Intelligent control method and system for cryo-against-the-top through based on deep learning

Abstract

The application provides an intelligent control method and system for frozen top-to-top through based on deep learning, and relates to the technical field of deep learning. The method comprises the following steps: acquiring the parameters, images, environmental data and construction target data of a pipe jacking machine at a construction site, determining the position of the pipe jacking machine head, and generating a pushing path based on a path planning module. During the construction process, the temperature and thickness distribution of the closing section position are acquired, and the historical freezing scheme is adjusted to obtain a freezing construction scheme for the closing section. When the frozen wall reaches the preset condition, a construction scheme for the machine head is constructed based on the first information, including a disassembly, welding and waterproof construction scheme. Finally, construction is carried out based on the construction scheme for the machine head until the waterproof effect reaches the preset standard, and a reinforced concrete and pressure compensation grouting scheme for the closing section is constructed according to the construction data, and the construction is finally completed. The application significantly improves the efficiency and precision of frozen top-to-top through construction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of deep learning, in particular to an intelligent control method and system for frozen freeze through based on deep learning. BACKGROUND

[0002] In the current field of underground engineering construction, especially in the process of pipe jacking construction, accurate control of freeze through is particularly important. In the prior art, the determination of the machine head position and path planning are usually performed by relying on manual experience and simple automatic control systems. This method has significant limitations, for example, the subjectivity of manual operation leads to insufficient positioning accuracy, and inaccurate path planning often leads to construction errors, affecting the quality of the project. In addition, the utilization rate of environmental data and machine head state data during construction is low, which makes the reaction ability to unforeseen problems insufficient, ultimately causing delays in construction progress and rising costs. In order to address these problems, some existing technologies introduce simple sensor monitoring and automatic control means, but due to the lack of support for deep learning and intelligent control technology, the adaptability and decision-making ability of these methods under complex working conditions are still limited, making it difficult to meet the construction needs under high-precision and complex geological conditions.

[0003] Therefore, there is an urgent need for an intelligent control method and system for frozen freeze through based on deep learning to solve the problems. SUMMARY

[0004] The present application aims to provide an intelligent control method and system for frozen freeze through based on deep learning to improve the problems. In order to achieve the purpose, the technical solutions adopted by the present application are as follows:

[0005] In a first aspect, the present application provides an intelligent control method for frozen freeze through based on deep learning, comprising:

[0006] Obtaining first information, the first information including parameter data, image data, environmental data and construction target data of all pipe jacking machines at the construction site;

[0007] Determining the machine head position of the pipe jacking machine based on the first information, and sending the determined machine head position and the first information to the path planning module for path planning to obtain pipe jacking machine advance path information;

[0008] Based on the pipe jacking machine advance path information, the construction is carried out until the machine heads of the two corresponding pipe jacking machines reach the preset closing section position, the temperature distribution and thickness distribution of the closing section position are obtained, and the preset historical freezing scheme is adjusted based on the temperature distribution and thickness distribution of the closing section position to obtain the freezing construction scheme of the closing section position;

[0009] Construction is carried out based on the freezing construction scheme of the joint section position until the freezing wall of the joint section position reaches the preset condition, the machine head construction scheme is constructed based on the first information, and the machine head construction scheme includes a machine head disassembly scheme, a machine head shell welding scheme and a machine head waterproof construction scheme;

[0010] Based on the machine head construction scheme, construction is carried out until the waterproof effect of the machine head shell reaches a preset threshold, the joint section reinforced concrete construction scheme and the pressure compensation grouting scheme are constructed based on the first information, and construction is sequentially carried out according to the joint section reinforced concrete construction scheme and the pressure compensation grouting scheme, to obtain a final construction result.

[0011] In a second aspect, the application also provides an intelligent control system for frozen top-to-top penetration based on deep learning, comprising:

[0012] An acquisition unit is configured to acquire first information, wherein the first information includes parameter data, image data, environment data and construction target data of all pipe jacking machines at a construction site;

[0013] A planning unit is configured to determine the machine head position of the pipe jacking machine based on the first information, and send the determined machine head position and the first information to a path planning module for path planning to obtain pipe jacking machine advancing path information;

[0014] An adjustment unit is configured to carry out construction based on the pipe jacking machine advancing path information until the machine heads of two corresponding pipe jacking machines reach a preset joint section position, acquire the temperature distribution and thickness distribution of the joint section position, and adjust the preset historical freezing scheme based on the temperature distribution and thickness distribution of the joint section position to obtain a freezing construction scheme of the joint section position;

[0015] A construction unit is configured to carry out construction based on the freezing construction scheme of the joint section position until the freezing wall of the joint section position reaches a preset condition, and construct a machine head construction scheme based on the first information, wherein the machine head construction scheme includes a machine head disassembly scheme, a machine head shell welding scheme and a machine head waterproof construction scheme;

[0016] A processing unit is configured to carry out construction based on the machine head construction scheme until the waterproof effect of the machine head shell reaches a preset threshold, construct a joint section reinforced concrete construction scheme and a pressure compensation grouting scheme based on the first information, and sequentially carry out construction according to the joint section reinforced concrete construction scheme and the pressure compensation grouting scheme to obtain a final construction result.

[0017] The application has the following beneficial effects:

[0018] The application acquires the pipe jacking machine parameters, image data, environment data and construction target data of the construction site, accurately positions the machine head position of the pipe jacking machine by using a deep learning algorithm, and plans the path based on a multi-objective optimization algorithm. Through dynamic adjustment of the temperature distribution and thickness data obtained during the construction process, the freezing construction scheme of the closure section is optimized. In addition, the method also combines advanced calculation methods such as analytic hierarchy process, particle swarm optimization algorithm and Bayesian optimization algorithm to construct a complete machine head disassembly, welding, waterproof construction scheme and closure section reinforced concrete construction scheme and pressure supplementing and grouting scheme. Compared with the prior art, the application can greatly improve the construction precision and efficiency, and significantly reduce the construction risk and cost.

