Pipe jacking machine cutter head for cutting reinforced concrete latticed column and cutting parameter optimization method

Abstract

The invention discloses a pipe jacking machine cutter head for cutting a reinforced concrete latticed column and a cutting parameter optimization method. The pipe jacking machine cutter head comprises a cutter head body, four double-edge inserted tooth hobs, two double-edge smooth surface hobs, two dish-shaped three-edge hobs, fifteen shell cutters and twenty-seven scrapers, wherein the double-edge smooth hobs and the double-edge inserted hobs are distributed at intervals along the longitudinal central axis of the cutterhead, the two double-edge smooth hobs are arranged in the middle, and the four double-edge inserted hobs are arranged on the two sides; the dish-shaped three-blade hobs are arranged on the two sides of the transverse central axis of the cutterhead, the scrapers are symmetrically and radially arranged at intervals along the two sides of the longitudinal central axis of the cutterhead, and the shell cutters are arranged in track intervals of the adjacent hobs. The problems of power adaptability and efficient cutting of the small pipe jacking machine under the working condition of cutting the reinforced concrete latticed column are effectively solved.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering technology, and in particular to a cutter head for a pipe jacking machine for cutting reinforced concrete lattice columns and a method for optimizing cutting parameters. Background Technology

[0002] With the development of underground space and the densification of underground networks, it is becoming increasingly common for newly constructed tunnels to encounter existing buildings and structures (piles, diaphragm walls). For large-section cut-and-cover tunnels, pipe jacking is often used as an advanced support to control deformation. In the process of dense pipe jacking construction, small steel pipe jacking machines are the main construction equipment. However, due to the relatively low rated thrust and torque parameters of small pipe jacking machines, direct cutting is extremely difficult when encountering rigid obstacles such as reinforced concrete piles, lattice columns, or diaphragm walls. According to on-site construction surveys, when encountering rigid obstacles such as reinforced concrete piles, lattice columns, or diaphragm walls, manual demolition is still the primary method, resulting in low efficiency and high construction risks. In addition, in the field of underground space construction, although tunnel boring machines (TBMs) can effectively cut through existing reinforced concrete piles or diaphragm walls due to their greater power performance and cutter modifications, the significant difference in power performance between TBMs and small steel pipe jacking machines makes it impossible to use the experience of TBM cutterheads and cutting parameters to improve small pipe jacking machines. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide a cutter head for a pipe jacking machine and a method for optimizing cutting parameters when cutting reinforced concrete lattice columns, so as to effectively solve the problems of power adaptability and efficient cutting of small pipe jacking machines when cutting reinforced concrete lattice columns.

[0004] Technical Solution: To achieve the above objectives, the pipe jacking machine cutterhead of the present invention includes a cutterhead and four double-edged toothed roller cutters, two double-edged smooth roller cutters, two disc-shaped three-edged roller cutters, 15 shell cutters, and 27 scrapers located on the cutterhead. The double-edged smooth roller cutters and double-edged toothed roller cutters are distributed at intervals along the longitudinal central axis of the cutterhead, with the two double-edged smooth roller cutters arranged in the center and the four double-edged toothed roller cutters arranged on both sides. The disc-shaped three-edged roller cutters are arranged on both sides of the transverse central axis of the cutterhead, the scrapers are arranged symmetrically and radially at intervals along both sides of the longitudinal central axis of the cutterhead, and the shell cutters are arranged within the trajectory spacing of adjacent roller cutters.

[0005] Optionally, the cutter head is a right-angle cutter head structure, which consists of a radial spoke structure, a small closed panel structure, and disc surface connecting ribs.

[0006] Optionally, the shell cutter is provided with drainage and wear-resistant plates symmetrically arranged on both sides.

[0007] Optionally, a flow-guiding wear-resistant plate is provided between two adjacent scrapers.

[0008] Optionally, 15 equally spaced concentric circles are drawn with the center of the cutter head as the center, and one shell cutter is arranged on each concentric circle. With any central axis on the cutter head as the reference line, the circumferential angle of each shell cutter relative to the reference line is different.

[0009] Optionally, the disc-shaped three-edged hob is placed at a 150° angle to the cutter head surface.

[0010] Optionally, the cutting edge height of the double-edged toothed hob and the double-edged smooth hob is 70mm, the cutting edge height of the shell cutter is 60mm, and the cutting edge height of the scraper is 45mm.

