A shield intelligent steering system and control method based on PID control

Abstract

The application discloses a kind of based on PID control's shield intelligent direction adjusting system and control method.System's shield pose measurement module gathers shield pose data and exports shield pose deviation value in combination with preset tunnel design axis, data acquisition module gathers and stores shield pose deviation value, PID controller receives shield pose deviation value and exports propulsion cylinder pressure increment control signal, and oil pressure pose module receives control signal and controls shield machine to execute action.The method comprises: simulation system designs PID controller, to build intelligent direction adjusting system;System real-time acquisition shield pose data and export control signal to control the propulsion cylinder of shield machine to execute action to continue tunnel excavation work.The system of the application uses PID controller to carry out shield direction adjusting operation, effectively solve a series of problems, such as misoperation, over-adjustment and adjustment not in time, etc., when manually operating shield machine direction adjusting, greatly improve the intelligent degree of shield.

Description

Technical Field

[0001] This invention relates to an intelligent tunnel boring machine (TBM) steering system, specifically to an intelligent TBM steering system and control method based on PID control. Background Technology

[0002] A tunnel boring machine (TBM) is a highly efficient, environmentally friendly, and safe large-scale tunnel excavation device. During construction, the TBM needs to excavate along the tunnel's design axis. Excessive deviation between the actual excavation axis and the design axis will directly lead to the tunnel's shape and dimensions deviating from design requirements, reducing construction quality, increasing safety risks, and raising construction costs. Therefore, adjusting the TBM's excavation direction according to the design axis, so that the actual excavation axis matches the design axis as closely as possible, is crucial to ensuring tunnel construction quality.

[0003] Traditionally, adjusting the tunneling direction of a tunnel boring machine (TBM) relies on manual operation by the operator. Based on the deviation between the TBM and the designed axis measured by the TBM's guidance system, the operator adjusts the pressure of different zones of the propulsion cylinders to change the TBM's tunneling direction. This process constitutes a human-in-the-loop (HIL) closed-loop control system, where the operator primarily analyzes deviations, makes decisions, and outputs control signals. Therefore, traditional TBM attitude control heavily depends on the operator's experience and intuition. However, due to the complex human-machine-environment interaction, the TBM's enormous inertia, and the numerous parameters monitored and operated by the operator, manual attitude control suffers from low accuracy, poor real-time performance, and lag, leading to prominent issues like serpentine tunneling in actual engineering projects. Furthermore, prolonged, high-intensity control work in high-noise construction environments easily causes operator fatigue and inattention, further exacerbating the difficulty of manual direction adjustment. The long training period and high cost of TBM operators, coupled with the shortage of skilled operators, have become a major challenge for the industry. In response to the shortcomings of traditional manual tunnel boring machine (TBM) steering, the development of automated and intelligent TBM steering technology is an industry trend and research hotspot, and a great deal of research and development work has been carried out both domestically and internationally.