[0019] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the present application according to the embodiments. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those skilled in the art, other related drawings can also be obtained from these drawings without creative labor.

[0021] Figure 1 The flow chart of the intelligent control method of frozen against top penetration based on deep learning described in the embodiments of the present application;

[0022] Figure 2 The structure schematic diagram of the intelligent control system of frozen against top penetration based on deep learning described in the embodiments of the present application.

[0023] In the figure: 701, acquisition unit; 702, planning unit; 703, adjustment unit; 704, construction unit; 705, processing unit. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0025] It should be noted that similar reference numerals and letters refer to similar items throughout the accompanying drawings, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", and the like are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0026] Embodiment 1

[0027] The embodiment provides an intelligent control method for frozen top-penetrating based on deep learning.

[0028] Referring to Figure 1 , the method includes steps S1, S2, S3, S4 and S5.

[0029] Step S1, acquiring first information, the first information including parameter data, image data, environment data and construction target data of all pipe jacking machines on the construction site;

[0030] It can be understood that acquiring the first information is the basis of the entire intelligent control method. These information includes parameter data, image data, environment data and construction target data of the pipe jacking machine, covering a comprehensive set of information required in the construction process. Specifically, the parameter data of the pipe jacking machine includes the specifications, performance indicators, real-time running state of the machine, etc., which can provide a basis for subsequent head position determination and path planning. The image data captures real-time images of the pipe jacking machine and the surrounding environment through high-precision cameras and sensors, and processes and analyzes them using computer vision technology to ensure visual monitoring of key positions in the construction process.

[0031] The environmental data relates to the temperature and humidity, soil properties, groundwater level and other key environmental variables of the construction site, which has a direct impact on path planning and construction method selection. The construction target data covers construction plans, design drawings, engineering specifications and other data, which ensures the consistency of the construction process with the established goals. Through the comprehensive acquisition and analysis of these data, a panoramic control of the construction site can be achieved, providing reliable data support for each subsequent step.

[0032] Step S2, determining the machine head position of the pipe jacking machine based on the first information, and sending the determined machine head position and the first information to the path planning module for path planning to obtain pipe jacking machine pushing path information;

[0033] It can be understood that through accurate machine head position determination and intelligent path planning, the path deviation of the pipe jacking machine in construction can be significantly reduced, the pushing path is optimized, the construction efficiency is improved, and the energy consumption and construction cost are reduced. This combination based on deep learning and intelligent path planning can greatly improve the automation and intelligent level of construction in complex environments. In this step, step S2 includes steps S21, S22, S23, S24 and S25.

[0034] Step S21, performing dimensionality reduction processing on the pipe jacking machine image data and environmental data obtained at the construction site to obtain machine head image data and machine head environmental data of the pipe jacking machine;

[0035] It can be understood that this step calculates the intra-class scatter matrix (S_w) and the inter-class scatter matrix (S_b) of various data for the high-dimensional pipe jacking machine image data and environmental data obtained at the construction site, where S_w represents the dispersion degree of the same class of data, and S_b represents the dispersion degree of different classes of data. Then, by maximizing the ratio of inter-class scatter to intra-class scatter, the linear discriminant analysis algorithm finds a projection direction, so that the distance between the projected data classes is as large as possible, and the intra-class distance is as small as possible. By solving the eigenvalues and eigenvectors, the eigenvectors that can maximize the inter-class separability are selected, and the original high-dimensional data is projected into a low-dimensional space, thereby obtaining the dimensionality reduction results of the machine head image data and environmental data of the pipe jacking machine. This process effectively extracts the most meaningful features for distinguishing tasks, greatly reduces the data dimension while retaining important information, and optimizes the calculation efficiency of subsequent path planning.

[0036] Step S22, analyzing the machine head image data and machine head environmental data of the pipe jacking machine based on the preset linear discriminant analysis method to obtain machine head position information of the pipe jacking machine;

[0037] It can be understood that this step is based on a preset linear discriminant analysis method to analyze the push bench image data and the push bench environment data, so as to accurately obtain the push bench position information. First, a classification model is constructed by the linear discriminant analysis method, and the reduced dimension image data and the environment data of the push bench are taken as inputs. The linear discriminant analysis method analyzes the within-class and between-class scatter of the data, finds the optimal linear combination, maximizes the separability between different classes, and thus maps the data to a new low-dimensional space. In this space, the originally indistinguishable classes are more separated in the new projection direction, and such processing greatly improves the accuracy of classification. Subsequently, the distance of each data point to the classification boundary is calculated in the reduced dimension data space to determine the position of the push bench. The unique feature of this method is that it can effectively reduce the dimension of the data while maintaining the key information of the data, improving the efficiency of data processing and analysis. The technical effect is reflected in accurately positioning the push bench position, ensuring the accuracy and reliability of the path planning, and helping the smooth development of the subsequent construction process.

[0038] Step S23, constructing a three-dimensional model of the push bench position information, environment data and construction target data of the push bench to obtain first sub-information, the first sub-information including three-dimensional coordinates of the push bench position, three-dimensional coordinates of the environment and three-dimensional coordinates of the construction target;

[0039] It can be understood that this step integrates the push bench position information, environment data and construction target data of the push bench, and generates first sub-information by constructing a three-dimensional model. Specifically, using three-dimensional modeling technology, the obtained push bench position, environmental features and construction target parameters are respectively mapped to points or surfaces in three-dimensional space to form a comprehensive model with spatial relationship. First, the push bench position data is analyzed as three-dimensional coordinates to represent the actual position in the construction space; then, the environment data including surrounding terrain, obstacles and other information are modeled as three-dimensional structures consistent with the actual scene; finally, the construction target data is converted into three-dimensional coordinates and spatial range of the target position. These data are integrated into a unified three-dimensional model through coordinate transformation and spatial registration technology, and the generated first sub-information not only contains the spatial coordinates of each element, but also reflects the spatial relationship between them. This process ensures the accuracy and adaptability of the push bench construction path planning, and provides a complete three-dimensional reference framework, laying a foundation for subsequent path planning and other construction decisions. The technical effect is reflected in comprehensively capturing the spatial layout of the construction site, which helps accurate construction path planning and risk prediction.

