A shield machine attitude autonomous adjustment control method and system

Abstract

The present invention discloses a shield attitude autonomous adjustment control method and system, belonging to the technical field of shield attitude adjustment. It includes: obtaining key tunneling parameters and shield attitude parameters during shield machine excavation and pre-processing the data; screening out relevant tunneling parameters to construct a shield attitude prediction model, taking relevant tunneling parameters as input and shield attitude parameters as output, and using the shield attitude prediction model to predict the tunneling predicted attitude of the shield machine within a predetermined range in the future; judging whether the predicted tunneling attitude meets the requirements, and determining the shield correction path planning curve based on the judgment result; establishing a grouped oil cylinder pressure prediction model, and predicting the pressure decision value of each grouped oil cylinder according to the determined shield correction path planning curve. The present invention provides a complete shield attitude autonomous control system, which solves the problems of inaccurate shield attitude judgment, unstable correction operation control, etc., avoids excessive shield attitude fluctuations and serpentine correction, and realizes precise control of shield attitude.

Description

Technical Field

[0001] The present invention belongs to the technical field of shield machine posture adjustment, and in particular relates to a shield machine posture autonomous adjustment control method and system. Background Art

[0002] The shield machine is a large-scale tunnel construction device that integrates multidisciplinary technologies, including mechanical, hydraulic, electrical, and control technologies. Due to its high degree of automation, excellent safety performance, and rapid construction speed, shield machines are widely used in projects such as urban subway tunnels, municipal pipeline networks, and cross-sea tunnels. However, with economic development, the construction difficulty of tunnel projects continues to increase. Shield machines face uncertain geological environments, difficult working conditions, and complex supporting equipment, often making it difficult to ensure tunnel construction quality. The degree of alignment between the actual axis of the shield machine and the designed axis of the tunnel is a key factor in determining tunnel construction quality.

[0003] Attitude control is paramount during shield tunneling. Excessive deviation can easily cause the actual tunneling trajectory to deviate from the designed axis, leading to segment misalignment and even safety hazards such as water leakage. During actual shield construction, the shield machine's attitude is primarily controlled by the shield driver, who controls the pressure of the shield machine's propulsion system's zoned cylinders based on measurement system results and the designed axis information. This correction often relies heavily on the shield driver's skill, experience, and commitment, making tunnel formation quality uncertain. Summary of the Invention

[0004] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a shield machine posture autonomous adjustment control method and system.

[0005] The present invention is achieved through the following technical solutions: A shield attitude autonomous adjustment control method comprises the following steps:

[0006] Obtain key tunneling parameters and shield attitude parameters during shield machine excavation, and perform data preprocessing on each to obtain data parameters of the same magnitude;

[0007] Screening out relevant excavation parameters based on data parameters of the same magnitude, wherein the relevant excavation parameters have a predetermined degree of correlation with shield attitude parameters;

[0008] Construct a shield machine attitude prediction model, take relevant tunneling parameters as input and shield machine attitude parameters as output, and use the shield machine attitude prediction model to predict the tunneling attitude of the shield machine within a predetermined range in the future;

[0009] Construct tunneling constraints, analyze the predicted parameter values of the tunneling prediction posture and the tunneling constraints to obtain a judgment result; determine the shield correction path planning curve based on the judgment result;

[0010] A group cylinder pressure prediction model is established to predict the pressure decision value of each group cylinder based on the determined shield correction path planning curve.

[0011] In a further embodiment, the following steps are also included:

[0012] Build an intelligent algorithm based PID Control model, build and shield machine PLC communication relationship;

[0013] The pressure decision value is transmitted to PID The control model is used to control the shield machine's grouped cylinder proportional valves and relief valves to achieve autonomous control of the shield machine's posture.