[0004] Currently, automated and intelligent shield tunneling alignment technologies are mainly divided into three categories. The first category is based on logic rules. This type of method establishes logical rules based on human experience, and then converts shield posture deviation information into zoned hydraulic cylinder control commands through these rules. This is equivalent to automating manual control using logic programming. Patent CN1800583 utilizes a total station equipped with a spatial vector method calculation module to automatically acquire the shield posture. Based on the selected correction mode, it uses automatic shield posture correction software to calculate the grouping position of the shield cut center under that posture. Finally, the calculation results and control commands are transmitted to the shield propulsion system's control module via a PLC (Programmable Logic Controller) module for corresponding adjustments, thereby achieving continuous automatic correction of the shield posture. This method reduces the number of operators, and the deviation fluctuation is lower than manual correction. Patent CN102518446A utilizes a laser total station to measure the real-time position and orientation information of the front, articulated, and rear sections of a tunnel boring machine (TBM). Based on this information, it calculates the vertical and horizontal deviations of the front, articulated, and rear sections relative to the tunnel's design axis, as well as the relative deviations and changes in relative deviation of the entire TBM in the vertical and horizontal planes. It also calculates the relative vertical and horizontal deflection angles of the front, rear, and overall TBM sections, and sets multiple preset thresholds, determining whether these deviations exceed these thresholds. When a threshold is exceeded, the controller uses pressure sensors to measure the thrust of the jacks in each section of the TBM or the main thrust of the TBM to determine the correction method and the position of the jacks performing the correction. Based on empirical formulas, it determines the pressure value of the jacks performing the correction and controls their extension and retraction to adjust the deflection attitude of the front and rear sections of the TBM, ultimately controlling the TBM's tunneling path. Patent CN102606165A utilizes an industrial control computer to receive real-time shield machine posture data transmitted from a laser automatic posture measurement system. This data is compared with the designed shield machine posture data to determine whether a correction control signal should be sent to the controller. Once the correction control signal is sent, the controller sends a drive signal to the drive mechanism, thereby controlling the propulsion cylinder and the central deflection hydraulic cylinder to perform corresponding actions, adjusting the shield machine posture and achieving automatic correction. The second type is based on machine learning methods. These methods train a model using a neural network to obtain a mapping from shield machine posture deviation to adjustment commands. Essentially, the neural network model uses information available during shield machine operation to predict posture deviations or control commands in advance, enabling shield machine regulation.Patent CN109779649A proposes a real-time shield tunneling axis correction system based on big data. This system performs preprocessing operations such as outlier handling and data fusion on data collected from the construction site, including shield machine cutterhead torque, cutterhead speed, propulsion speed, sectional jack stroke, total thrust, sectional hydraulic valve opening, sectional output hydraulic pressure value, horizontal / vertical deviation of the cut / tail, shield slope angle, excavation face earth pressure, tunnel design axis coordinate changes, tail clearance, and excavation face soil information. It then uses a Long Short-Term Memory (LSTM) self-learning model to train a trajectory strategy module. This module determines the target position the shield machine will adjust to in the next construction unit, while the hydraulic attitude module outputs the required sectional hydraulic cylinder pressure values ​​based on the target position, controlling the shield to perform the alignment operation. Furthermore, this method can optimize the trajectory strategy module based on new data generated at the work site, aiming to achieve real-time shield axis alignment based on big data. Chinese patent CN108868807A proposes an intelligent control method for shield tunneling deviation correction. This method first divides the shield propulsion cylinders into multiple zones and obtains the pressure distribution values ​​for each zone's propulsion cylinders based on the segment burial depth and historical shield construction data. Next, the controller extracts features from the real-time data of shield posture and tail clearance obtained by the automatic measurement system and calculates the deviation correction curve and shield propulsion distance using a custom-defined deviation correction curve equation and deviation correction distance formula. Then, based on the shield propulsion distance, combined with the burial depth and historical construction data, the controller uses a random forest algorithm to obtain the adjustment values ​​for the hydraulic pressure of each zone's propulsion cylinders and outputs the corresponding zone's hydraulic pressure to adjust the shield posture during propulsion. After the shield machine advances a certain distance, the shield posture extends to the next adjustment distance. The difference between this shield posture and the deviation correction curve is calculated to obtain the deviation correction evaluation quantity, which is added to the shield posture adjustment quantity. Furthermore, the output hydraulic pressure and posture changes during the shield posture adjustment process are stored as historical data, and this data is used to train a random forest prediction model every 10 rings of tunneling. Patent CN112922609A trains a Gate Recurrent Unit (GRU) model using collected tunnel boring machine (TBM) excavation and guidance sample data to obtain a TBM attitude prediction model. This attitude prediction model can process current excavation and guidance data in real time and obtain the zoned pressure value and average propulsion speed that the TBM should output under this state, thereby controlling the TBM's propulsion and realizing intelligent TBM excavation.Patent CN110195592A establishes a hybrid deep learning model, WCNN-LSTM (Wide Convolutional Neural Networks-Long Short Term Memory), for predicting tunnel boring machine (TBM) pose. This model takes the TBM tunneling parameters at time t+j or the time interval from t+1 to t+j as input and the TBM pose signal at time t+j or the time interval from t+1 to t+j as output. Therefore, it can predict the TBM pose value for the time interval from t+1 to t+j under the current input values ​​and compare it with the design values ​​to obtain the deviation. If the deviation exceeds the allowable range, the TBM operator can adjust the input values ​​applied to the TBM at time t+j or the time interval from t+1 to t+j in advance, achieving pre-emptive correction before the TBM tunneling goes astray. This method provides a decision-making reference for the TBM operator when performing correction operations, but manual operation is still required. Currently, a big data-driven shield tunneling machine (TBM) attitude control method has been proposed. This method has two sub-models: one for setting the TBM attitude control target and the other for reflecting the relationship between the TBM propulsion cylinder pressure and the TBM posture deviation. The two sub-models work together to generate a TBM attitude adjustment strategy, providing decision-making reference for the TBM operator when performing steering operations. The above technologies are all based on supervised learning methods. Furthermore, patent CN114019795A constructs a TBM simulation correction environment that maps TBM posture deviation to steering commands using a reinforcement learning framework, and establishes a TBM correction decision model. Through agent evaluation methods and value function network structures, the TBM correction decision model is trained multiple times in the simulation environment to obtain the final model. The final model outputs the propulsion cylinder pressure values ​​for each TBM section and the cutterhead steering decision scheme, providing the TBM operator with a reference and assisting in steering operations. The third category is based on fuzzy control methods. This type of method establishes fuzzy rules based on human experience, and uses these fuzzy rules to convert TBM posture deviation information into section cylinder control commands, achieving automatic TBM posture adjustment. In the current research on shield tunneling attitude control based on the fuzzy PID (Proportion Integration Differentiation) method, a dual closed-loop feedback fuzzy PID control strategy is used to adjust the speed of the shield propulsion hydraulic cylinder to achieve accurate control of the shield tunneling trajectory.Patent CN113931648A proposes an automatic control system for adjusting the attitude of a tunnel boring machine (TBM). This system consists of a host computer, a slave computer, a TBM attitude measurement system, a TBM propulsion system, and feedback sensors. The TBM attitude measurement system transmits the current attitude deviation of the TBM to the slave computer in real time. The host computer reads this data and key TBM tunneling parameters from the slave computer and calculates recommended pressure values ​​for the four propulsion cylinders of the TBM based on a fuzzy control algorithm. The slave computer then controls the TBM propulsion system to perform corresponding actions based on these recommended values, thereby achieving automatic orientation adjustment of the TBM. Patent CN106522973A proposes an automatic deviation correction system for a TBM. This system collects the current operating status of the TBM through a guide measurement device and sends it to a computing device. The control device uses IPC (Inter-Process Communication) software programming to convert the parameter values ​​from the current operating status of the TBM stored in the computing device into control commands according to fuzzy rules, thereby achieving automatic deviation correction of the TBM. However, since the fuzzy rules of fuzzy control need to be manually formulated, it is essentially a method based on human experience. Furthermore, patent CN113586075A proposes an automatic attitude correction system and method for a tunnel boring machine (TBM) axis relative to the tunnel axis. This system includes a deviation angle calculation unit and an angle correction control unit. The deviation angle calculation unit can collect and determine the horizontal and vertical deviation angles of the TBM axis relative to the tunnel axis in real time. The system's adaptive module can generate adjustment strategies for the TBM's various force components based on these deviation angles, adjusting the horizontal and vertical deviation angles. Then, the angle correction control unit controls the horizontal and vertical deflection of the TBM. Simultaneously, the system also includes a thrust correction unit and an advance control unit. The thrust correction unit can monitor the actual total thrust of the TBM and compare the actual horizontal and vertical components of the actual total thrust with the target components generated by the adaptive module to obtain the corresponding correction amount. Finally, the advance control unit corrects the current actual total thrust based on the current correction amount, controlling the TBM's advance. However, this method fails to consider the impact of different geological conditions.

[0005] In existing automatic shield tunneling (TBM) steering technologies, logic-based methods rely solely on the TBM's posture deviation information to calculate the pressure of the TBM's propulsion cylinders for steering / correction operations. This approach fails to fully utilize other valuable information generated during TBM operation, such as geological information. Furthermore, due to the complexity of geological conditions and actual control processes, logic-based TBM steering technologies suffer from unstable control and difficulty adapting to complex and changing geological conditions. Moreover, the formulation of logic rules still depends on human experience, requiring manual intervention. Fuzzy control-based methods also require manual formulation of fuzzy rules and cannot adapt to different geological conditions, exhibiting limited intelligence. Machine learning-based methods suffer from reliance on large amounts of historical data for training and poor generalization characteristics; they can only provide decision-making assistance for the TBM operator and cannot completely replace manual steering operations. Therefore, existing automatic TBM steering technologies cannot escape the series of problems inherent in manual operation. Summary of the Invention