[0040] Step S24, mapping the first sub-information to a one-dimensional sequence corresponding to a preset Hilbert curve, and generating a push bench pushing path through the mapping;

[0041] It can be understood that the Hilbert curve in this step is a curve that continuously fills the space, and has the ability to map multi-dimensional data into one-dimensional space while maintaining the spatial proximity of the data as much as possible. Specifically, first, the three-dimensional coordinates of the machine head position, environment and construction target in the first sub-information are converted into discrete three-dimensional grid points. Then, by constructing a Hilbert curve, the grid points are arranged in the order of the path of the curve, thereby mapping the three-dimensional data into a one-dimensional space. This one-dimensional sequence not only simplifies the data structure, but also preserves the geometric relationship in the original three-dimensional space. Next, by analyzing the continuity and trend of the one-dimensional sequence, the pushing path of the pipe jacking machine is generated, which can better fit the actual environment during construction and optimize the path length and obstacle avoidance ability. The technical effect lies in the use of the space preserving property of the Hilbert curve, effectively reducing the data dimension and reducing the computational complexity, while ensuring the accuracy and reliability of the construction path.

[0042] Step S25, based on the convex hull algorithm, the pushing path of the pipe jacking machine is analyzed for feasibility, and the pushing path of the pipe jacking machine with the shortest pushing distance is obtained.

[0043] It can be understood that this step restores the one-dimensional sequence path data generated in the above steps to a set of path points in three-dimensional space. Then, the convex hull algorithm is applied to construct the smallest convex polyhedron covering all path points, which represents the boundary range of all possible paths. In this process, by extracting the vertices of the convex hull, the extreme points and turning points in the path are identified, and the total length of the path is calculated and optimized. By comparing the lengths of all possible paths, the path with the shortest pushing distance is selected as the final pushing path. The technical effect of this method lies in the fact that the convex hull algorithm can quickly and effectively process multi-dimensional data and identify key points in the path, ensuring that the pipe jacking machine selects the optimal path in complex construction environments, reducing construction time and resource consumption, while improving the accuracy and efficiency of path planning.

[0044] Step S3, based on the pipe jacking machine pushing path information, construction is carried out until the machine heads of the two corresponding pipe jacking machines reach the preset jointing section position, the temperature distribution and thickness distribution of the jointing section position are obtained, and the preset historical freezing scheme is adjusted based on the temperature distribution and thickness distribution of the jointing section position, to obtain a freezing construction scheme for the jointing section position;

[0045] It can be understood that this step dynamically adjusts the freezing construction scheme through precise temperature monitoring and thickness analysis, thereby improving the effect and accuracy of freezing construction, reducing construction risks, and effectively avoiding construction problems caused by uneven temperature or soil changes. In this step, step S3 includes steps S31, S32, S33, S34 and S35.

[0046] Step S31, constructing a finite element model based on the preset geological data information of the closure section position, wherein the attributes and boundary conditions of the geological data are defined to obtain the established finite element model;

[0047] It can be understood that this step first constructs a finite element model based on the preset geological data information of the closure section position. The core purpose of this process is to accurately simulate the physical properties of the soil around the closure section and the complex processes such as heat conduction and frozen wall formation during construction. The first step in constructing the finite element model is to extract the geological data of the closure section, including physical properties such as soil type, density, humidity, and thermal conductivity, as well as information such as groundwater distribution that affects the freezing process.

[0048] On this basis, the boundary conditions of the model are defined. The boundary conditions include the temperature field distribution of the closure section position, the temperature of the freezing medium during the freezing construction process, and the temperature of the surrounding environment. These boundary conditions will affect the heat conduction equation in the model, thereby determining the thickness, expansion speed of the frozen wall, and its interaction with the surrounding soil. The setting of boundary conditions usually needs to combine with the on-site environmental monitoring data and historical construction experience to ensure that the model can as accurately as possible reflect the actual situation during the construction process.

[0049] The heat conduction equation is as follows:

[0050]

[0051] where, is the Laplacian operator of the temperature field, representing the spatial second-order partial derivative of the temperature field, describing the propagation of heat in space. α is the thermal diffusivity, representing the ability of heat diffusion, usually determined by the thermal conductivity and specific heat capacity of the material, T represents the temperature field, t represents the freezing construction time, and Q represents the heat source term.

[0052] Through the definition of the above geological properties and boundary conditions, the finite element model is established. This model not only can simulate the formation process of the frozen wall, but also can predict the temperature change, heat conduction efficiency, and formation speed of the frozen wall during the freezing process, providing theoretical support for subsequent construction scheme optimization. Through high-precision finite element simulation, various uncertain factors in the freezing construction can be predicted in advance, providing scientific basis for the adjustment and optimization of the construction scheme, ensuring the efficiency and safety of the freezing construction.

[0053] Step S32, meshing the established finite element model, and inputting the temperature information collected by each grid into the finite element model to solve the temperature field of the heat conduction equation, obtaining the temperature field distribution information of the closure section;

[0054] It can be understood that this step first carries out meshing on the established finite element model. The purpose of this process is to discretize the continuous physical space into a plurality of small finite elements, so as to accurately calculate the physical process inside each element. The fineness of meshing directly affects the accuracy of simulation results and the efficiency of calculation. Generally speaking, the finer the mesh, the higher the calculation accuracy, but the calculation amount will also increase accordingly. Next, the temperature information collected by each grid is input into the finite element model to solve the heat conduction equation. The heat conduction equation is a basic equation describing the propagation of heat in a medium, and its form is usually a partial differential equation with respect to time and space variables. In this process, the initial temperature conditions of each grid (such as the soil temperature before construction and the ambient temperature) and the boundary conditions (such as the temperature of the freezing medium and the temperature change during freezing construction) are input to solve the heat conduction equation.