[0014] In a further embodiment, the key excavation parameters include at least: A Cylinder pressure, B Cylinder pressure, C Cylinder pressure, D Cylinder pressure, E Cylinder pressure, F Cylinder pressure, driving thrust, propulsion speed, cutterhead torque, cutterhead speed, penetration, shield tail gap, grouting volume, grease injection volume, slurry inflow, slurry outflow, excavation chamber pressure and air cushion chamber pressure;

[0015] The shield posture parameters include at least: a shield head horizontal deviation value, a shield head vertical deviation value, a shield tail horizontal deviation value and a shield tail vertical deviation value.

[0016] In a further embodiment, the preprocessing process of the data parameters of the same magnitude includes:

[0017] The key excavation parameters and shield attitude parameters are partitioned, cleaned and eliminated to obtain valid parameter data;

[0018] The effective parameter data are locally dimensionless processed to make them at the same order of magnitude.

[0019] In a further embodiment, the screening process of the relevant excavation parameters is as follows:

[0020] Calculating the importance scores of key tunneling parameters using gradient decision trees and score the key excavation parameters according to their importance Sort the parameters in descending order; define the key excavation parameters ranked in the front as relevant excavation parameters;

[0021] Among them, the importance score The calculation formula is as follows:

[0022] ;

[0023] Where, Represents a node, represents the number of all trees, express m The number of non-leaf nodes in a tree, Indicates the m The first j The partitioning features of non-leaf nodes, is the loss function, 、 Respectively indicate the m The first j The sum of the first-order derivatives and the sum of the second-order derivatives of all samples on non-leaf nodes, 、 Respectively indicate the m On the tree j The sum of the first-order derivatives on the left node L and the first-order derivatives on the right node R of the non-leaf nodes, 、 Respectively indicate the m On the tree j The sum of the second-order derivatives on the left node L and the right node R of the non-leaf nodes, represents the hyperparameter of the regularization term.

[0024] In a further embodiment, the excavation constraint conditions include: a constraint condition for horizontal deviation of the shield head, a constraint condition for vertical deviation of the shield head, a constraint condition for horizontal deviation of the shield tail, and a constraint condition for vertical deviation of the shield tail;

[0025] The predicted parameter values include: shield head horizontal deviation predicted value , shield head vertical deviation prediction value , shield tail horizontal deviation prediction value and the predicted value of shield tail horizontal deviation ;

[0026] Correspondingly, the analysis process of the judgment result is: if the judgment formula is satisfied, the judgment result is that the excavation requirements are met; otherwise, the judgment result is that the excavation requirements are not met;

[0027] The judgment formula is expressed as follows:

[0028] ;

[0029] Where, 、 、 and They are the standard value of horizontal deviation of the shield head, the standard value of vertical deviation of the shield head, the standard value of horizontal deviation of the shield tail and the standard value of horizontal deviation of the shield tail.

[0030] In a further embodiment, the method for determining the shield deviation correction path planning curve is as follows:

[0031] If the result is that the tunneling requirements are met, tunneling will be carried out according to the current shield deviation correction path planning curve;

[0032] If the judgment result is that it does not meet the excavation requirements, a shield posture correction path planning parameter equation is established based on the current posture information and excavation constraints; a new shield correction path planning curve is determined according to the shield posture correction path planning parameter equation, and excavation is carried out according to the new shield correction path planning curve.

[0033] In a further embodiment, the grouped cylinder pressure prediction model is expressed as follows:

[0034] ;

[0035] Where, is the sample size, Represents a sample Parameter value, Before Decision trees are used to analyze the samples The predicted value of Indicates the t Decision tree model The complexity of is the loss function, is the objective function;

[0036] Correspondingly, the process of determining the pressure decision value is as follows: taking the shield posture parameters on the determined shield correction path planning curve as input, and using the grouped cylinder pressure prediction model to obtain the pressure decision value of each grouped cylinder.