[0006] To address the problems existing in the background technology, this invention provides a shield tunneling intelligent steering system and control method based on PID control. The PID-based intelligent steering system mainly consists of four parts: a shield posture measurement module, a data acquisition module, a PID intelligent steering controller, and a hydraulic posture module. It can realize three functions: real-time acquisition of shield posture data, control signal output, and shield machine posture change. The shield machine's posture data is mainly measured in real time by the shield posture measurement module and compared with the preset tunnel design axis. After calculation, the shield posture deviation is sent to the data acquisition module on the shield machine. The intelligent steering controller based on the PID control algorithm is simple to design and implement, has good robustness, and a wide range of applications. To optimize the PID-based intelligent steering controller, this invention also builds a shield intelligent steering simulation system, which mainly includes four parts: a shield-environment dynamic interaction simulation module, a signal conversion module, a tunneling parameter prediction module, and a PID controller. The signal conversion module has three functions: signal input, data storage, and signal transmission. The signal input function primarily receives the output signal from the PID controller. The data storage function saves the previous output signal from the signal conversion module. The signal transmission function accumulates the received PID controller output signal with the saved previous output signal and then sends it to the tunneling parameter prediction module and the shield-environment dynamic interactive simulation module. After receiving the parameters from the signal conversion module, the tunneling parameter prediction module outputs the corresponding predicted values ​​to the shield-environment dynamic interactive simulation module, primarily providing some input parameters for it. The shield-environment dynamic interactive simulation module is a shield rock machine interaction module built using a Long Short-Term Memory (LSTM) model. It primarily provides the control object for the signal conversion module during the design phase and provides input and a high-precision training simulation environment for the PID controller, optimizing the PID controller's design parameters and verifying its performance. The intelligent tunnel boring machine (TBM) steering method based on PID control works as follows: At a set time point, the intelligent TBM steering controller receives the current shield posture deviation measured by the shield posture measurement module from the data acquisition module of the TBM. Based on these deviations, it generates incremental pressure signals for the propulsion cylinders and outputs them to the hydraulic posture module. The hydraulic posture module adjusts the pressure of the corresponding propulsion cylinders according to the control signals, thereby controlling the movement of the corresponding propulsion cylinders to change the TBM posture. After the shield posture changes, the intelligent TBM steering controller receives updated data from the shield posture measurement module and the data acquisition module again at the next set time point, and then outputs the optimal incremental pressure signals for the propulsion cylinders of each shield section to the hydraulic posture module again. This continuous cycle enables intelligent adjustment of the TBM's tunneling direction without human intervention.

[0007] The technical solution adopted in this invention is:

[0008] I. An intelligent steering system based on PID control, comprising:

[0009] The shield posture measurement module is used to collect shield posture data during tunnel excavation and then output the current posture deviation value of the shield based on the shield posture data and the preset tunnel design axis.

[0010] The data acquisition module is used to collect and store the current positional deviation value of the tunnel boring machine.

[0011] The PID controller is used to receive the current posture deviation value of the tunnel boring machine stored in the data acquisition module and then output the pressure increment control signal of the propulsion cylinder of each section.

[0012] The hydraulic posture module is used to receive the pressure increment control signals of the propulsion cylinders of each section of the tunnel boring machine and control the propulsion cylinders of each section of the tunnel boring machine to perform actions to continue tunnel excavation.

[0013] Real-time acquisition of tunnel boring machine (TBM) posture data is achieved through the TBM's posture measurement module. After calculation, the posture measurement module transmits the TBM posture deviation to the TBM's data acquisition module. Upon receiving the current posture deviation, the PID controller calculates and outputs control signals to the hydraulic control valves of each section of the TBM's hydraulic posture module, adjusting the pressure of the hydraulic cylinders in each section to regulate the TBM's tunneling direction. When the TBM posture changes, the posture measurement module feeds back the new posture deviation to the PID controller, which then continues to adjust. The PID controller's adjustment frequency depends on the set frequency, improving the efficiency of adjusting the TBM's tunneling direction. The entire control process is a closed-loop control, effectively improving the accuracy of TBM direction adjustment.

[0014] The PID controller includes the following units:

[0015] The three comparison units are used to determine whether the shield's horizontal tendency, vertical tendency, and pitch angle in the shield's current posture deviation value are greater than the limit value.

[0016] n PID units are used to output incremental control signals for the propulsion cylinder pressure of each zone based on the current posture deviation value of the shield.

[0017] The number of PID units is the same as the number of shield propulsion cylinder zones in the tunnel boring machine, which is 4 or 6.

[0018] II. A control method for an intelligent steering system based on PID control, comprising:

[0019] Step 1) Design a PID controller using the shield tunneling intelligent steering simulation system, and build an intelligent steering system based on the PID controller.

[0020] Step 2) When the tunnel boring machine is excavating the tunnel, the intelligent steering system collects the shield position and posture data of the shield in real time and outputs the pressure increment control signal of each section of the shield's propulsion cylinder in real time in combination with the preset tunnel design axis. The system controls the propulsion cylinders of each section of the shield to perform actions to continue the tunnel excavation work.

[0021] In step 1), the intelligent tunnel boring machine (TBM) directional adjustment simulation system includes a signal conversion module, a tunneling parameter prediction module, and a dynamic interactive simulation module for the TBM environment. A PID controller is designed using the intelligent TBM directional adjustment simulation system, as detailed below:

[0022] An initial PID controller is constructed using three comparison units and n PID units. The current simulated shield posture deviation value is input into the PID controller. The PID controller outputs the incremental control signal of the propulsion cylinder pressure for each zone to the signal conversion module. The signal conversion module superimposes the incremental control signal of the propulsion cylinder pressure for each zone with the previously output pressure signal of the propulsion cylinder for each zone to obtain the current pressure signal of the propulsion cylinder for each zone. This signal is saved and output to the shield-environment dynamic interactive simulation module and the tunneling parameter prediction module, respectively. At the same time, the first shield tunneling data and geological parameters are input into the tunneling parameter prediction module. The tunneling parameter prediction module outputs the total propulsion force, propulsion speed, cutterhead torque, and cutterhead rotation speed of the shield to the shield-environment dynamic interactive simulation module. Simultaneously, the second shield tunneling data, shield geometric parameters, geological parameters, and the current simulated shield posture deviation value are input into the shield-environment dynamic interactive simulation module. The shield-environment dynamic interactive simulation module outputs the current simulated shield posture deviation value to itself and the PID controller to complete the closed loop, realizing the training and optimization of the PID controller. Finally, the final PID controller is designed and obtained.

[0023] The signal conversion module has four functions: signal input, signal accumulation, data storage, and signal transmission. The input to the signal conversion module is the pressure increment of the four sets of propulsion cylinders of the tunnel boring machine (TBM) output from the PID controller. Internally, the signal conversion module accumulates the input pressure increment of the four sets of propulsion cylinders with their initial pressure values, and then uses the signal transmission function to output the pressure values ​​of the four sets of propulsion cylinders to the tunneling parameter prediction module and the TBM-environment dynamic interactive simulation module. Simultaneously, the signal conversion module also uses its data storage function to save the output pressure values ​​of the four sets of propulsion cylinders as the initial pressure values ​​to be used in the next iteration of the signal conversion module.

[0024] The tunneling parameter prediction module is built using a Long Short-Term Memory (LSTM) model. Its input data features include the pressure of the propulsion cylinders in each section of the shield tunnel, geological parameters, and the tunneling parameters of the first shield tunnel. The data output from the tunneling parameter prediction module and the data output from the signal conversion module are used as part of the input to the shield-environment dynamic interactive simulation module, effectively solving the problem that other affected parameters do not change after the pressure of the propulsion cylinders in each section changes.

[0025] The shield tunneling machine-environment dynamic interactive simulation module is a shield tunneling machine interactive module built using a Long Short Term Memory (LSTM) model.