[0055] The temperature field distribution information of the closure section is calculated by the model, and the temperature distribution in the entire closure section area is displayed. These temperature field information can intuitively reflect the formation process of the frozen wall, and can reveal the temperature variation trend of each region in the soil or rock. In this way, the construction personnel can understand which regions have better freezing effect and which regions may have temperature deviation, so as to timely adjust the freezing scheme.

[0056] In step S33, the temperature nodes related to the frozen wall are selected based on the temperature field distribution information of the closure section, and regression analysis is performed based on the preset historical frozen wall data to obtain a mathematical relationship between the temperature and the thickness of the frozen wall;

[0057] The mathematical relationship is as follows:

[0058]

[0059] Where d(t) represents the thickness of the frozen soil, represents the gradient of the soil temperature field, t represents the freezing construction time, k represents the thermal conductivity of the soil, p represents the density of the soil, c represents the specific heat capacity of the soil, T env represents the temperature of the surrounding environment, T freeze represents the temperature of the freezing medium.

[0060] It can be understood that this step is first based on the temperature field distribution information of the closure section to select the temperature nodes related to the frozen wall. The temperature node refers to a specific point or area in the finite element model where temperature information is collected. In the temperature field distribution of the closure section, the formation of the frozen wall is closely related to the temperature change of these temperature nodes, so it is necessary to select those important temperature nodes that can represent the thickness change of the frozen wall. These temperature nodes are located at the boundary or near the frozen wall area, where the temperature change is most significant. By selecting these nodes, accurate monitoring of the changes in the frozen wall can be ensured. Then, regression analysis is performed based on the preset historical frozen wall data. Historical frozen wall data is usually obtained through previous construction experience or experimental data, which reflects how the thickness of the frozen wall changes under different temperature conditions. Regression analysis provides an effective mathematical model for predicting the thickness of the frozen wall by combining historical data with current temperature field information. This method not only enhances the accuracy of construction, but also improves the controllability of the freezing construction, avoiding unstable freezing effects due to temperature fluctuations, and providing a basis for subsequent freezing construction scheme adjustments.

[0061] Step S34, predicting the thickness of the frozen wall of each grid area based on the mathematical relationship, and determining the optimal freezing parameter combination based on the thickness of the frozen wall of each grid area and the temperature field distribution information, the freezing parameter combination including the combination of freezing medium temperature, freezing time and freezing spacing;

[0062] It can be understood that this step can calculate the thickness of the frozen wall of each area by inputting the temperature information of each grid area into the mathematical model. Next, based on the thickness of the frozen wall of each grid area and the temperature field distribution information, the optimal freezing parameter combination is determined. The optimal freezing parameter combination is to ensure that the frozen wall reaches the expected thickness and remains stable during construction. The freezing parameter combination includes three important parameters: freezing medium temperature, freezing time and freezing spacing. Among them, the freezing medium temperature determines the freezing rate, too low temperature may lead to too fast freezing, while too high temperature may lead to incomplete freezing; freezing time affects the duration of the freezing process, appropriate freezing time helps to form a stable frozen wall; freezing spacing is related to the arrangement of freezing equipment and freezing efficiency, reasonable spacing can improve the freezing effect and avoid energy waste. In order to determine the optimal freezing parameter combination, this step adopts a multi-objective optimization method, considering factors such as frozen wall thickness, construction efficiency and energy consumption, to find the optimal solution under these conditions.

[0063] It can be understood that the objective function of optimization in this step is as follows:

[0064]

[0065] where A represents the optimal solution, w1 represents the weight coefficient of the frozen wall thickness, d(t) represents the thickness of the frozen wall, w2 represents the weight coefficient of the construction time, T construction represents the total time required for construction, w3 represents the weight coefficient of energy consumption, Q freeze (T freeze ) represents the flow rate of the freezing medium, t freeze represents the time interval of the freezing process, and dt represents the small time increment in the integral.

[0066] This step can achieve accurate control of the frozen wall thickness and optimize the selection of freezing parameters through the prediction of mathematical relationships and the application of optimization algorithms, thereby improving the accuracy and stability of the freezing construction. In this way, the uncertainty in the construction process can be greatly reduced, avoiding construction delays or resource waste caused by uneven freezing or excessive freezing time, and achieving a more efficient and economical freezing construction process. At the same time, this also provides a basis for real-time adjustment and monitoring on the construction site, ensuring the smooth achievement of construction goals.

[0067] Step S35, based on the optimal freezing parameter combination and historical freezing scheme, sends to the preset neural network structure for learning and training, and predicts the freezing construction scheme of the closure section position.

[0068] It can be understood that the neural network in this step is a deep neural network, and the structure of the deep neural network includes an input layer, a hidden layer, and an output layer, where each layer of neurons is connected to the neurons of the previous and next layers. In this step, the input layer accepts data of the optimal freezing parameter combination and historical freezing scheme, including freezing medium temperature, freezing time, freezing interval, historical freezing construction data, etc. The hidden layer abstracts and learns the complex relationship between these input data through multiple nonlinear transformations, and the final output layer will predict the freezing construction scheme of the closure section position, including predicted freezing parameters and construction steps. By inputting the optimal freezing parameter combination and historical freezing scheme, the deep neural network adjusts its weights and biases through the backpropagation algorithm to minimize the error between the predicted value and the actual demand.

[0069] The deep neural network will continuously adjust the model parameters during the training process, learning the rules in historical data, not only adapting to different construction environments, but also providing reasonable schemes for freezing construction in new environments. This deep learning-based approach makes the formulation of freezing construction schemes more intelligent, avoiding the limitations of relying on experience and manual adjustments.

[0070] Step S4, based on the frozen construction scheme of the closure section position, construction is carried out until the frozen wall of the closure section position reaches the preset condition, and a machine head construction scheme is constructed based on the first information, the machine head construction scheme including a machine head disassembly scheme, a machine head shell welding scheme and a machine head waterproof construction scheme;

[0071] It can be understood that through the construction and implementation of the above three schemes, the disassembly, welding and waterproof construction process of the machine head can be effectively ensured to be completed smoothly, and technical support is provided for subsequent construction links. In this step, step S4 includes step S41, step S42, step S43, step S44, step S45 and step S46.