[0037] In a further embodiment, the process of establishing the shield posture correction path planning parameter equation is as follows:

[0038] Obtain current posture information, including the axis coordinates of the shield tunnel and the minimum turning radius of the shield machine , Minimum turning radius of the tunnel , shield tail gap , segment length and cylinder stroke difference;

[0039] Taking the shortest path as the goal, the minimum correction curvature radius is calculated using the following formula: :

[0040] ;

[0041] Where, is the allowable correction radius of the shield tail gap, and its calculation formula is: ;

[0042] is the minimum correction radius allowed by the cylinder stroke difference, and its calculation formula is: , is the radius of the thrust cylinder, The distance from the cutterhead to the shield tail;

[0043] Based on the minimum correction curvature radius , the shield posture correction path planning parameter equation that meets the tunneling constraints is constructed and expressed as:

[0044] ;

[0045] ;

[0046] Where, 、 are the starting and ending coordinates of the correction curve, is the coordinate of the fitting point after the end point, is the center coordinate of the shield tail, is the curvature radius of the correction curve, 、 The equations are The first and second derivatives of 、 、 and d are coefficients, Indicates that it is restricted.

[0047] A shield machine attitude autonomous adjustment control system, used to implement the shield machine attitude autonomous adjustment control method as described above, comprising:

[0048] The data acquisition module is configured to obtain key tunneling parameters and shield attitude parameters during tunneling of the shield machine;

[0049] The data processing module is configured to perform data preprocessing on key tunneling parameters and shield attitude parameters to obtain data parameters of the same magnitude; and to screen out relevant tunneling parameters based on the data parameters of the same magnitude, wherein the relevant tunneling parameters have a predetermined degree of correlation with the shield attitude parameters;

[0050] The intelligent decision-making module is configured to construct a shield machine attitude prediction model, taking relevant tunneling parameters as input and shield machine attitude parameters as output. The shield machine attitude prediction model is used to predict the tunneling prediction attitude of the shield machine within a predetermined range in the future. The predicted parameter values of the tunneling prediction attitude are analyzed in correspondence with the tunneling constraint conditions to obtain a judgment result. Based on the judgment result, a shield machine deviation correction path planning curve is determined. A group cylinder pressure prediction model is established to predict the pressure decision value of each group cylinder based on the determined shield machine deviation correction path planning curve.

[0051] Autonomous control module, is set to build an intelligent algorithm based PID Control model, build and shield machine PLC Communication relationship; transmitting the pressure decision value to the PID control model for controlling the shield machine group cylinder proportional valve and relief valve to achieve autonomous control of the shield posture.

[0052] Beneficial effects of the present invention: The present invention provides a shield attitude autonomous control method, which predicts the shield attitude within a certain mileage range in the future by constructing a shield attitude prediction model, and judges whether the shield attitude within a certain mileage range in the future meets the requirements based on the actual construction shield tunneling constraints, and then establishes a shield attitude correction path planning parameter equation, and provides a shield attitude correction path planning curve. Subsequently, based on the shield attitude correction path planning curve, combined with the shield attitude parameters, a correlation between the shield correction curve related parameters and the grouped cylinders is established, and the pressure and stroke decision values of each grouped cylinder are given. Finally, by establishing a correlation with the shield machine PLC Communication relationship, use PID The controller realizes the control of zoned cylinder pressure and stroke, and ultimately achieves autonomous control of the shield machine's posture.

[0053] The present invention provides a complete autonomous control system for shield machine attitude, which solves the problems of inaccurate judgment of shield machine attitude and unstable control of deviation correction operation during manual operation. It can effectively avoid excessive fluctuation of shield machine attitude and serpentine deviation correction, and realize autonomous and precise control of shield machine attitude. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of the shield machine posture autonomous adjustment control method of Example 1.

[0055] Figure 2 This is the architecture diagram of the shield posture autonomous adjustment control system of Example 2. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the examples of the present invention. This description only represents some embodiments of the present invention and does not represent all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without expending creative work are within the scope of protection of the present invention.