[0026] The second shield tunneling parameters in the input features of the shield-environment dynamic interactive simulation module include parameters such as grouting pressure and articulation pressure; shield geometric parameters include parameters such as cutterhead diameter, opening ratio, total length of the main machine, diameter of the front shield, diameter of the middle shield, diameter of the tail shield, diameter of the propulsion cylinder, propulsion stroke, diameter of the articulation hydraulic cylinder, and articulation stroke; geological parameters include parameters such as elevation, natural density, soil particle specific gravity, internal friction angle, cohesion, natural compressive strength, saturated compressive strength, permeability coefficient, Poisson's ratio, elastic model, and shear modulus.

[0027] The first shield tunneling parameter in the input features of the tunneling parameter prediction module includes parameters such as the total thrust of the shield machine, thrust speed, cutterhead rotation speed, cutterhead torque, grouting pressure, and articulation pressure. The geological parameters are the same as those in the input features of the shield-environment dynamic interactive simulation module.

[0028] In step 2), when the tunnel boring machine (TBM) is excavating the tunnel, the TBM posture measurement module collects the TBM posture data and outputs the current posture deviation value of the TBM based on the TBM posture data and the preset tunnel design axis. The data acquisition module collects and stores the current posture deviation value of the TBM. The PID controller receives the current posture deviation value of the TBM stored by the data acquisition module and outputs the pressure increment control signal of each zone propulsion cylinder. The TBM hydraulic posture module receives the pressure increment control signal of each zone propulsion cylinder and controls the propulsion cylinders of each zone of the TBM to perform actions.

[0029] The shield's posture deviation includes the shield head's horizontal deviation, vertical deviation, horizontal tendency, vertical tendency, and pitch angle. When the shield machine's propulsion cylinders are divided into four groups (A, B, C, and D), group A is the right-side propulsion cylinder group, group B is the lower-side propulsion cylinder group, group C is the left-side propulsion cylinder group, and group D is the upper-side propulsion cylinder group. In this case, the PID controller includes four PID units corresponding to the four groups of propulsion cylinders (A, B, C, and D). The specific control process of the PID controller is as follows:

[0030] The three comparison units of the PID controller receive the shield's horizontal tendency, vertical tendency, and pitch angle transmitted from the shield data acquisition module, and determine whether they exceed their respective preset limits. When the shield's horizontal tendency, vertical tendency, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal tendency to the first PID unit A and the third PID unit C, the second comparison unit inputs the shield's vertical tendency to the second PID unit B and the fourth PID unit D, and the third comparison unit inputs the pitch angle to the second PID unit B and the fourth PID unit D. Simultaneously, the shield head horizontal deviation is input to the first PID unit A and the third PID unit C, and the shield head vertical deviation is input to the second PID unit B and the fourth PID unit D. Finally, the four PID units output the propulsion cylinder pressure increment signals of the four sets of propulsion cylinders (A, B, C, and D) to the shield machine's hydraulic posture module. When the shield's horizontal tendency, vertical tendency, and pitch angle are less than the preset limits, the comparison units do not output anything.

[0031] When the number of shield propulsion cylinder zones in the tunnel boring machine (TBM) is 6, that is, the shield propulsion cylinders are divided into six groups: A, B, C, D, E, and F. Group A is the upper right zone propulsion cylinder, group B is the lower right zone propulsion cylinder, group C is the lower zone propulsion cylinder, group D is the lower left zone propulsion cylinder, group E is the upper left zone propulsion cylinder, and group F is the upper zone propulsion cylinder. In this case, the PID controller includes 6 PID units corresponding to the six groups of propulsion cylinders A, B, C, D, E, and F respectively. The specific control process of the PID controller is as follows:

[0032] The three comparison units of the PID controller receive the shield's horizontal direction, vertical direction, and pitch angle transmitted from the shield data acquisition module, and determine whether each value exceeds its preset limit. When the shield's horizontal direction, vertical direction, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal direction to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E, respectively; the second comparison unit inputs the shield's vertical direction to the third PID unit C and the sixth PID unit F, respectively; and the third comparison unit... The unit inputs the pitch angle to the third PID unit C and the sixth PID unit F respectively, while the shield head horizontal deviation is input to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E respectively, and the shield head vertical deviation is input to the third PID unit C and the sixth PID unit F respectively; finally, the six PID units output the propulsion cylinder pressure increment signals of the six groups of propulsion cylinders A, B, C, D, E, and F to the hydraulic posture module of the tunnel boring machine; when the shield horizontal tendency, shield vertical tendency, and pitch angle are less than the preset limit values, the comparison unit does not output.

[0033] Compared with other controllers, the design and implementation of the shield tunneling intelligent orientation controller based on PID control is relatively simple. By reasonably adjusting the PID parameters, it can ensure that the shield can quickly recover to the set target orientation when the shield posture deviates. It has a certain degree of real-time performance and robustness, and can avoid the untimely adjustment of the shield posture when manually adjusting it, which greatly improves the intelligence level of the shield.

[0034] The intelligent shield tunneling machine (TBM) steering system and control method based on PID control described in this invention enables unmanned autonomous steering of TBMs, significantly improving the intelligence level of TBMs. The intelligent TBM steering system and its design method based on PID control have the following technical advantages and innovations: First, the PID controller replaces the TBM operator in performing decision-making tasks and outputting control commands during the TBM steering process, eliminating the need for operator intervention and effectively avoiding a series of drawbacks caused by manual TBM steering, such as misoperation, over-adjustment, and untimely adjustment. Second, the shield-environment dynamic interactive simulation module and the tunneling parameter prediction module are trained using historical tunneling parameters from multiple TBMs, improving the accuracy, applicability, and precision of each module. Third, the intelligent TBM steering simulation system built based on the shield-environment dynamic interactive simulation module can optimize not only the PID controller but also other controllers, expanding the applicability of the intelligent TBM steering controller design method. In summary, this invention has great application potential in intelligent TBM steering systems and their design methods.

[0035] The beneficial effects of this invention are:

[0036] The intelligent shield tunneling machine (TBM) steering system proposed in this invention utilizes a PID controller to replace the TBM operator in steering operations, effectively solving a series of problems that occur during manual TBM steering, such as misoperation, over-adjustment, and untimely adjustment, thus significantly improving the intelligence level of the TBM. Furthermore, the constructed intelligent TBM steering simulation system can be used to design various different controllers, greatly expanding its applicability. In summary, this invention has great application potential in the design of intelligent TBM steering systems and their controllers. Attached Figure Description

[0037] Figure 1 This is a structural diagram of a shield tunnel intelligent directional controller based on PID control.

[0038] Figure 2 This is a schematic diagram of a shield tunneling intelligent directional adjustment simulation system based on PID control.

[0039] Figure 3 Diagram showing the zoning of the tunnel boring machine's propulsion cylinders;

[0040] Figure 4 This is a schematic diagram of a shield tunneling intelligent orientation adjustment system based on PID control. Detailed Implementation

[0041] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] like Figure 4 As shown, the intelligent orientation system based on PID control of the present invention includes: a shield posture measurement module, used to collect shield posture data during tunnel excavation and output the current shield posture deviation value based on the shield posture data and the preset tunnel design axis; a data acquisition module, used to collect and store the current shield posture deviation value; a PID controller, used to receive the current shield posture deviation value stored in the data acquisition module and output the pressure increment control signal of each zone propulsion cylinder; and a hydraulic posture module, used to receive the pressure increment control signal of each zone propulsion cylinder of the shield machine and control each zone propulsion cylinder of the shield machine to perform actions to continue tunnel excavation.