[0072] Step S41, obtaining the machine head structure data information of the parameter data of the pipe jacking machine in the first information, and using the analytic hierarchy process to decompose the preset machine head disassembly task into a multi-level structure to obtain a hierarchical model of the disassembly steps;

[0073] It can be understood that in this step, the parameter data of the pipe jacking machine is first extracted from the first information, especially the structure data information of the machine head. These structure data include the geometric shape of the machine head, the size of each part, the weight distribution, and the connection mode with other components, etc. These information is the basis for disassembly task design, ensuring that the disassembly scheme can adapt to the structural characteristics of the machine head and effectively proceed.

[0074] Next, the preset machine head disassembly task is decomposed into a multi-level structure using the analytic hierarchy process. The analytic hierarchy process is a multi-objective decision-making method, which can assign a relative weight to each step according to the importance, difficulty, time requirement and other factors of each disassembly step. In specific operation, first identify all the steps in the disassembly process, such as disassembling the machine head shell, disassembling the internal components, cleaning and inspecting each component, etc. Then, sort these steps and divide them into multiple levels, ensuring that the priority and execution order of each step are appropriate. For example, the first level of the disassembly task may be to disassemble the shell, and the second level may be to handle the internal electrical system or pipe connection.

[0075] The analytic hierarchy process quantifies the weights of different disassembly tasks through expert judgment and mathematical models, forming a hierarchical structure model. This structure model not only helps to optimize the disassembly process and reduce unnecessary operations, but also improves the efficiency and safety of the disassembly process. It can provide a clear task allocation and implementation framework for subsequent construction and adjustment.

[0076] Step S42, calculating the weight of each disassembly task step based on the relative importance weight of the preset machine head disassembly task, and sorting each machine head disassembly task according to the weight calculation result, wherein the sorting is performed in the order of relative importance weight from large to small, to obtain a machine head disassembly scheme;

[0077] It can be understood that in this step, the expert team first compares and evaluates the role of each step in the entire disassembly process according to the preset nose disassembly task. For example, some steps may be crucial in the disassembly process, such as ensuring that important internal components are not damaged when disassembling the nose shell, while other steps may be secondary, such as cleaning work or detailed adjustments. Through the analytic hierarchy process, the expert assigns an importance weight to each disassembly task, reflecting the influence of the step on safety, efficiency, or successful completion of the task in the disassembly process. Or directly assign an importance weight to each task based on historical data.

[0078] Next, according to these weight values, all disassembly tasks are sorted in order of relative importance from large to small. The sorting result will provide a clear priority order for the nose disassembly, ensuring that the most critical steps are executed first. The order and priority of each step will directly affect the efficiency and accuracy of the disassembly process. For example, if the step of disassembling the nose shell has a higher weight than other steps, the disassembly of the nose shell will be prioritized in the disassembly process.

[0079] Through this process, the obtained nose disassembly scheme not only ensures the efficient execution of disassembly tasks, but also maximizes the reduction of errors and risks. The sorted disassembly scheme helps construction personnel understand the step arrangement of disassembly, thereby improving the smoothness and safety of operation. The technical effect is reflected in optimizing the disassembly process, saving time, and improving overall construction efficiency through reasonable task allocation, reducing mechanical damage and personnel accidents.

[0080] Step S43, obtaining the geometry of the pipe jacking machine in the image data of the pipe jacking machine in the first information and preset welding requirement information;

[0081] It can be understood that image data is usually collected by high-resolution cameras or laser scanners, covering detailed images of various parts of the pipe jacking machine. In the image processing stage, computer vision technology is used for feature extraction. Through edge detection, contour recognition, image segmentation, etc., the shape information and internal structure of the pipe jacking machine are extracted, and a geometric model of the nose is constructed. This process is realized by using YOLOV3 algorithm in this invention. This step also combines the preset welding requirement information with the geometric shape data. These welding requirement information usually includes the welding method (such as TIG welding, MIG welding, etc.), the quality standard of welding, the joint type (such as fillet weld, butt weld, etc.), and the temperature and pressure requirements during welding, etc. Welding requirements can be extracted from historical data, construction standards and design specifications. The welding requirements and the geometric shape data of the nose are fused to ensure that the welding scheme can meet the requirements of structural strength, durability, and operation convenience, etc.

[0082] Step S44, constructing a welding path model based on the geometry of the pipe jacking machine and the preset welding requirement information, wherein a plurality of welding path information is obtained by defining the starting point, the ending point and the constraint condition of the path;

[0083] It can be understood that this step analyzes the geometric data of the pipe jacking machine to determine the morphological characteristics of the welding position, especially the shape and size of the connection area and the possible complex geometric conditions. The head structure of the pipe jacking machine usually includes multiple connection parts, such as the shell, support, joint, etc. In this process, using computational geometry and three-dimensional modeling techniques, a welding area model of the head is constructed according to the external and internal structure of the pipe jacking machine. Next, the starting point and the ending point of the welding path are defined. The starting point of the welding path is usually selected at the beginning of the joint, and the ending point is the position where the welding ends. When selecting these starting and ending points, it is necessary to ensure that the welding operation can cover the entire joint area and meet the preset welding requirements, such as welding depth, weld width, etc. In addition, in order to ensure the welding quality and construction efficiency, the path model also needs to define a series of constraint conditions. These constraint conditions include welding speed, welding temperature range, welding angle, joint type and external environmental factors (such as climate, pressure, etc.). These constraint conditions ensure that the welding process can be carried out within a safe and accurate range, avoiding welding defects.