[0011] The cutting parameter optimization method of the present invention includes the following steps: (1) A simulation test of cutting reinforced concrete lattice columns was conducted using a small steel pipe curtain jacking machine with the cutterhead described above. In the first stage, the cutting progress was a certain distance, and the test data for the first stage were collected. The test data for the first stage included the advance speed, cutterhead speed, rotation angle, inclination angle, soil pressure, jacking machine thrust, and jacking machine cutterhead torque. In the second stage, the cutting progress was a certain distance, and the test data for the second stage were collected. The test data for the second stage included the advance speed, cutterhead speed, rotation angle, inclination angle, and soil pressure. The rotation angle refers to the rotation angle of the jacking machine around its axis, and the inclination angle refers to the angle between the axis of the jacking machine and the horizontal plane. (2) Construct the Transformer model and train it using the experimental data from the first phase; (3) Determine the advance speed of the pipe jacking machine and the speed of the cutter head of the pipe jacking machine The range of values, , Randomly generate m sets of candidate cutting parameters within the value range. Each set of candidate cutting parameters is used as the active cutting parameters set at the initial moment of the second stage. These parameters, along with the rotation angle, inclination angle, and soil pressure collected at the initial moment of the second stage, and historical test data from N moments prior to the initial moment of the second stage (including propulsion speed, cutterhead speed, rotation angle, inclination angle, and soil pressure), are input into the trained Transformer model to predict m sets of pipe jacking machine loads at the initial moment of the second stage. The pipe jacking machine loads include the pipe jacking machine thrust. Torque of the pipe jacking machine cutter head ; (4) Calculate the roof deformation using m sets of pipe jacking machine loads, and remove pipe jacking machine loads, pipe jacking machine propulsion speeds, and pipe jacking machine cutterhead speeds that do not meet the constraints for the corresponding roof deformation. (5) Define the target evaluation function, which is the cutting efficiency evaluation function. and cutting energy consumption evaluation function ; Cutting efficiency evaluation function The penetration depth is used to evaluate the cutting efficiency of the pipe jacking machine. The cutting efficiency evaluation function is as follows: The expression is: , In the formula Indicates propulsion speed, in units of , Indicates the rotational speed of the cutter head, in units of ; Cutting energy consumption evaluation function This refers to the work done by the pipe jacking machine on the workpiece per unit time, defined as the energy consumed by the pipe jacking machine per unit time divided by the amount of soil cut per unit time. It is the cutting energy consumption evaluation function. The expression is: , In the formula This indicates that the propulsive force does work. Indicates the work done by torque. The radius of the cutterhead of the pipe jacking machine is in meters. The thrust of the pipe jacking machine is expressed in units of [unit missing]. , The torque of the cutter head is expressed in units of... ; (6) Based on the objective evaluation function, the alligator ambush optimization algorithm is used for iterative optimization. When the iteration reaches the maximum number of iterations, the iteration is terminated. The optimal pipe jacking machine propulsion speed and pipe jacking machine cutter head speed with the optimal objective evaluation function value in the population are extracted. This is the optimal combination of cutting parameters that satisfies the dual objectives of "maximizing efficiency and minimizing energy consumption" and meets the roof stability constraints. (7) Based on the optimal jacking machine advance speed, jacking machine cutterhead speed and jacking machine load at the initial moment of the second stage, predict the jacking machine load at the next moment in a time-sequential cycle. That is, m sets of candidate cutting parameters are used as the active cutting parameters set at the current moment of the second stage. They are input into the trained Transformer model together with the rotation angle, tilt angle and soil pressure collected at the current moment of the second stage and the historical test data of N moments before the current moment. The m sets of jacking machine loads at the current moment of the second stage are predicted. Repeat steps (4) to (6) to obtain the optimal jacking machine advance speed and jacking machine cutterhead speed at the current moment.

[0012] Optionally, the training of the Transformer model specifically includes the following steps: (2.1) The propulsion speed, cutterhead rotation speed, rotation angle, tilt angle, soil chamber pressure, pipe jacking machine thrust and pipe jacking machine cutterhead torque collected in the first stage are preprocessed to form a basic dataset; (2.2) Improve the existing Transformer model by incorporating soil mechanics constraints and power balance constraints; , , , , In the formula , , and These represent the total loss value, foundation prediction error, soil mechanical constraints, and power balance constraints, respectively. , , This indicates the weight of each loss relative to the total loss; For sample size; and These represent the predicted value and the actual value, respectively. To correct linear units, specifically activation functions; The diameter of the cutter head is in mm. The unit is the weight of the soil. , The cohesion of the soil, in units of , and All are bearing capacity coefficients; To indicate the speed of advancement, the unit is... , For the power efficiency of the pipe jacking machine, This refers to the rated power of the pipe jacking machine; (2.4) The basic dataset is randomly divided into training and test sets according to the proportion. The training set data is input into the Transformer model and passes through the input layer, Transformer layer and output layer in sequence to generate predicted values ​​of the pipe jacking machine thrust and cutterhead torque. The loss value is calculated based on the predicted value and the true value. Then, the weights and bias parameters of the Transformer model are updated through the backpropagation mechanism. The iteration is repeated until the preset maximum number of iterations is reached. The generalization performance of the Transformer model is evaluated using the test set. The core evaluation index is the coefficient of determination. , This yields the trained Transformer model.