[0057] Example 1

[0058] like Figure 1 As shown, this embodiment discloses a shield posture autonomous adjustment control method, comprising the following steps:

[0059] Obtain key tunneling parameters and shield attitude parameters during shield machine excavation, and perform data preprocessing on each to obtain data parameters of the same magnitude;

[0060] Screening out relevant excavation parameters based on data parameters of the same magnitude, wherein the relevant excavation parameters have a predetermined degree of correlation with shield attitude parameters;

[0061] use GRU The neural network constructs a shield posture prediction model, takes relevant tunneling parameters as input and shield posture parameters as output, and uses the shield posture prediction model to predict the tunneling predicted posture of the shield machine within a future predetermined range; the future predetermined range is in meters, which can be 20 meters, 30 meters, etc.

[0062] Construct tunneling constraints, analyze the predicted parameter values of the tunneling prediction posture and the tunneling constraints to obtain a judgment result; determine the shield correction path planning curve based on the judgment result;

[0063] A group cylinder pressure prediction model is established to predict the pressure decision value of each group cylinder based on the determined shield correction path planning curve.

[0064] In a further embodiment, the key excavation parameters include at least: A Cylinder pressure, B Cylinder pressure, C Cylinder pressure, D Cylinder pressure, E Cylinder pressure, F Cylinder pressure, driving thrust, propulsion speed, cutterhead torque, cutterhead speed, penetration, shield tail gap, grouting volume, grease injection volume, slurry inflow, slurry outflow, excavation chamber pressure and air cushion chamber pressure;

[0065] The shield posture parameters include at least: a shield head horizontal deviation value, a shield head vertical deviation value, a shield tail horizontal deviation value and a shield tail vertical deviation value.

[0066] Key tunneling parameters and shield attitude parameters often have significantly different original magnitudes due to their physical meaning and engineering realities. For example, tunneling thrust is generally large, perhaps measured in kilonewtons or even larger, while shield tail clearance is relatively small, measured in millimeters. When conducting statistical analyses such as calculating the coefficient of variation of each parameter to measure data dispersion and extracting key features through principal component analysis, if these parameters are not dimensionlessly normalized to the same magnitude, the larger parameters will dominate the calculated results, obscuring the true data characteristics and variations of the smaller parameters. This can lead to inaccurate conclusions about the impact of each parameter on the overall tunneling performance of the shield machine.

[0067] Therefore, the processing process of data parameters of the same order of magnitude includes: data partitioning, data cleaning, and data elimination of key tunneling parameters and shield posture parameters to obtain valid parameter data; and local dimensionless processing of the valid parameter data to make them at the same order of magnitude.

[0068] The data partitioning process can be performed based on the different strata (e.g., soft soil, rock, etc.) that the shield machine traverses. This is because the performance and variation patterns of various parameters often differ significantly under different strata. This facilitates more targeted analysis of the data characteristics within each area, identifying parameter trends and interrelationships under specific operating conditions. Further understanding is that when the stratum type changes, it marks the starting point of a new partition. All parameter data within the same stratum interval is then grouped into a single partition, and the data for each partition is stored separately for subsequent processing and analysis.

[0069] Data cleaning involves inspecting each partitioned data section and removing any data that is obviously erroneous or inconsistent with actual physical meaning. Examples of erroneous data in shield machine parameters include sensor failure causing a sudden, abnormally high value in tunneling thrust (far exceeding the acceptable range for this parameter under normal operation), or loss of shield tail clearance data due to communication interference. Cleaning this data ensures data quality for subsequent analysis and processing, preventing misleading results from erroneous data.

[0070] Correspondingly, after data cleaning, some data may still exist that, while not obviously incorrect, is not particularly relevant to the analysis objectives, or is redundant in specific analysis scenarios. Removing this data allows the remaining data to focus more on the key tunneling performance and attitude characteristics of the shield machine. For example, when studying the relationship between attitude parameters and tunneling speed of a shield machine in stable formations, some temporary data recorded due to brief external interference can be eliminated if it is not relevant to the overall analysis.

[0071] The dimensionless processing mentioned in this embodiment may be normalization processing or standardization processing.