[0084] Step S45, optimizing and selecting all welding path information based on a particle swarm optimization algorithm, wherein each particle represents a path selection, and the fitness of each path is evaluated iteratively to calculate its optimization degree until the optimal welding path information is obtained, and the optimal welding path information is used as the head shell welding scheme;

[0085] It can be understood that in this step, the particle swarm optimization algorithm regards each welding path as a "particle". Each particle represents a possible path selection and has a position (path selection) and a speed (path adjustment direction). Initially, all particles are randomly distributed in the search space, and the fitness of each particle is evaluated by a preset objective function. The goal of fitness evaluation is to optimize the performance of the welding path, such as welding accuracy, welding speed, welding cost, etc.

[0086] It can be understood that the fitness function in this step is as follows:

[0087] f(X) = a·P precision +b·V welding +c·C welding

[0088] Where X is the position of the particle, representing the selection of the welding path, P precision is the welding accuracy, which calculates the deviation of the path from the target path, the higher the accuracy, the smaller the value, V weldingis the welding speed, C welding is the welding cost, a is the weight coefficient of welding precision, b is the weight coefficient of welding speed, and c is the weight coefficient of welding cost.

[0089] In the optimization process, each particle adjusts its current position based on its own historical best position and the best position of all particles in the swarm. With each iteration, the particle swarm gradually converges, and the particles focus on the optimal path selection. The fitness of each particle is continuously evaluated, and its personal best position and global best position are updated. Through multiple iterations, the particle swarm will eventually converge to an optimal solution, i.e., the optimal welding path. During the iteration process of the particle swarm optimization algorithm, the degree of path optimization is quantified by the objective function. The objective function usually includes multiple dimensions, such as the distance of the welding path, welding time, welding quality, energy consumption, and possible interference during the welding process. By considering these factors comprehensively, the particle swarm algorithm can assign a fitness value to each path, and then select the path with the lowest comprehensive cost and the highest welding quality.

[0090] Step S46, selecting all the preset waterproof materials based on the environmental data of the pipe jacking machine in the first information, wherein the waterproof material with the minimum cost and meeting the preset conditions is selected as the final selected waterproof material, and the preset construction scheme corresponding to the final selected waterproof material is taken as the machine head waterproof construction scheme.

[0091] It can be understood that this step first needs to screen the preset waterproof materials according to the environmental data. Each waterproof material has different performance parameters, such as pressure resistance, impermeability, chemical stability, temperature and humidity resistance, etc. Through these performance indicators, materials that do not meet the environmental condition requirements can be filtered out. For example, if the soil on site has high chemical corrosiveness, some materials may be excluded due to insufficient corrosion resistance. Next, by selecting waterproof materials with cost-effectiveness and meeting environmental condition requirements, combined with the preset construction budget and resource conditions, the cost performance of the materials is further evaluated. At this time, the cost-effectiveness consideration is very important because there are differences in the procurement and construction costs of different materials. By considering the technical performance and economy of the waterproof materials, the waterproof material with the highest cost performance is finally selected.

[0092] Step S5, based on the machine head construction scheme, construction is carried out until the waterproof effect of the machine head shell reaches the preset threshold, a closing section reinforced concrete construction scheme and a pressure supplementing grouting scheme are constructed based on the first information, and construction is carried out in sequence according to the closing section reinforced concrete construction scheme and the pressure supplementing grouting scheme to obtain the final construction result.

[0093] It can be understood that this step combines the optimization of the closure section reinforced concrete construction scheme and the pressure supplementing grouting scheme, which can effectively improve the construction efficiency, reduce the construction risk, and ensure that the engineering quality meets the expected standard. Finally, the entire construction process ensures that the project is completed according to the plan while ensuring construction safety and quality, providing a solid foundation for the subsequent construction stage. In this step, step S5 includes step S51, step S52, and step S53.

[0094] Step S51, extracting geological parameters, environmental parameters, and construction target data based on environmental data at the closure section location;

[0095] It can be understood that by combining the extraction of geological parameters, environmental parameters, and construction target data, comprehensive information support can be provided for subsequent construction scheme optimization and execution. This step ensures that the development of the construction scheme is more scientific and reasonable through accurate data analysis and parameter extraction, avoiding problems caused by neglecting certain parameters in actual construction. Ultimately, these data will help the construction team make more flexible and accurate adjustments when facing complex construction environments, ensuring the smooth progress of the entire project.

[0096] Step S52, constructing a multi-objective optimization function based on the geological conditions, environmental parameters, and construction target data at the closure section location, wherein the multi-objective optimization model takes construction intensity, construction cost, and construction time as optimization objectives;

[0097] It can be understood that the construction intensity in this step mainly refers to the structural strength requirement during construction, to ensure that the materials and processes used in construction can meet the engineering safety requirements. For the construction of the closure section, the intensity not only involves the compressive strength of reinforced concrete, but also involves the supporting capacity of the surrounding soil or rock mass. The differences in geological conditions, such as soil density, groundwater level and rock stability, will directly affect the required construction intensity. By incorporating construction intensity into the optimization function, it can be ensured that the use of materials during construction can meet the structural safety requirements. Construction cost is one of the key factors of the entire project, covering the procurement cost of construction materials, labor cost, equipment usage cost, etc. Different construction methods and technical choices will have different impacts on cost. For example, using high-strength materials and advanced equipment can improve construction efficiency, but also increase the cost of individual materials and equipment. Therefore, optimizing construction cost means that under the premise of ensuring construction intensity and construction time, unnecessary costs such as material waste and equipment downtime should be minimized. Construction time is an important indicator for evaluating construction progress, and directly affects the delivery cycle of the project. Reasonably arranging the construction sequence, optimizing the construction process, and reducing the waiting and downtime during construction can effectively reduce the construction time. Especially in complex geological environments and extreme weather conditions, the optimization of construction time is particularly important. By introducing construction time as an optimization objective, it can avoid the waste of resources and the increase of construction cost caused by too long construction period. By constructing and solving the multi-objective optimization function, the construction team can weigh the pros and cons of each aspect while considering geological conditions, environmental factors and project goals, and finally obtain an optimal construction scheme.

[0098] Step S53, solving the multi-objective optimization function to obtain the optimal reinforced concrete construction scheme for the closure section.