[0013] Optionally, the formula for calculating the deformation of the top plate is: , , The constraints are: , in The amount of roof deformation calculated based on the thrust of the pipe jacking machine. The amount of roof deformation calculated based on the cutterhead torque of the pipe jacking machine. , , , , , These are the fitting coefficients.

[0014] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: Targeting engineering scenarios where pipe jacking machines cut rigid obstacles such as lattice columns, this invention addresses the issue from two core dimensions: the power constraints of the pipe jacking machine and the structural characteristics of the lattice column. By optimizing the layout design of the cutting tools, it effectively solves two key problems: power adaptability of the pipe jacking machine under such working conditions and efficient cutting of rigid obstacles. Furthermore, by specifically designing the installation position of the wear-resistant plate, this invention constructs a collaborative protection mechanism adapted to scraper and shell cutter cutting operations, effectively suppressing the pulling phenomenon of the cutting tools on the reinforcing steel during cutting, ensuring the continuity of cutting operations and the stability of tool operation. Finally, this invention proposes a multi-dimensional collaborative cutting parameter optimization method integrating field experiments, numerical simulation, machine learning, and multi-objective optimization, constructing a complete technical chain of "data support - model prediction - parameter optimization," providing a scientific path for the accurate acquisition of reasonable cutting parameters. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the cutter head structure in this invention; Figure 2 This is a schematic diagram of the arrangement of the disc-shaped three-flute hob in this invention; Figure 3 This is a schematic diagram showing the arrangement of the hobbing cutter and the shell cutting cutter in this invention; Figure 4 This is a schematic diagram showing the arrangement of the wear-resistant plate and the scraper in this invention; Figure 5 This is a schematic diagram showing the arrangement of the wear-resistant plate and the shell cutter in this invention; Figure 6 This is a schematic diagram of the Transformer model in this invention; Figure 7 This is a flowchart illustrating the cutting parameter optimization method of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0017] Example 1: As Figure 1As shown, the present invention provides a cutter head for a pipe jacking machine for cutting reinforced concrete lattice columns, including a cutter head 1 and four double-edged toothed hobs 4, two double-edged smooth hobs 5, two disc-shaped triple-edged hobs 6, 15 shell cutters 7, and 27 scrapers 8 located on the cutter head. The double-edged smooth hobs 5 and double-edged toothed hobs 4 are distributed at intervals along the longitudinal central axis of the cutter head, with the two double-edged smooth hobs 5 arranged in the center and the four double-edged toothed hobs 5 arranged on both sides. The disc-shaped triple-edged hobs 6 are arranged on both sides of the transverse central axis of the cutter head. The scrapers 8 are arranged symmetrically and radially at intervals on both sides of the longitudinal central axis of the cutter head. The shell cutters 7 are arranged in 15 concentric circles with equal spacing around the center of the cutter head, with one shell cutter arranged on each concentric circle.

[0018] In one embodiment, the cutter head 1 is a right-angled cutter head structure, which consists of a radial spoke structure 2, a small closed panel structure 3, and a disc surface connecting rib 10.

[0019] In one embodiment, the shell cutter 7 is symmetrically provided with drainage and wear-resistant plates 9 on both sides, such as... Figure 5 As shown.

[0020] In one embodiment, a flow-guiding wear-resistant plate 9 is provided between two adjacent scrapers 8, such as... Figure 4 As shown.

[0021] In one embodiment, with any central axis on the cutter head as the reference line, each shell cutter has a different circumferential angle relative to the reference line.

[0022] In one embodiment, the disc-shaped three-flute hob 6 is placed at a 150° angle to the cutter head surface, such as... Figure 2 As shown.

[0023] In one embodiment, the cutting edge height of the double-edged toothed hob 4 and the double-edged smooth hob 5 is 70mm, the cutting edge height of the shell cutter is 60mm, and the cutting edge height of the scraper is 45mm.

[0024] The right-angle cutterhead structure, with its 90° vertical transition at the edges, ensures that the cutterhead edge makes positive contact with the pile body during cutting, with the contact direction approaching the cutterhead axis; at this time, the lateral component of the soil reaction force... The force exerted by the arc-shaped cutter head is negligible and unlikely to deviate; the edge of the arc-shaped cutter head structure has an arc transition, and the contact between the edge of the cutter head and the pile body during cutting is an oblique line contact, and the reaction force of the pile body on the cutter head is negligible. It can be decomposed into axial force and lateral component The axial force is in the direction of propulsion, the lateral component is perpendicular to the direction of propulsion, and the inclination angle of the curved surface is... This will result in a lateral component force. Furthermore, due to the different inclination angles at various points of the arc, the magnitude and direction of the lateral force will change periodically when the cutter head rotates. When the lateral force cannot be balanced by the cutter head structure or the propulsion system, it will cause the cutter head to deviate to the side.