[0043] Real-time acquisition of tunnel boring machine (TBM) posture data is achieved through the TBM's posture measurement module. After calculation, the posture measurement module transmits the TBM posture deviation to the TBM's data acquisition module. Upon receiving the current posture deviation, the PID controller calculates and outputs control signals to the hydraulic control valves of each section of the TBM's hydraulic posture module, adjusting the pressure of the hydraulic cylinders in each section to regulate the TBM's tunneling direction. When the TBM posture changes, the posture measurement module feeds back the new posture deviation to the PID controller, which then continues to adjust. The PID controller's adjustment frequency depends on the set frequency, improving the efficiency of adjusting the TBM's tunneling direction. The entire control process is a closed-loop control, effectively improving the accuracy of TBM direction adjustment.

[0044] like Figure 1 As shown, the PID controller includes: three comparison units, which are used to determine whether the shield horizontal tendency, shield vertical tendency and pitch angle in the current shield posture deviation value of the tunnel boring machine are greater than the limit value; and n PID units, which are used to output the pressure increment control signal of the propulsion cylinder of each zone according to the current shield posture deviation value.

[0045] The number of PID units is the same as the number of shield propulsion cylinder zones in the tunnel boring machine, which is 4 or 6.

[0046] The control method for an intelligent steering system based on PID control of the present invention includes:

[0047] Step 1) Design a PID controller using the shield tunneling intelligent steering simulation system, and build an intelligent steering system based on the PID controller.

[0048] like Figure 2 As shown in step 1), the intelligent tunnel boring machine (TBM) steering simulation system includes a signal conversion module, a tunneling parameter prediction module, and a dynamic interactive simulation module for the TBM environment. A PID controller is designed using the intelligent TBM steering simulation system, as detailed below:

[0049] An initial PID controller is constructed using three comparison units and n PID units. The current simulated shield posture deviation value is input into the PID controller. The PID controller outputs the incremental control signal of the propulsion cylinder pressure for each zone to the signal conversion module. The signal conversion module superimposes the incremental control signal of the propulsion cylinder pressure for each zone with the previously output pressure signal of the propulsion cylinder for each zone to obtain the current pressure signal of the propulsion cylinder for each zone. This signal is saved and output to the shield-environment dynamic interactive simulation module and the tunneling parameter prediction module, respectively. At the same time, the first shield tunneling data and geological parameters are input into the tunneling parameter prediction module. The tunneling parameter prediction module outputs the total propulsion force, propulsion speed, cutterhead torque, and cutterhead rotation speed of the shield to the shield-environment dynamic interactive simulation module. Simultaneously, the second shield tunneling data, shield geometric parameters, geological parameters, and the current simulated shield posture deviation value are input into the shield-environment dynamic interactive simulation module. The shield-environment dynamic interactive simulation module outputs the current simulated shield posture deviation value to itself and the PID controller to complete the closed loop, realizing the training and optimization of the PID controller. Finally, the final PID controller is designed and obtained.

[0050] The signal conversion module has four functions: signal input, signal accumulation, data storage, and signal transmission. The input to the signal conversion module is the pressure increment of the four sets of propulsion cylinders of the tunnel boring machine (TBM) output from the PID controller. Internally, the signal conversion module accumulates the input pressure increment of the four sets of propulsion cylinders with their initial pressure values, and then uses the signal transmission function to output the pressure values ​​of the four sets of propulsion cylinders to the tunneling parameter prediction module and the TBM-environment dynamic interactive simulation module. Simultaneously, the signal conversion module also uses its data storage function to save the output pressure values ​​of the four sets of propulsion cylinders as the initial pressure values ​​to be used in the next iteration of the signal conversion module.

[0051] The tunneling parameter prediction module is built using a long short-term memory model. Its input data features include the pressure of the propulsion cylinders in each section of the shield tunnel, geological parameters, and the tunneling parameters of the first shield tunnel. The data output from the tunneling parameter prediction module and the data output from the signal conversion module are used as part of the input to the shield-environment dynamic interactive simulation module, effectively solving the problem that other affected parameters do not change after the pressure of the propulsion cylinders in each section changes.

[0052] The shield tunneling machine-environment dynamic interactive simulation module is a shield tunneling machine interactive module built using a long short-term memory model.

[0053] The second shield tunneling parameter in the input features of the shield-environment dynamic interactive simulation module includes parameters such as grouting pressure and articulation pressure; shield geometric parameters include parameters such as cutterhead diameter, opening ratio, total length of the main machine, diameter of the front shield, diameter of the middle shield, diameter of the tail shield, diameter of the propulsion cylinder, propulsion stroke, diameter of the articulation hydraulic cylinder, and articulation stroke; geological parameters include parameters such as elevation, natural density, soil particle specific gravity, internal friction angle, cohesion, natural compressive strength, saturated compressive strength, permeability coefficient, Poisson's ratio, elastic model, and shear modulus.

[0054] The first shield tunneling parameter in the input features of the tunneling parameter prediction module includes parameters such as the total thrust of the shield machine, thrust speed, cutterhead rotation speed, cutterhead torque, grouting pressure, and articulation pressure. The geological parameters are the same as those in the input features of the shield-environment dynamic interactive simulation module.

[0055] Step 2) When the tunnel boring machine is excavating the tunnel, the intelligent steering system collects the shield position and posture data of the shield in real time and outputs the pressure increment control signal of each section of the shield's propulsion cylinder in real time in combination with the preset tunnel design axis. The system controls the propulsion cylinders of each section of the shield to perform actions to continue the tunnel excavation work.

[0056] In step 2), when the tunnel boring machine (TBM) is excavating the tunnel, the TBM posture measurement module collects the TBM posture data and outputs the current posture deviation value of the TBM based on the TBM posture data and the preset tunnel design axis. The data acquisition module collects and stores the current posture deviation value of the TBM. The PID controller receives the current posture deviation value of the TBM stored in the data acquisition module and outputs the pressure increment control signal of each zone propulsion cylinder. The TBM hydraulic posture module receives the pressure increment control signal of each zone propulsion cylinder and controls the propulsion cylinders of each zone of the TBM to perform actions.