[0099] It can be understood that this step, when solving the multi-objective optimization problem of the reinforced concrete construction scheme of the closure section, randomly generates an initial solution and gradually adjusts it through the simulated annealing algorithm to find the optimal balance of construction intensity, construction cost and construction time, etc.

[0100] First, initialize a solution S0, and set the initial temperature T0 and the cooling parameter a. Then, generate a neighborhood solution S' based on the current solution S, calculate the energy difference DE = E(S') - E(S), where E(S) is the objective function value of the current solution, and E(S') is the objective function value of the new solution. If the new solution S' is better (DE < 0), it is directly accepted; otherwise, with a probability Accepting inferior solutions to avoid falling into local optimal solutions. As the temperature T gradually decreases, the algorithm gradually converges to the global optimal solution. The stopping condition of simulated annealing is usually that the temperature reaches a set minimum value or the number of iterations reaches an upper limit, and the final output is the optimal solution S*, such as a construction plan that balances cost, strength and time. By continuously adjusting the weights and parameters of the objective function, the simulated annealing algorithm can provide an effective global search solution in multi-objective optimization, and find the optimal construction path and material configuration.

[0101] In this step, step S5 further comprises step S54, step S55 and step S56.

[0102] Step S54, based on environmental data, extracts geological parameters, fracture distribution and water pressure information of the closure section, and takes the geological parameters, fracture distribution and water pressure information as input variables;

[0103] It can be understood that after taking the extracted geological parameters, fracture distribution and water pressure information as input variables, this step can provide accurate basic data for subsequent pressure compensation grouting scheme. These input variables will help build a detailed underground environment model, ensure that the pressure compensation grouting scheme can effectively deal with the instability of underground structure and water pressure changes in actual construction, so as to optimize the key parameters such as grouting amount, pressure and flow rate, and ensure the smooth progress of construction. Through this process, the construction environment under underground conditions can be accurately simulated to provide theoretical support for subsequent grouting operation and freezing construction, and ensure the safety and effectiveness of the construction scheme.

[0104] Step S55, based on the preset historical pressure compensation grouting data, an initial grouting model is constructed, and grouting amount, grouting pressure and grouting flow rate parameters are defined;

[0105] It can be understood that this step helps to establish a preliminary grouting model by analyzing past projects, such as groundwater permeability, fracture expansion and underground stability, etc. In the process of constructing the initial grouting model, three key parameters of grouting amount, grouting pressure and grouting flow rate need to be defined. Grouting amount refers to the total amount of grouting liquid required in each grouting process, which directly affects the construction effect and material cost. Grouting pressure refers to the pressure applied during grouting.

[0106] Step S56, inputting the input variables into the initial grouting model for processing, wherein the initial grouting model is optimized based on the Bayesian optimization algorithm, and the optimal pressure compensation grouting construction scheme is predicted.

[0107] It can be understood that in this step, Bayesian optimization is a method of global optimization through a probability model, which is usually used to solve optimization problems with high dimension, high computational cost and no explicit analytical solution. In this step, the Bayesian optimization algorithm establishes a probability relationship between the input variables and the output effect (the relationship between the grouting effect and the grouting parameters), and infers the optimal parameter combination according to this relationship. Bayesian optimization can gradually find the parameter combination that can maximize the grouting effect through continuous updating and improvement of the probability model with fewer test times.

[0108] In the Bayesian optimization process, this step first defines a Gaussian process model, and then uses the model to predict the output of the initial grouting model. By optimizing the proxy model to minimize the expected loss target, Bayesian optimization can effectively determine which parameter combination is most likely to bring the best grouting effect, and provide the optimal pressure compensation grouting scheme for actual construction.

[0109] Through this optimization method, not only the number of experiments can be reduced, the construction cost can be reduced, but also the accuracy and effect of construction can be improved, and the optimal grouting effect can be achieved in complex construction environment, and the requirements of geological conditions, fracture distribution and other aspects can be met to the greatest extent.

[0110] Embodiment 2:

[0111] As shown in Figure 2 The embodiment provides an intelligent control system for frozen counter-jacking through based on deep learning, which is shown in Figure 2 The system comprises an acquisition unit 701, a planning unit 702, an adjustment unit 703, a construction unit 704 and a processing unit 705.

[0112] The acquisition unit 701 is configured to acquire first information, wherein the first information comprises parameter data, image data, environment data and construction target data of all jacking machines at a construction site.

[0113] The planning unit 702 is configured to determine the position of the jacking machine head based on the first information, and send the determined position of the jacking machine head and the first information to a path planning module for path planning to obtain jacking machine advancing path information.

[0114] The adjustment unit 703 is configured to perform construction based on the jacking machine advancing path information, until the positions of the heads of two corresponding jacking machines reach a preset converging section position, acquire the temperature distribution and thickness distribution of the converging section position, and adjust the preset historical freezing scheme based on the temperature distribution and thickness distribution of the converging section position to obtain a freezing construction scheme of the converging section position.

[0115] Construction unit 704 is used to carry out construction based on the freezing construction plan at the closure section position until the frozen wall at the closure section position reaches the preset conditions. Based on the first information, a machine head construction plan is constructed. The machine head construction plan includes a machine head disassembly plan, a machine head shell welding plan, and a machine head waterproofing construction plan.

[0116] The processing unit 705 is used to carry out construction based on the machine head construction plan until the waterproof effect of the machine head shell reaches a preset threshold. Based on the first information, it constructs a closure section reinforced concrete construction plan and a pressure injection grouting plan, and carries out construction in sequence according to the closure section reinforced concrete construction plan and the pressure injection grouting plan to obtain the final construction result.