[0025] The shell cutter 7 consists of 15 equally spaced concentric circles centered on the cutter head. One shell cutter is placed on each concentric circle. With any central axis on the cutter head as the reference line, the circumferential angle between each shell cutter and the reference line is different. The circumferential angles can be 0°, 24°, 48°, 72°, 96°, 120°, 144°, 168°, 192°, 216°, 240°, 264°, 288°, 312°, and 336°.

[0026] Addressing the current technical limitations where existing experiments primarily focus on cutting rigid obstacles with tunnel boring machines (TBMs), this invention innovatively proposes a power-adaptive stacked cutting system to overcome the core constraint that the rated power of the tunnel jacking machine is far lower than that of the TBM. Based on the design concept of "energy hierarchical allocation," a hierarchical cutting structure of "primary crushing - secondary crushing - trimming and slag removal" is constructed through differentiated parameter configurations of cutter blade heights: 70mm for the roller cutter, 60mm for the shell cutter, and 45mm for the scraper cutter.

[0027] like Figure 3 As shown, to achieve efficient crushing of reinforced concrete piles, this invention adopts a functionally graded cutter synergistic arrangement strategy: double-edged smooth cutters 5 and double-edged toothed cutters 4 are longitudinally distributed in a hierarchical manner along the spoke structure 2, with two double-edged smooth cutters 5 centrally positioned and four double-edged toothed cutters 5 arranged on both sides. Utilizing the high pressure advantage of the double-edged smooth cutters 5, they can be pressed into the pile body first, generating a core crushing pit and radial cracks in the reinforced concrete structure. The toothed cutters 4 arranged on both sides utilize their wedge-shaped alloy teeth to efficiently grasp and expand the formed cracks, thoroughly crushing and peeling off large, loose pieces of concrete. Based on the mechanical characteristics that the cracks generated when the cutters compress the medium are mainly concentrated within a 2-3 cm range on both sides of the blade, to ensure that the lateral cracks in the area between adjacent cutter tips can penetrate each other and form a continuous through-damage zone, the cutter trajectory lines are evenly distributed at 60 mm intervals. The disc-shaped three-flute hob 6 is placed on both sides of the cutter head at a 150° angle to the cutter head surface, so that the effective cutting outer diameter of the cutter head is greater than the size limit of the cutter head body, thus realizing the super-boundary expansion of the cutting surface.

[0028] To precisely adapt to the unique structure and failure characteristics of reinforced concrete lattice columns, this invention employs a precise trajectory gap layout scheme, placing the shell-shaped cutter 7 within the trajectory spacing of adjacent roller cutters to construct a targeted cutting system. This design is based on in-depth analysis and experimental verification of the differentiated mechanical responses of lattice columns, concrete, and reinforcing steel: In the initial stage of the experiment, without the shell-shaped cutter, it was found that only the lattice column within the roller cutter trajectory coverage area could be ground down, while other areas were difficult to break effectively; pure concrete, under the extrusion of the roller cutters, would form through-cracks along the outer edge of the trajectory and peel off in blocks; the reinforcing steel, after being cut, was in a discrete state and could not continuously hinder the advancement of the pipe jacking machine. As a columnar continuous load-bearing structure, the lattice column has high overall rigidity and is difficult to form through-cracks through the extrusion of the roller cutters. Its four sides are connected by rigid nodes to form a closed frame, and the overall structural integrity is only completely destroyed when the pipe jacking machine is fully advanced to its axial end. Therefore, it is necessary to fill the gap in the roller cutter trajectory with a shell cutter to achieve efficient full-range cutting of the lattice column. This will break through the propulsion bottleneck caused by the continuous structure of the lattice column from the perspective of cutting mechanism, and provide key technical support for small pipe jacking machines to achieve efficient and stable cutting of lattice columns.

[0029] Scrapers 8 are arranged radially at intervals on both sides of the spoke structure 2, 27 of them, to clean up the debris generated after the hobbing cutter and shell cutter cutter cut, to avoid debris accumulation and secondary friction, and to ensure the flatness of the cutting surface.

[0030] During the experiment, phenomena such as scraper pulling on reinforcing bars and momentary jamming of the cutter head occurred. To solve the technical bottleneck of "cutter-reinforcing bar coupling interference" when the pipe jacking machine cuts reinforced concrete lattice columns, this invention proposes a collaborative protection design with drainage wear-resistant plates 9. By symmetrically welding drainage wear-resistant plates 9 on both sides of the shell cutter 7, a reinforcing bar guiding and protection system is constructed. From the perspective of mechanical action mechanism, if the reinforcing bar directly contacts one side of the vertical cutting surface of the shell cutter, the "rigid collision" effect can easily cause momentary jamming of the cutter head, disrupting the cutting continuity. However, by adopting the drainage wear-resistant plates 9 with a triangular structure, the hypotenuse can construct a "gradient guiding channel," allowing the contacting reinforcing bar to achieve a "sliding-detachment" dynamic response along the hypotenuse. By changing the contact form between the reinforcing bar and the cutter, the risk of the shell cutter pulling and trapping the reinforcing bar is avoided from the root. At the same time, to address the problem of reinforcing bar easily getting stuck in the gap between the scrapers, this invention synchronously welds drainage wear-resistant plates 9 between two adjacent scrapers, using the solid structure of the wear-resistant plates to fill the gap between the scrapers, forming a "physical barrier layer." When the free reinforcing bar contacts the cutter head, the barrier can directly block the intrusion path of the reinforcing bar into the scraper gap, avoid the "entanglement-accumulation" effect of the reinforcing bar in the gap, significantly reduce the probability of coupling interference between the reinforcing bar and the scraper, and provide structural protection support for the continuous and stable cutting of the cutter head.