[0072] Combined with the above description, the screening process of relevant excavation parameters is as follows:

[0073] Calculating the importance scores of key tunneling parameters using gradient decision trees and score the key excavation parameters according to their importance Sort the parameters in descending order; define the key excavation parameters ranked in the front as relevant excavation parameters;

[0074] Among them, the importance score The calculation formula is as follows:

[0075] ;

[0076] Where, Represents a node, represents the number of all trees, express m The number of non-leaf nodes in a tree, Indicates the m The first j The partitioning features of non-leaf nodes, is the loss function, 、 Respectively indicate the m The first j The sum of the first-order derivatives and the sum of the second-order derivatives of all samples on non-leaf nodes, 、 Respectively indicate the m On the tree j The sum of the first-order derivatives on the left node L and the first-order derivatives on the right node R of the non-leaf nodes, 、 , respectively, represent the m On the tree j The sum of the second-order derivatives on the left node L and the right node R of the non-leaf nodes, represents the hyperparameter of the regularization term.

[0077] For example, according to the current excavation scene and geological conditions, A - F Taking the group of oil cylinders as an example, the importance score of the key excavation parameters of each group of oil cylinders is calculated. The top three are preset before customization. A The excavation thrust of the group cylinder, A The propulsion speed of the group cylinder, C The cutter head torque of the group cylinder is selected as the relevant excavation parameter.

[0078] To further illustrate, the specific calculation scores of the importance of each key parameter are: A Cylinder pressure 0.13min, B Cylinder pressure 0.13min, C Cylinder pressure 0.14min, D Cylinder pressure 0.14min, E Cylinder pressure 0.15min, F The group cylinder pressure is 0.14 points, the excavation thrust is 0.06 points, the propulsion speed is 0.04 points, the cutter head torque is 0.03 points, the cutter head speed is 0.03 points, the penetration rate is 0.003 points, the shield tail gap is 0.004 points, the grouting volume is 0.0008 points, the grease injection volume is 0.0007 points, the slurry inflow rate is 0.0004 points, the slurry outflow rate is 0.0004 points, the excavation chamber pressure is 0.00043 points and the air cushion chamber pressure is 0.00033 points.

[0079] In this embodiment, the top 10 parameters with the highest scores are selected as key tunneling parameters, which are: A Cylinder pressure, B Cylinder pressure, C Cylinder pressure, D Cylinder pressure, E Cylinder pressure, F Group cylinder pressure, tunneling thrust, propulsion speed, cutterhead torque, cutterhead speed, penetration and shield tail clearance.

[0080] In a further embodiment, the excavation constraint conditions include: a constraint condition for horizontal deviation of the shield head, a constraint condition for vertical deviation of the shield head, a constraint condition for horizontal deviation of the shield tail, and a constraint condition for vertical deviation of the shield tail.

[0081] Correspondingly, the predicted parameter values of the tunneling prediction posture are analyzed correspondingly with the tunneling constraint conditions to obtain the judgment result, wherein the predicted parameter values include: the predicted value of the shield head horizontal deviation , shield head vertical deviation prediction value , shield tail horizontal deviation prediction value and the predicted value of shield tail horizontal deviation ;

[0082] If the following formula is satisfied, the result is considered to meet the requirements, otherwise it does not meet the requirements:

[0083] ;

[0084] Where, 、 、 and They are the standard value of horizontal deviation of the shield head, the standard value of vertical deviation of the shield head, the standard value of horizontal deviation of the shield tail and the standard value of horizontal deviation of the shield tail.

[0085] For example, in this embodiment 、 、 and The values are as follows: , so the above formula is updated to:

[0086] .

[0087] Furthermore, the method for determining the shield deviation correction path planning curve is as follows:

[0088] If the result is that the tunneling requirements are met, tunneling will be carried out according to the current shield deviation correction path planning curve;

[0089] If the judgment result is that it does not meet the excavation requirements, a shield posture correction path planning parameter equation is established based on the current posture information and excavation constraints; a new shield correction path planning curve is determined according to the shield posture correction path planning parameter equation, and excavation is carried out according to the new shield correction path planning curve.