[0057] Shield tunneling machine posture deviations include shield head horizontal deviation, shield head vertical deviation, shield horizontal tendency, shield vertical tendency, and pitch angle; such as Figure 3 As shown, when the number of shield propulsion cylinders in the tunnel boring machine is 4, that is, the shield propulsion cylinders are divided into four groups: A, B, C, and D. Group A is the right shield propulsion cylinder, group B is the lower shield propulsion cylinder, group C is the left shield propulsion cylinder, and group D is the upper shield propulsion cylinder. At this time, the PID controller includes 4 PID units corresponding to the four groups of propulsion cylinders A, B, C, and D respectively. The specific control process of the PID controller is as follows:

[0058] The three comparison units of the PID controller receive the shield's horizontal tendency, vertical tendency, and pitch angle transmitted from the shield data acquisition module, and determine whether they exceed their respective preset limits. When the shield's horizontal tendency, vertical tendency, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal tendency to the first PID unit A and the third PID unit C, the second comparison unit inputs the shield's vertical tendency to the second PID unit B and the fourth PID unit D, and the third comparison unit inputs the pitch angle to the second PID unit B and the fourth PID unit D. Simultaneously, the shield head horizontal deviation is input to the first PID unit A and the third PID unit C, and the shield head vertical deviation is input to the second PID unit B and the fourth PID unit D. Finally, the four PID units output the propulsion cylinder pressure increment signals of the four sets of propulsion cylinders (A, B, C, and D) to the shield machine's hydraulic posture module. When the shield's horizontal tendency, vertical tendency, and pitch angle are less than the preset limits, the comparison units do not output anything.

[0059] When the number of shield propulsion cylinder zones in the tunnel boring machine (TBM) is 6, that is, the shield propulsion cylinders are divided into six groups: A, B, C, D, E, and F. Group A is the upper right zone propulsion cylinder, group B is the lower right zone propulsion cylinder, group C is the lower zone propulsion cylinder, group D is the lower left zone propulsion cylinder, group E is the upper left zone propulsion cylinder, and group F is the upper zone propulsion cylinder. In this case, the PID controller includes 6 PID units corresponding to the six groups of propulsion cylinders A, B, C, D, E, and F respectively. The specific control process of the PID controller is as follows:

[0060] The three comparison units of the PID controller receive the shield's horizontal direction, vertical direction, and pitch angle transmitted from the shield data acquisition module, and determine whether each value exceeds its preset limit. When the shield's horizontal direction, vertical direction, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal direction to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E, respectively; the second comparison unit inputs the shield's vertical direction to the third PID unit C and the sixth PID unit F, respectively; and the third comparison unit... The unit inputs the pitch angle to the third PID unit C and the sixth PID unit F respectively, while the shield head horizontal deviation is input to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E respectively, and the shield head vertical deviation is input to the third PID unit C and the sixth PID unit F respectively; finally, the six PID units output the propulsion cylinder pressure increment signals of the six groups of propulsion cylinders A, B, C, D, E, and F to the hydraulic posture module of the tunnel boring machine; when the shield horizontal tendency, shield vertical tendency, and pitch angle are less than the preset limit values, the comparison unit does not output.

[0061] Compared with other controllers, the design and implementation of the shield tunneling intelligent orientation controller based on PID control is relatively simple. By reasonably adjusting the PID parameters, it can ensure that the shield can quickly recover to the set target orientation when the shield posture deviates. It has a certain degree of real-time performance and robustness, and can avoid the untimely adjustment of the shield posture when manually adjusting it, which greatly improves the intelligence level of the shield.

[0062] Taking a tunnel boring machine (TBM) with its propulsion cylinders divided into four zones as an example, this paper explains the working principle of each part of the intelligent tunnel boring machine orientation simulation system. The four-zone diagram of the propulsion cylinders of the TBM is shown below. Figure 3As shown, the tunnel is mainly divided into four zones: A, B, C, and D. Following the general principle of tunnel boring machine (TBM) direction adjustment, when the TBM deviates to the right relative to the tunnel's design axis, the pressure of the right-side propulsion cylinders is appropriately increased (i.e., the pressure of group A propulsion cylinders is increased); when the TBM deviates to the left relative to the tunnel's design axis, the pressure of the left-side propulsion cylinders is appropriately increased (i.e., the pressure of group C propulsion cylinders is increased); when the TBM deviates downward relative to the tunnel's design axis, the pressure of the lower propulsion cylinders is appropriately increased (i.e., the pressure of group B propulsion cylinders is increased); and when the TBM deviates upward relative to the tunnel's design axis, the pressure of the upper hydraulic cylinders is appropriately increased (i.e., the pressure of group D propulsion cylinders is increased). Based on this rule for adjusting the TBM's tunneling direction, the intelligent TBM direction adjustment controller based on PID control is optimized. The intelligent TBM direction adjustment controller based on PID control mainly consists of two parts: a comparison unit and a PID unit. The comparison unit has three components, which receive the TBM's horizontal tendency, vertical tendency, and pitch angle output from the TBM-environment dynamic interactive simulation module, and determine whether they exceed the set limit values. If the value exceeds the set limit, the comparison unit transmits the received data to the PID unit; if the value is less than the set limit, the comparison unit does not output data. The PID unit mainly includes four PIDs: PID(A), PID(B), PID(C), and PID(D), which correspond to the four propulsion cylinders of shield tunneling groups A, B, C, and D, respectively. The inputs to PID(A) are the shield head horizontal deviation and the shield tunneling horizontal orientation; the output of PID(A) is the pressure increment of the propulsion cylinders in group A. The inputs to PID(B) are the shield head vertical deviation, the shield tunneling vertical orientation, and the pitch angle; the output of PID(B) is the pressure increment of the propulsion cylinders in group B. The inputs to PID(C) are the shield head horizontal deviation and the shield tunneling horizontal orientation; the output of PID(C) is the pressure increment of the propulsion cylinders in group C. The inputs to PID(D) are the shield head vertical deviation, the shield tunneling vertical orientation, and the pitch angle; the output of PID(D) is the pressure increment of the propulsion cylinders in group D.