[0031] Example 2: During pipe jacking machine construction, the selection of cutting parameters mainly relies on engineering experience, lacking theoretical standards. Therefore, this invention addresses this issue by combining numerical simulation, machine learning methods, and multi-objective optimization methods in pipe jacking machine cutting tests. It establishes a multi-objective tunneling parameter optimization method based on a pipe jacking machine load (pipe jacking machine thrust and cutterhead torque) prediction model, with the dual objectives of maximizing pipe jacking machine efficiency and minimizing its energy consumption, and the constraint of keeping the vertical displacement of the roof within control values. This provides theoretical support and technical solutions for setting cutting parameters.

[0032] like Figure 7 As shown, the cutting parameter optimization method of the present invention includes the following steps: (1) A simulation test of cutting reinforced concrete lattice columns was conducted using a small steel pipe curtain jacking machine equipped with stacked right-angle cutterheads. The first stage cutting advance was 500 mm, and the test data for the first stage were collected. The test data for the first stage included the advance speed, cutterhead rotation speed, rotation angle, inclination angle, soil pressure, jacking machine thrust, and jacking machine cutterhead torque. The second stage cutting advance was 500 mm, and the test data for the second stage were collected. The test data for the second stage included the advance speed, cutterhead rotation speed, rotation angle, inclination angle, and soil pressure. The rotation angle refers to the rotation angle of the jacking machine around its axis, and the inclination angle refers to the angle between the axis of the jacking machine and the horizontal plane. (2) Construct the Transformer model and train it using the experimental data from the first phase, such as... Figure 6 As shown, the specific steps include the following: (2.1) The propulsion speed, cutterhead rotation speed, rotation angle, tilt angle, soil chamber pressure, pipe jacking machine thrust and pipe jacking machine cutterhead torque collected in the first stage are preprocessed to form a basic dataset; (2.2) Improve the existing Transformer model by incorporating soil mechanics constraints and power balance constraints; , , , , In the formula , , and These represent the total loss value, foundation prediction error, soil mechanical constraints, and power balance constraints, respectively. , , This indicates the weight of each loss relative to the total loss; For sample size; and These represent the predicted value and the actual value, respectively. To correct linear units, specifically activation functions; The diameter of the cutter head is in mm. The unit is the weight of the soil. , The cohesion of the soil, in units of , and All are bearing capacity coefficients; To indicate the speed of advancement, the unit is... , For the power efficiency of the pipe jacking machine, This refers to the rated power of the pipe jacking machine; (2.3) The inputs to the Transformer model are: cutterhead rotation speed, propulsion speed, rotation angle, inclination angle, and soil chamber pressure. To eliminate the interference of different parameter dimensions on model training, the max-min normalization method is used to normalize the raw data collected in the experiment, mapping the values ​​of each parameter to a uniform scale. The basic dataset is randomly divided into training and test sets according to the proportion. The training set data is input into the Transformer model and passes through the input layer, Transformer layer, and output layer in sequence to generate predicted values ​​of the pipe jacking machine thrust and cutterhead torque. The loss value is calculated based on the predicted value and the true value. Then, the weights and bias parameters of the Transformer model are updated through the backpropagation mechanism. The iteration is repeated until the preset maximum number of iterations is reached. The generalization performance of the Transformer model is evaluated using the test set. The core evaluation index is the coefficient of determination. , This yields the trained Transformer model.

[0033] (3) Determine the advance speed of the pipe jacking machine and the speed of the cutter head of the pipe jacking machine The range of values, , Randomly generate m sets of candidate cutting parameters within the value range. Each set of candidate cutting parameters is used as the active cutting parameters set at the initial moment of the second stage. These parameters, along with the rotation angle, inclination angle, and soil pressure collected at the initial moment of the second stage, and historical test data from N moments prior to the initial moment of the second stage (including propulsion speed, cutterhead speed, rotation angle, inclination angle, and soil pressure), are input into the trained Transformer model to predict m sets of pipe jacking machine loads at the initial moment of the second stage. The pipe jacking machine loads include the pipe jacking machine thrust. Torque of the pipe jacking machine cutter head ; (4) Calculate the roof deformation by predicting the m sets of pipe jacking machine loads, and remove the pipe jacking machine loads, pipe jacking machine propulsion speeds and pipe jacking machine cutterhead speeds that do not meet the constraints for the corresponding roof deformation. The stability constraint of the top plate needs to be established through numerical simulation to establish the influence of a large number of pipe jacking machines cutting reinforced concrete lattice columns on the upper top plate. The variables are thrust and torque, and the output is the deformation of the top plate structure. The relationship between thrust, torque and top plate structure deformation is quantified, the fitting coefficient is obtained, and the formula for calculating the deformation of the top plate is obtained through fitting. The formula for calculating the deformation of the top plate is: , , The constraints are: , in The amount of roof deformation calculated based on the thrust of the pipe jacking machine. The amount of roof deformation calculated based on the cutterhead torque of the pipe jacking machine. For control values, , , , , , These are the fitting coefficients.