[0090] The expression of the grouped cylinder pressure prediction model is as follows:

[0091] ;

[0092] Where, is the sample size, Represents a sample Parameter value, Before Decision trees are used to analyze the samples The predicted value of Indicates the t Decision tree model The complexity of is the loss function, is the objective function;

[0093] Correspondingly, the process of determining the pressure decision value is as follows: taking the shield posture parameters (shield head horizontal deviation value, shield head vertical deviation value, shield tail horizontal deviation value and shield tail vertical deviation value) on the determined shield correction path planning curve as input, and using the grouped cylinder pressure prediction model to obtain the pressure decision value of each grouped cylinder. Combined with the above example, the specific pressure decision value is A - F The pressure decision value of the group cylinder.

[0094] Furthermore, the process of establishing the shield posture correction path planning parameter equation is as follows:

[0095] Obtain current posture information, including the axis coordinates of the shield tunnel and the minimum turning radius of the shield machine , Minimum turning radius of the tunnel , shield tail gap , segment length and cylinder stroke difference;

[0096] Taking the shortest path as the goal, the minimum correction curvature radius is calculated using the following formula: :

[0097] ;

[0098] Where, is the allowable correction radius of the shield tail gap, and its calculation formula is: ;

[0099] is the minimum correction radius allowed by the cylinder stroke difference, and its calculation formula is: , is the radius of the thrust cylinder, The distance from the cutterhead to the shield tail;

[0100] Based on the minimum correction curvature radius , construct the shield posture correction path planning parameter equation that meets the tunneling constraints;

[0101] The shield posture correction path planning parameter equation is constructed using a cubic parabola, and the equation is:

[0102] ;

[0103] In order to make the shield machine deviation correction planning curve smooth and facilitate the shield machine to fit the tunnel design axis to achieve deviation correction, the following conditions must be met:

[0104] ;

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] Where, 、 are the starting and ending coordinates of the correction curve, is the coordinate of the fitting point after the end point, is the center coordinate of the shield tail, is the curvature radius of the correction curve, 、 The equations are The first and second derivatives of 、 、 and d All are coefficients.

[0110] In a further embodiment, based on the pressure decision values of each grouped oil cylinder obtained in the above steps, it also includes: constructing a system based on an intelligent algorithm PID Control model, build and shield machine PLC communication relationship;

[0111] The pressure decision value is transmitted to the PID control model to control the shield machine's grouped cylinder proportional valves and relief valves to achieve autonomous control of the shield machine's posture.

[0112] Example 2

[0113] This embodiment discloses a shield attitude autonomous adjustment control system, which is used to implement the shield attitude autonomous adjustment control method as described in Example 1. Figure 2 Shown include:

[0114] The data acquisition module is configured to obtain key tunneling parameters and shield attitude parameters during tunneling of the shield machine;

[0115] The data processing module is configured to perform data preprocessing on key tunneling parameters and shield attitude parameters to obtain data parameters of the same magnitude; and to screen out relevant tunneling parameters based on the data parameters of the same magnitude, wherein the relevant tunneling parameters have a predetermined degree of correlation with the shield attitude parameters;

[0116] The intelligent decision-making module is configured to construct a shield machine attitude prediction model, taking relevant tunneling parameters as input and shield machine attitude parameters as output. The shield machine attitude prediction model is used to predict the tunneling attitude of the shield machine within a predetermined range in the future. The module also constructs tunneling constraints to determine whether the predicted tunneling attitude meets the requirements and determines the shield machine deviation correction path planning curve based on the judgment result. The module also establishes a group cylinder pressure prediction model to predict the pressure decision value of each group cylinder based on the determined shield machine deviation correction path planning curve.