[0063] The optimization process of the shield tunneling intelligent directional simulation system based on PID control for the shield tunneling intelligent directional controller includes three steps. First, a set of historical tunneling data from a shield machine containing multiple continuous tunneling sections, which has not been used in training the shield-environment dynamic interaction simulation module and the tunneling parameter prediction module, is prepared. This data includes shield machine geometric parameters, geological parameters, shield posture deviation, and second shield tunneling parameters, which are used as partial inputs to the shield-environment dynamic interaction simulation module. Simultaneously, the aforementioned geological parameters and first shield tunneling parameters are also used as partial inputs to the tunneling parameter prediction module. Then, a set of initial values ​​for the PID controller training is provided, including the propulsion cylinder pressures for each shield section and the shield posture deviation. Specifically, the given propulsion cylinder pressures for each shield section are input to the signal conversion module as initial values, and the shield posture deviation is input to the shield-environment dynamic interaction simulation module as initial input values. Second, the signal conversion module saves the obtained initial values ​​of the propulsion cylinder pressures for each shield section and outputs them to the tunneling parameter prediction module and the shield-environment dynamic interaction simulation module, respectively, as partial inputs to these modules. After obtaining geological parameters from historical tunneling data of the tunnel boring machine (TBM), the first TBM tunneling parameters, and the pressure of the propulsion cylinders in each section of the TBM output from the signal conversion module, the tunneling parameter prediction module outputs the predicted total propulsion force, propulsion speed, cutterhead torque, and cutterhead speed to the TBM-environment dynamic interactive simulation module. After obtaining the pressure of the propulsion cylinders in each section of the TBM output from the signal conversion module, the total propulsion force, propulsion speed, cutterhead torque, and cutterhead speed output from the tunneling parameter prediction module, and the given TBM geometric parameters, current geological parameters, second TBM tunneling parameters, and the given initial input value of the TBM posture deviation, the TBM-environment dynamic interactive simulation module outputs the simulated shield head horizontal deviation, shield head vertical deviation, shield horizontal tendency, shield vertical tendency, and pitch angle to the PID controller, and simultaneously feeds them back to the TBM-environment dynamic interactive simulation module as partial input for the next simulation. The shield's horizontal orientation, vertical orientation, and pitch angle have maximum allowable values. If these values ​​are exceeded, it becomes difficult to adjust the shield's posture deviation to the desired value, and it may even cause the shield shell to collide. Therefore, when training the PID controller, a comparison unit needs to be set up separately, and the limit values ​​for the shield's horizontal orientation, vertical orientation, and pitch angle are set in the comparison unit. If the shield's horizontal orientation, vertical orientation, and pitch angle output from the shield-environment dynamic interactive simulation module to the PID controller do not exceed their respective limit values, the comparison unit outputs a value of 0 to the corresponding PID controller. If the shield's horizontal orientation, vertical orientation, and pitch angle output from the shield-environment dynamic interactive simulation module to the PID controller exceed their respective limit values, the comparison unit outputs the feature exceeding the limit value to the corresponding PID controller.Specifically, when the horizontal trend of the shield tunneling machine is not greater than the limit value, the comparison unit outputs 0 values ​​to PID(A) and PID(C); when the horizontal trend of the shield tunneling machine is greater than the limit value, the comparison unit outputs the horizontal trend of the shield tunneling machine to PID(A) and PID(C). When the vertical trend of the shield tunneling machine is not greater than the limit value, the comparison unit outputs 0 values ​​to PID(B) and PID(D); when the vertical trend of the shield tunneling machine is greater than the limit value, the comparison unit outputs the vertical trend of the shield tunneling machine to PID(B) and PID(D). When the pitch angle is not greater than the limit value, the comparison unit outputs 0 values ​​to PID(B) and PID(D); when the pitch angle is greater than the limit value, the comparison unit outputs the vertical trend of the shield tunneling machine to PID(B) and PID(D). After PID(A), PID(B), PID(C), and PID(D) receive their respective input signals, they calculate and output the corresponding pressure increments of the A-group propulsion cylinder, B-group propulsion cylinder, C-group propulsion cylinder, and D-group propulsion cylinder to the signal conversion module. The signal conversion module sums the pressure increments of propulsion cylinders A, B, C, and D with the initial pressure values ​​of the same groups, and then outputs the sums to the tunneling parameter prediction module and the shield-environment dynamic interactive simulation module. This summation is then saved as the initial pressure values ​​for each section of the shield's propulsion cylinders in the next simulation. This completes one simulation. This simulation process is repeated until the prepared data is exhausted, or the shield's horizontal and vertical deviations both approach zero. The third step involves repeatedly performing the training process from step two, changing the PID parameters and PID controller adjustment frequency each time, until the shield's posture deviations approach zero and the adjustment frequency meets the requirements, while preventing the shield from exhibiting serpentine movement.

[0064] This invention's system enables three functions: real-time acquisition of shield tunneling machine (TBM) posture data, output of control signals, and adjustment of TBM posture. Traditional methods, where the TBM operator manually adjusts the tunneling direction, require the operator to periodically check the shield's posture deviation on the machine's control panel to determine the machine's offset direction. The operator then relies on experience to adjust the pressure of the corresponding hydraulic cylinders in the shield's thrust sections to adjust the tunneling direction. However, the PID-controlled intelligent TBM steering system proposed in this invention receives the current shield posture deviation measured by the shield posture measurement module from the data acquisition module at a set time point. It then outputs corresponding control signals to the hydraulic posture module to adjust the thrust cylinders in the corresponding sections. In other words, the PID controller replaces the operator in adjusting the tunneling direction.

[0065] The schematic diagram of the intelligent directional adjustment system control method for tunnel boring machines based on PID control of the present invention is shown below. Figure 4 As shown, the PID-based intelligent tunnel boring machine (TBM) steering controller receives the current posture deviation information of the TBM from the data acquisition module at a set time point. Based on these parameters, it generates incremental pressure signals for the propulsion cylinders of each section of the TBM and outputs them to the hydraulic attitude module. The hydraulic attitude module adjusts the pressure of the corresponding propulsion cylinders according to the signals output by the PID-based intelligent steering controller, thereby controlling the movement of the corresponding propulsion cylinders to change the TBM posture. After the TBM posture changes, the PID-based intelligent steering controller receives updated data from the TBM posture measurement module and the data acquisition module at the next set time point, and then outputs the optimal incremental pressure control signals for the propulsion cylinders of each section of the TBM under the current state to the hydraulic attitude module again. This continuous cycle enables intelligent adjustment of the TBM's tunneling direction without human intervention. Furthermore, the control method of this intelligent TBM steering system retains a manual adjustment channel for the TBM operator in case of emergencies.

[0066] This method takes a tunnel boring machine (TBM) with four zones for its propulsion cylinders as an example, but it is not only applicable to TBMs with four zones for their propulsion cylinders. It is also applicable to other zone types, such as six-zone TBMs. Only the number of PID controllers, the input and output structure of the signal conversion module, the input structure of the tunneling parameter prediction module, and the input structure of the TBM-environment dynamic interactive simulation module need to be changed.

[0067] Furthermore, the controller design method of this intelligent tunnel boring machine steering system is not limited to designing controllers equipped with PID algorithms, but is also applicable to controllers equipped with other intelligent algorithms. It is only necessary to change the corresponding input characteristics according to the nature of the designed controller.

Claims

1. An intelligent steering system based on PID control, characterized in that, include: The shield posture measurement module is used to collect shield posture data during tunnel excavation and then output the current posture deviation value of the shield based on the shield posture data and the preset tunnel design axis. The data acquisition module is used to collect and store the current positional deviation value of the tunnel boring machine. The PID controller is used to receive the current posture deviation value of the shield stored in the data acquisition module and then output the pressure increment control signal of the propulsion cylinder of each section. The hydraulic posture module is used to receive the pressure increment control signal of the propulsion cylinders of each section of the tunnel boring machine and control the propulsion cylinders of each section of the tunnel boring machine to perform actions to continue tunnel excavation; The PID controller includes the following units: The three comparison units are used to determine whether the shield's horizontal tendency, vertical tendency, and pitch angle in the shield's current posture deviation value are greater than the limit value. n PID units are used to output the pressure increment control signal of the propulsion cylinder of each zone according to the current posture deviation value of the shield. The number of PID units is the same as the number of shield propulsion cylinder partitions of the tunnel boring machine, which is 4 or 6. When the shield's position changes, the intelligent shield orientation controller based on PID control receives updated data from the shield's position measurement module and data acquisition module at the next set time point. Then, it outputs the optimal incremental pressure signal of the shield machine's propulsion cylinders in each zone to the hydraulic position module, continuously cycling to achieve intelligent adjustment of the shield machine's tunneling direction without human intervention.