[0034] (5) Define the target evaluation function, which is the cutting efficiency evaluation function. and cutting energy consumption evaluation function ; Cutting efficiency evaluation function The penetration depth is used to evaluate the cutting efficiency of the pipe jacking machine. The cutting efficiency evaluation function is as follows: The expression is: , In the formula Indicates propulsion speed, in units of , Indicates the rotational speed of the cutter head, in units of ; Cutting energy consumption evaluation function This refers to the work done by the pipe jacking machine on the workpiece per unit time, defined as the energy consumed by the pipe jacking machine per unit time divided by the amount of soil cut per unit time. It is the cutting energy consumption evaluation function. The expression is: , In the formula This indicates that the propulsive force does work. Indicates the work done by torque. The radius of the cutterhead of the pipe jacking machine is in meters. The thrust of the pipe jacking machine is expressed in units of [unit missing]. , The torque of the cutter head is expressed in units of... .

[0035] (6) Based on the objective evaluation function, the alligator ambush optimization algorithm is used for iterative optimization. When the iteration reaches the maximum number of iterations, the iteration is terminated. The optimal jacking machine propulsion speed and jacking machine cutterhead speed with the optimal objective evaluation function value in the population are extracted. These are the optimal jacking machine propulsion speed and jacking machine cutterhead speed at the initial moment of the second stage. This is the optimal combination of cutting parameters that satisfies the dual objectives of "maximizing efficiency and minimizing energy consumption" and meets the roof stability constraints. This realizes the continuous dynamic fine-tuning suggestion of the cutting parameters of the jacking machine during the working stage.

[0036] (7) Based on the optimal jacking machine advance speed, jacking machine cutterhead speed and jacking machine load at the initial moment of the second stage, predict the jacking machine load at the next moment in a time-sequential cycle. That is, m sets of candidate cutting parameters are used as the active cutting parameters set at the current moment of the second stage. They are input into the trained Transformer model together with the rotation angle, tilt angle and soil pressure collected at the current moment of the second stage and the historical test data of N moments before the current moment. The m sets of jacking machine loads at the current moment of the second stage are predicted. Repeat steps (4) to (6) to obtain the optimal jacking machine advance speed and jacking machine cutterhead speed at the current moment.

[0037] This invention uses machine learning to predict the load of a pipe jacking machine. The main active cutting parameters of the pipe jacking machine are the propulsion speed and the cutterhead rotation speed. The propulsion speed and the cutterhead rotation speed are key cutting parameters that are controlled by the operator. The passive cutting parameters generated during the cutting process are the pipe jacking machine thrust and the cutterhead torque. Therefore, it is necessary to train a Transformer model using pipe jacking machine test data to predict the load of the pipe jacking machine.

[0038] Experimental data was collected and cleaned: Key cutting parameters such as cutterhead rotation speed, pipe jacking machine thrust, cutterhead torque, feed speed, rotation angle, tilt angle, and soil chamber pressure were collected in real time using the equipment's built-in PLC sensing system. This data was then synchronously transmitted to a local database for structured storage, providing a foundational dataset for subsequent Transformer model training. For missing and outlier values ​​in the dataset, mean interpolation was used for data repair, effectively ensuring the accuracy and reliability of subsequent model training and data analysis.

[0039] Parameter selection and preprocessing: Taking into account the core influencing factors of pipe jacking machine cutting operation, the input of the Transformer model is determined to be: cutterhead rotation speed, feed speed, rotation angle, tilt angle and soil chamber pressure. In order to eliminate the interference of different parameter dimensions on model training, the max-min standardization method is used to normalize the raw data collected in the experiment and map the values ​​of each parameter to a uniform scale.

[0040] Improvements to the Transformer model: Incorporate soil mechanics constraints and power balance constraints to make the predicted load values ​​of the pipe jacking machine conform to physical laws, thereby enhancing credibility and interpretability; Model Training and Testing: The dataset is randomly divided into training and testing sets in a 7:3 ratio. The training data is sequentially processed through the input layer, Transformer layer, and output layer to generate predicted values ​​for the pipe jacking machine thrust and cutterhead torque. The loss value is calculated based on the predicted and actual values, and the model weights and bias parameters are updated through backpropagation. This process is repeated until the preset maximum number of iterations is reached. The generalization performance of the model is evaluated using the test set, with the coefficient of determination being the core performance indicator. It is necessary to ensure It is then embedded as a reliable Transformer module into subsequent multi-objective optimization models.

