The execution and feedback correction module is connected to the MPC-PID dual-mode intelligent controller. It is used to send the generated optimal control commands to the PLC actuator, adjust the corresponding operating parameters of the tunnel boring machine, and feed back the actual torque response to the system to form a closed-loop control.

7. The real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine according to claim 6, characterized in that: The multi-source heterogeneous data acquisition module includes: An online viscometer, installed in the mud circulation feed pipeline, is used to measure the plastic viscosity and yield stress of the mud in real time. Coriolis mass flow meters are installed in the slurry inlet and outlet pipelines respectively to measure the specific gravity and inlet / outlet flow rate of the slurry in real time. The tunnel boring machine's PLC system is used to collect data in real time, including cutterhead rotation speed, propulsion speed v, penetration depth P, and total thrust F.

8. The real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine according to claim 6, characterized in that: The multi-component torque mechanism model construction module is specifically used to decouple the total torque of the cutterhead into the torque of the cutter cutting the soil. Mud viscosity damping torque Static pressure difference damping torque and mechanical friction torque Four independent physical components; the coupled weight quantization module is specifically used to measure the torque of the tool cutting the soil. A model was established by introducing formation strength coefficient and softening coefficient, and a model was developed to address the viscous damping torque of mud. A model is established based on the rotational rheological theory of non-Newtonian fluids.

9. The real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine according to claim 6, characterized in that: The MPC-PID dual-mode intelligent controller includes: The MPC main controller is used to handle problems with large time delay and multiple variables. It takes minimizing the deviation between the predicted torque and the set safety threshold as the objective function and solves the optimal control sequence that includes the optimal cutterhead speed and the optimal mud flow rate. The PID secondary controller, connected to the MPC main controller, is used to handle high-frequency interference, fine-tune the control quantity output by the MPC main controller, and quickly compensate for model errors.

10. The real-time prediction and intelligent control system for the cutterhead torque of a slurry balance shield tunneling machine according to claim 6, characterized in that: The execution and feedback correction module includes: The PLC actuator is used to receive control commands and adjust the frequency of the variable frequency motor and the speed of the mud pump. A torque sensor, mounted on the cutter head drive shaft, is used to measure the actual cutter head torque value in real time; The feedback loop is used to feed back the actual measured cutter head torque value to the rolling time domain prediction module and the MPC-PID dual-mode intelligent controller to achieve closed-loop control.