[0117] Autonomous control module, is set to build an intelligent algorithm based PID Control model, build and shield machine PLC Communication relationship; transmitting the pressure decision value to the PID control model for controlling the shield machine group cylinder proportional valve and relief valve to achieve autonomous control of the shield posture.

Claims

1. A shield machine posture autonomous adjustment control method, characterized in that: The following steps are involved: Obtain key tunneling parameters and shield attitude parameters during shield machine excavation, and perform data preprocessing on each to obtain data parameters of the same magnitude; Based on data parameters of the same magnitude, relevant excavation parameters are screened out, wherein the relevant excavation parameters have a predetermined degree of correlation with shield attitude parameters. The screening process of the relevant excavation parameters is as follows: Calculating the importance scores of key tunneling parameters using gradient decision trees and score the key excavation parameters according to their importance Sort the parameters in descending order; define the key excavation parameters ranked in the front as relevant excavation parameters; Construct a shield machine attitude prediction model, take relevant tunneling parameters as input and shield machine attitude parameters as output, and use the shield machine attitude prediction model to predict the tunneling attitude of the shield machine within a predetermined range in the future; Constructing excavation constraints, performing corresponding analysis on the predicted parameter values of the excavation prediction posture and the excavation constraints to obtain a judgment result; determining the shield deviation correction path planning curve based on the judgment result, and the method for determining the shield deviation correction path planning curve is as follows: If the result is that the tunneling requirements are met, tunneling will be carried out according to the current shield deviation correction path planning curve; If the judgment result is that it does not meet the excavation requirements, a shield posture correction path planning parameter equation is established based on the current posture information and excavation constraints; Determine a new shield deviation correction path planning curve according to the shield posture deviation correction path planning parameter equation, and perform tunneling according to the new shield deviation correction path planning curve; A grouped cylinder pressure prediction model is established to predict the pressure decision value of each grouped cylinder based on the determined shield deviation correction path planning curve; the grouped cylinder pressure prediction model is expressed as follows: ； Where, is the sample size, Representation sample Parameter value, Before Decision trees are used to analyze the samples The predicted value of Indicates the t Decision tree model The complexity of is the loss function, is the objective function.

2. A shield attitude autonomous adjustment control method according to claim 1, characterized in that: The following steps are also included: Build an intelligent algorithm based PID Control model, build and shield machine PLC communication relationship; The pressure decision value is transmitted to PID The control model is used to control the shield machine's grouped cylinder proportional valves and relief valves to achieve autonomous control of the shield machine's posture.

3. The shield machine posture autonomous adjustment control method according to claim 1, characterized in that: The key excavation parameters include at least: A Cylinder pressure, B Cylinder pressure, C Cylinder pressure, D Cylinder pressure, E Cylinder pressure, F Cylinder pressure, driving thrust, propulsion speed, cutterhead torque, cutterhead speed, penetration, shield tail gap, grouting volume, grease injection volume, slurry inflow, slurry outflow, excavation chamber pressure and air cushion chamber pressure; The shield posture parameters include at least: a shield head horizontal deviation value, a shield head vertical deviation value, a shield tail horizontal deviation value and a shield tail vertical deviation value.

4. The shield machine posture autonomous adjustment control method according to claim 1, characterized in that: The preprocessing process of the data parameters of the same magnitude includes: The key excavation parameters and shield attitude parameters are partitioned, cleaned and eliminated to obtain valid parameter data; The effective parameter data are locally dimensionless processed to make them at the same order of magnitude.

5. The shield machine posture autonomous adjustment control method according to claim 1, characterized in that: The importance score The calculation formula is as follows: ； Where, Represents a node, represents the number of all trees, express m The number of non-leaf nodes in a tree, Indicates the m The first j The partitioning features of non-leaf nodes, is the loss function, 、 Respectively indicate the m The first j The sum of the first-order derivatives and the sum of the second-order derivatives of all samples on non-leaf nodes, 、 Respectively indicate the m On the tree j The sum of the first-order derivatives on the left node L and the first-order derivatives on the right node R of the non-leaf nodes, 、 Respectively indicate the m On the tree j The sum of the second-order derivatives on the left node L and the right node R of the non-leaf nodes, represents the hyperparameter of the regularization term.