[0041] This invention constructs a multi-objective optimization model based on the crocodile ambush optimization algorithm, specifically as follows: The core mechanism of the crocodile ambush optimization algorithm: a1 Population Initialization: The parameters to be optimized (propulsion speed and cutterhead rotation speed) are encoded as "the position of the crocodile individual". Multiple individuals are randomly generated within the parameter value range. At the same time, each individual is assigned an initial energy value. The initial fitness (objective function value) is calculated and the optimal individual is selected as the initial leader. a2 Energy Dynamics Simulation: Simulates the energy consumption of a crocodile during an ambush, with each individual's energy decreasing as it moves further. When energy is plentiful, individuals focus on "global exploration," while when energy is insufficient, they shift to "local exploitation," balancing search diversity and convergence. , In the formula , For the first The energy of an individual at time t+1; Energy consumption coefficient; For the first The distance traveled by each individual; a3 ambush point selection: The probability of an individual becoming a "leader" is calculated based on fitness (objective function value). Individuals with better fitness (lower objective function value) are more likely to become leaders, guiding the population to move towards regions with better parameters. , In the formula No. Individual, and These are the probability and the objective function, respectively. A4 attack timing decision: Monitor the magnitude of change in individual fitness (objective function value), and trigger adaptive movement when the change exceeds a threshold (≤0.1); The a5 individual position update combines "leader guidance (development)" and "random perturbation (exploration)". , In the formula , , For time t+1 and t, the first... The best individual; Controlling the leader's influence; Control the intensity of exploration; It is a random number between (0,1).

[0042] The a6 iteration terminates and repeats the above steps until the maximum number of iterations is reached, finally outputting the optimal position (i.e., the optimal parameters) in the population.

[0043] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. All components not explicitly stated in this embodiment can be implemented using existing technology.

Claims

1. A cutterhead for a pipe jacking machine used for cutting reinforced concrete lattice columns, characterized in that, It includes a cutter head (1) and four double-edged toothed hobs (4), two double-edged smooth hobs (5), two disc-shaped three-edged hobs (6), 15 shell cutters (7) and 27 scrapers (8) located on the cutter head. The double-edged smooth hobs (5) and double-edged toothed hobs (4) are distributed at intervals along the longitudinal central axis of the cutter head. The two double-edged smooth hobs (5) are arranged in the center and the four double-edged toothed hobs (5) are arranged on both sides. The disc-shaped three-edged hobs (6) are arranged on both sides of the transverse central axis of the cutter head. The scrapers (8) are arranged symmetrically and radially at intervals on both sides of the longitudinal central axis of the cutter head. The shell cutters (7) are arranged within the trajectory spacing of adjacent hobs.

2. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, The cutter head (1) is a right-angle cutter head structure, which is composed of a radial spoke structure (2), a small closed panel structure (3) and a disc surface connecting rib (10).

3. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, The shell cutter (7) is symmetrically provided with drainage wear-resistant plates (9) on both sides.

4. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, A flow-guiding wear-resistant plate (9) is provided between two adjacent scrapers (8).

5. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, Draw 15 concentric circles with equal spacing around the center of the cutter head. Place one shell cutter on each concentric circle. Take any central axis on the cutter head (1) as the reference line. The circumferential angle of each shell cutter (7) relative to the reference line is different.

6. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, The disc-shaped three-edged hob (6) is placed at a 150° angle to the cutter head surface.

7. The cutterhead of the pipe jacking machine according to claim 1, characterized in that, The blade height of the double-edged toothed hob (4) and the double-edged smooth hob (5) is 70mm, the blade height of the shell cutter is 60mm, and the blade height of the scraper is 45mm.