2. The control method for an intelligent steering system based on PID control according to claim 1, characterized in that, include: Step 1) Design a PID controller using the shield tunneling intelligent steering simulation system, and build an intelligent steering system based on the PID controller; Step 2) When the tunnel boring machine is excavating the tunnel, the intelligent steering system collects the shield position and posture data of the shield in real time and outputs the pressure increment control signal of each section of the shield's propulsion cylinder in real time in combination with the preset tunnel design axis. The system controls the propulsion cylinders of each section of the shield to perform actions to continue the tunnel excavation work. In step 1), the intelligent tunnel boring machine (TBM) directional adjustment simulation system includes a signal conversion module, a tunneling parameter prediction module, and a dynamic interactive simulation module for the TBM environment. A PID controller is designed using the intelligent TBM directional adjustment simulation system, as detailed below: An initial PID controller is constructed using three comparison units and n PID units. The current simulated shield posture deviation value is input into the PID controller. The PID controller outputs the incremental control signal of the propulsion cylinder pressure for each zone to the signal conversion module. The signal conversion module superimposes the incremental control signal of the propulsion cylinder pressure for each zone with the previously output pressure signal of the propulsion cylinder for each zone to obtain the current pressure signal of the propulsion cylinder for each zone. This signal is saved and output to the shield-environment dynamic interactive simulation module and the tunneling parameter prediction module, respectively. At the same time, the first shield tunneling data and geological parameters are input into the tunneling parameter prediction module. The tunneling parameter prediction module outputs the total propulsion force, propulsion speed, cutterhead torque, and cutterhead rotation speed of the shield to the shield-environment dynamic interactive simulation module. Simultaneously, the second shield tunneling data, shield geometric parameters, geological parameters, and the current simulated shield posture deviation value are input into the shield-environment dynamic interactive simulation module. The shield-environment dynamic interactive simulation module outputs the current simulated shield posture deviation value to itself and the PID controller to complete the closed loop, realizing the training and optimization of the PID controller. Finally, the final PID controller is designed and obtained.

3. The control method for an intelligent steering system based on PID control according to claim 2, characterized in that: The second shield tunneling parameter in the input features of the shield-environment dynamic interactive simulation module includes grouting pressure and articulation pressure parameters; shield geometric parameters include cutterhead diameter, opening ratio, main machine length, front shield diameter, middle shield diameter, tail shield diameter, propulsion cylinder diameter, propulsion stroke, articulation hydraulic cylinder diameter and articulation stroke parameters; geological parameters include elevation, natural density, soil particle specific gravity, internal friction angle, cohesion, natural compressive strength, saturated compressive strength, permeability coefficient, Poisson's ratio, elastic model and shear modulus parameters; The first shield tunneling parameter in the input features of the tunneling parameter prediction module includes the total thrust of the shield machine, thrust speed, cutterhead rotation speed, cutterhead torque, grouting pressure, and articulation pressure parameters.

4. The control method for an intelligent steering system based on PID control according to claim 2, characterized in that: In step 2), when the tunnel boring machine is excavating the tunnel, the shield posture measurement module collects the shield posture data of the shield machine and outputs the current posture deviation value of the shield machine based on the shield posture data and the preset tunnel design axis. The data acquisition module collects and stores the current posture deviation value of the shield machine. The PID controller receives the current posture deviation value of the shield machine stored by the data acquisition module and outputs the pressure increment control signal of each zone propulsion cylinder. The shield hydraulic posture module receives the pressure increment control signal of each zone propulsion cylinder and controls each zone propulsion cylinder of the shield machine to perform actions. The shield's posture deviation includes the shield head's horizontal deviation, vertical deviation, horizontal tendency, vertical tendency, and pitch angle. When the shield machine's propulsion cylinders are divided into four groups (A, B, C, and D), group A is the right-side propulsion cylinder group, group B is the lower-side propulsion cylinder group, group C is the left-side propulsion cylinder group, and group D is the upper-side propulsion cylinder group. In this case, the PID controller includes four PID units corresponding to the four groups of propulsion cylinders (A, B, C, and D). The specific control process of the PID controller is as follows: The three comparison units of the PID controller receive the shield's horizontal direction, vertical direction, and pitch angle from the shield data acquisition module, and determine whether they exceed their respective preset limits. When the shield's horizontal direction, vertical direction, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal direction to the first PID unit A and the third PID unit C, the second comparison unit inputs the shield's vertical direction to the second PID unit B and the fourth PID unit D, and the third comparison unit inputs the pitch angle to the second PID unit B and the fourth PID unit D. Simultaneously, the shield head horizontal deviation is input to the first PID unit A and the third PID unit C, and the shield head vertical deviation is input to the second PID unit B and the fourth PID unit D. Finally, the four PID units output the propulsion cylinder pressure increment signals of the four sets of propulsion cylinders (A, B, C, and D) to the shield machine's hydraulic posture module. When the shield's horizontal direction, vertical direction, and pitch angle are less than the preset limits, the comparison units do not output anything. When the number of shield propulsion cylinder zones in the tunnel boring machine (TBM) is 6, that is, the shield propulsion cylinders are divided into six groups: A, B, C, D, E, and F. Group A is the upper right zone propulsion cylinder, group B is the lower right zone propulsion cylinder, group C is the lower zone propulsion cylinder, group D is the lower left zone propulsion cylinder, group E is the upper left zone propulsion cylinder, and group F is the upper zone propulsion cylinder. In this case, the PID controller includes 6 PID units corresponding to the six groups of propulsion cylinders A, B, C, D, E, and F respectively. The specific control process of the PID controller is as follows: The three comparison units of the PID controller receive the shield's horizontal direction, vertical direction, and pitch angle transmitted from the shield data acquisition module, and determine whether each value exceeds its preset limit. When the shield's horizontal direction, vertical direction, and pitch angle exceed their respective preset limits, the first comparison unit inputs the shield's horizontal direction to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E, respectively; the second comparison unit inputs the shield's vertical direction to the third PID unit C and the sixth PID unit F, respectively; and the third comparison unit... The unit inputs the pitch angle to the third PID unit C and the sixth PID unit F respectively, while the shield head horizontal deviation is input to the first PID unit A, the second PID unit B, the fourth PID unit D, and the fifth PID unit E respectively, and the shield head vertical deviation is input to the third PID unit C and the sixth PID unit F respectively; finally, the six PID units output the propulsion cylinder pressure increment signals of the six groups of propulsion cylinders A, B, C, D, E, and F to the hydraulic posture module of the tunnel boring machine; when the shield horizontal tendency, shield vertical tendency, and pitch angle are less than the preset limit values, the comparison unit does not output.

Images