6. The shield machine posture autonomous adjustment control method according to claim 1, characterized in that: The tunneling constraints include: constraints on the horizontal deviation of the shield head, constraints on the vertical deviation of the shield head, constraints on the horizontal deviation of the shield tail, and constraints on the vertical deviation of the shield tail; The predicted parameter values include: shield head horizontal deviation predicted value , shield head vertical deviation prediction value , shield tail horizontal deviation prediction value and the predicted value of shield tail horizontal deviation ; Correspondingly, the analysis process of the judgment result is: if the judgment formula is satisfied, the judgment result is that the excavation requirements are met; otherwise, the judgment result is that the excavation requirements are not met; The judgment formula is expressed as follows: ； Where, 、 、 and They are the standard value of horizontal deviation of the shield head, the standard value of vertical deviation of the shield head, the standard value of horizontal deviation of the shield tail and the standard value of horizontal deviation of the shield tail.

7. The shield machine posture autonomous adjustment control method according to claim 1, characterized in that: The process of determining the pressure decision value is as follows: taking the shield posture parameters on the determined shield correction path planning curve as input, and using the grouped cylinder pressure prediction model to obtain the pressure decision value of each grouped cylinder.

8. The shield machine posture autonomous adjustment and control method according to claim 6, characterized in that: The process of establishing the shield posture correction path planning parameter equation is as follows: Obtain current posture information, including the axis coordinates of the shield tunnel and the minimum turning radius of the shield machine , Minimum turning radius of the tunnel , shield tail gap , segment length and cylinder stroke difference; Taking the shortest path as the goal, the minimum correction curvature radius is calculated using the following formula: : ； Where, is the allowable correction radius of the shield tail gap, and its calculation formula is: ; is the minimum correction radius allowed by the cylinder stroke difference, and its calculation formula is: , is the radius of the thrust cylinder, The distance from the cutterhead to the shield tail; Based on the minimum correction curvature radius , the shield posture correction path planning parameter equation that meets the tunneling constraints is constructed and expressed as: ； ； Where, 、 are the starting and ending coordinates of the correction curve, is the coordinate of the fitting point after the end point, is the center coordinate of the shield tail, is the curvature radius of the correction curve, 、 The equations are The first and second derivatives of 、 、 and d are coefficients, Indicates that it is restricted.

9. A shield machine attitude autonomous adjustment control system, used to implement the shield machine attitude autonomous adjustment control method according to any one of claims 1 to 8, characterized in that: include: The data acquisition module is configured to obtain key tunneling parameters and shield attitude parameters during tunneling of the shield machine; The data processing module is configured to perform data preprocessing on key tunneling parameters and shield attitude parameters to obtain data parameters of the same magnitude; and to screen out relevant tunneling parameters based on the data parameters of the same magnitude, wherein the relevant tunneling parameters have a predetermined degree of correlation with the shield attitude parameters; The intelligent decision-making module is configured to construct a shield machine attitude prediction model, taking relevant tunneling parameters as input and shield machine attitude parameters as output, and using the shield machine attitude prediction model to predict the tunneling attitude of the shield machine within a predetermined range in the future; The predicted parameter values of the tunneling prediction posture are analyzed in correspondence with the tunneling constraints to obtain a judgment result; based on the judgment result, a shield correction path planning curve is determined; a group cylinder pressure prediction model is established, and the pressure decision value of each group cylinder is predicted based on the determined shield correction path planning curve; Autonomous control module, is set to build an intelligent algorithm based PID Control model, build and shield machine PLC Communication relationship; transmitting the pressure decision value to the PID control model for controlling the shield machine group cylinder proportional valve and relief valve to achieve autonomous control of the shield posture.

Images