8. A method for optimizing cutting parameters in cutting reinforced concrete lattice columns, characterized in that, Includes the following steps: (1) A simulation test of cutting reinforced concrete lattice columns was conducted using a small steel pipe curtain jacking machine with a cutterhead as described in any one of claims 1 to 7. In the first stage, the cutting advance was a certain distance, and the test data for the first stage were collected. The test data for the first stage included the advance speed, cutterhead rotation speed, rotation angle, inclination angle, soil pressure, jacking machine thrust, and jacking machine cutterhead torque. In the second stage, the cutting advance was a certain distance, and the test data for the second stage were collected. The test data for the second stage included the advance speed, cutterhead rotation speed, rotation angle, inclination angle, and soil pressure. The rotation angle refers to the rotation angle of the jacking machine around its axis, and the inclination angle refers to the angle between the axis of the jacking machine and the horizontal plane. The cutterhead of the jacking machine; (2) Construct the Transformer model and train it using the experimental data from the first phase; (3) Determine the advance speed of the pipe jacking machine and the speed of the cutter head of the pipe jacking machine The range of values, , Randomly generate m sets of candidate cutting parameters within the value range. Each set of candidate cutting parameters is used as the active cutting parameters set at the initial moment of the second stage. These parameters, along with the rotation angle, inclination angle, and soil pressure collected at the initial moment of the second stage, and historical test data from N moments prior to the initial moment of the second stage (including propulsion speed, cutterhead speed, rotation angle, inclination angle, and soil pressure), are input into the trained Transformer model to predict m sets of pipe jacking machine loads at the initial moment of the second stage. The pipe jacking machine loads include the pipe jacking machine thrust. Torque of the pipe jacking machine cutter head ; (4) Calculate the roof deformation using m sets of pipe jacking machine loads, and remove pipe jacking machine loads, pipe jacking machine propulsion speeds, and pipe jacking machine cutterhead speeds that do not meet the constraints for the corresponding roof deformation. (5) Define the target evaluation function, which is the cutting efficiency evaluation function. and cutting energy consumption evaluation function ; Cutting efficiency evaluation function The penetration depth is used to evaluate the cutting efficiency of the pipe jacking machine. The cutting efficiency evaluation function is as follows: The expression is: ， In the formula Indicates propulsion speed, in units of , Indicates the rotational speed of the cutter head, in units of ; Cutting energy consumption evaluation function This refers to the work done by the pipe jacking machine on the workpiece per unit time, defined as the energy consumed by the pipe jacking machine per unit time divided by the amount of soil cut per unit time. It is the cutting energy consumption evaluation function. The expression is: ， In the formula This indicates that the propulsive force does work. Indicates the work done by torque. The radius of the cutterhead of the pipe jacking machine is in meters. The thrust of the pipe jacking machine is expressed in units of [unit missing]. , The torque of the cutter head is expressed in units of... ; (6) Based on the objective evaluation function, the alligator ambush optimization algorithm is used for iterative optimization. When the iteration reaches the maximum number of iterations, the iteration is terminated. The optimal pipe jacking machine propulsion speed and pipe jacking machine cutter head speed with the optimal objective evaluation function value in the population are extracted. This is the optimal combination of cutting parameters that satisfies the dual objectives of "maximizing efficiency and minimizing energy consumption" and meets the roof stability constraints. (7) Based on the optimal jacking machine advance speed, jacking machine cutterhead speed and jacking machine load at the initial moment of the second stage, predict the jacking machine load at the next moment in a time-sequential cycle. That is, m sets of candidate cutting parameters are used as the active cutting parameters set at the current moment of the second stage. They are input into the trained Transformer model together with the rotation angle, tilt angle and soil pressure collected at the current moment of the second stage and the historical test data of N moments before the current moment. The m sets of jacking machine loads at the current moment of the second stage are predicted. Repeat steps (4) to (6) to obtain the optimal jacking machine advance speed and jacking machine cutterhead speed at the current moment.

9. The cutting parameter optimization method according to claim 8, characterized in that, The specific steps involved in training the Transformer model are as follows: (2.1) The propulsion speed, cutterhead rotation speed, rotation angle, tilt angle, soil chamber pressure, pipe jacking machine thrust and pipe jacking machine cutterhead torque collected in the first stage are preprocessed to form a basic dataset; (2.2) Improve the existing Transformer model by incorporating soil mechanics constraints and power balance constraints; ， ， ， ， In the formula , , and These represent the total loss value, foundation prediction error, soil mechanical constraints, and power balance constraints, respectively. , , This indicates the weight of each loss relative to the total loss; For sample size; and These represent the predicted value and the actual value, respectively. To correct linear units, specifically activation functions; This refers to the diameter of the cutter head, in mm. The unit is the weight of the soil. , The cohesion of the soil, in units of , and All are bearing capacity coefficients; To indicate the speed of advancement, the unit is... , For the power efficiency of the pipe jacking machine, This refers to the rated power of the pipe jacking machine; (2.3) The basic dataset is randomly divided into training set and test set according to the proportion. The training set data is input into the Transformer model and passes through the input layer, Transformer layer and output layer in sequence to generate predicted values ​​of pipe jacking machine thrust and cutterhead torque. The loss value is calculated based on the predicted value and the true value. Then, the Transformer model weights and bias parameters are updated through the backpropagation mechanism. The iteration is repeated until the preset maximum number of iterations is reached. The generalization performance of the Transformer model is evaluated using a test set, with the coefficient of determination as the core performance indicator. , This yields the trained Transformer model.

10. The cutting parameter optimization method according to claim 8, characterized in that, The formula for calculating the deformation of the top plate is: ， ， The constraints are: , in The amount of roof deformation calculated based on the thrust of the pipe jacking machine. The amount of roof deformation calculated based on the cutterhead torque of the pipe jacking machine. , , , , , These are the fitting coefficients.

Images