[0117] It should be noted that the specific ways in which each module performs operations in the system described in the embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0118] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0119] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An intelligent control method for frozen flip-chip based on deep learning, characterized in that, The application comprises the following steps: acquiring first information, which comprises parameter data, image data, environmental data and construction target data of all pipe jacking machines in the construction site; determining the position of the machine head of the pipe jacking machine based on the first information, and sending the determined position of the machine head and the first information to a path planning module for path planning to obtain pipe jacking machine advancing path information; based on the pipe jacking machine advancing path information, construction is carried out until the machine heads of two corresponding pipe jacking machines reach a preset jointing section position, the temperature distribution and thickness distribution of the jointing section position are acquired, and the preset historical freezing scheme is adjusted based on the temperature distribution and thickness distribution of the jointing section position to obtain a freezing construction scheme of the jointing section position; based on the freezing construction scheme of the jointing section position, construction is carried out until the freezing wall of the jointing section position reaches a preset condition, a machine head construction scheme is constructed based on the first information, and the machine head construction scheme comprises a machine head disassembly scheme, a machine head shell welding scheme and a machine head waterproof construction scheme; based on the machine head construction scheme, construction is carried out until the waterproof effect of the machine head shell reaches a preset threshold, a jointing section reinforced concrete construction scheme and a pressure compensation grouting scheme are constructed based on the first information, and construction is sequentially carried out according to the jointing section reinforced concrete construction scheme and the pressure compensation grouting scheme to obtain a final construction result; wherein, acquiring the temperature distribution and thickness distribution of the jointing section position, and adjusting the preset historical freezing scheme based on the temperature distribution and thickness distribution of the jointing section position to obtain the freezing construction scheme of the jointing section position comprises: constructing a finite element model based on preset geological data information of the jointing section position, wherein the properties and boundary conditions of the geological data are defined to obtain a completed finite element model; dividing the completed finite element model into grids, and inputting the temperature information collected by each grid into the finite element model to solve the temperature field of the heat conduction equation to obtain the temperature field distribution information of the jointing section; selecting temperature nodes related to the freezing wall based on the temperature field distribution information of the jointing section, and performing regression analysis based on preset historical freezing wall data to obtain a mathematical relationship between temperature and freezing wall thickness; predicting the freezing wall thickness of each grid area based on the mathematical relationship, and determining the optimal freezing parameter combination based on the freezing wall thickness of each grid area and the temperature field distribution information, wherein the freezing parameter combination comprises a combination of freezing medium temperature, freezing time and freezing spacing; sending the optimal freezing parameter combination and the historical freezing scheme to a preset neural network structure for learning and training, and predicting the freezing construction scheme of the jointing section position. 2.The deep learning-based intelligent control method of freeze-through of a cold plate according to claim 1, wherein based on the first information, the position of the machine head of the pipe jacking machine is determined, and the determined position of the machine head and the first information are sent to a path planning module for path planning to obtain pipe jacking machine advancing path information, which comprises the following steps: dimension reduction processing is performed on the pipe jacking machine image data and environmental data acquired at the construction site to obtain machine head image data and machine head environmental data of the pipe jacking machine; based on a preset linear discriminant analysis method, the machine head image data and machine head environmental data of the pipe jacking machine are analyzed to obtain machine head position information of the pipe jacking machine; The head position information, environment data and construction target data of the pipe jacking machine are constructed into a three-dimensional model to obtain first sub-information, which includes head position three-dimensional coordinates, environment three-dimensional coordinates and construction target three-dimensional coordinates; The first sub-information is mapped into a one-dimensional sequence corresponding to a preset Hilbert curve, and a one-dimensional sequence is generated to obtain a pushing path of the pipe jacking machine; The pushing path of the pipe jacking machine is analyzed for feasibility based on a convex hull algorithm to obtain a pushing path of the pipe jacking machine with the shortest pushing distance. 3.The deep learning-based intelligent control method of freeze-through of a cold plate according to claim 1, wherein A head construction scheme is constructed based on the first information, and the head construction scheme includes a head disassembly scheme, a head shell welding scheme and a head waterproof construction scheme, including: Head structure data information of the parameter data of the pipe jacking machine in the first information is obtained, and an analytic hierarchy process is used to decompose a preset head disassembly task into a multi-level structure to obtain a hierarchical model of disassembly steps; The weight of each disassembly task step is calculated based on the relative importance weight of the preset head disassembly task, and each head disassembly task is sorted according to the weight calculation result, wherein the sorting is performed in the order from large to small of the relative importance weight, to obtain the head disassembly scheme; The geometry of the pipe jacking machine in the image data of the first information and preset welding requirement information are obtained; A welding path model is constructed based on the geometry of the pipe jacking machine and the preset welding requirement information, wherein a plurality of welding path information is obtained by defining the starting point, the ending point and the constraint condition of the path; All welding path information is selected by using a particle swarm optimization algorithm, wherein each particle represents a path selection, and each path is evaluated for fitness by iteration until the optimal welding path information is obtained, and the optimal welding path information is used as the head shell welding scheme; All preset waterproof materials are selected based on the environment data of the pipe jacking machine in the first information, wherein the waterproof material with the minimum cost and meeting the preset conditions is selected as the final selected waterproof material, and the preset construction scheme corresponding to the final selected waterproof material is used as the head waterproof construction scheme. 4.The intelligent control method of deep learning based freeze-through of a cold plate according to claim 1, wherein A closing segment reinforced concrete construction scheme and a pressure supplementing grouting scheme are constructed based on the first information, and the closing segment reinforced concrete construction scheme and the pressure supplementing grouting scheme are sequentially constructed to obtain a final construction result, including: Geological parameters, environment parameters and construction target data of the closing segment position are extracted based on the environment data; A multi-objective optimization function is constructed based on the geological conditions, environment parameters and construction target data of the closing segment position, wherein the multi-objective optimization model takes construction intensity, construction cost and construction time as optimization objectives; The multi-objective optimization function is solved to obtain an optimal closing segment reinforced concrete construction scheme.

5. An intelligent control system for frozen flip-chip based on deep learning, characterized in that, The unit, configured to obtain first information, the first information including parameter data, image data, environment data and construction target data of all pipe jacking machines at the construction site; ​ The planning unit is configured to determine the position of the machine head of the pipe jacking machine based on the first information, and send the determined position of the machine head and the first information to a path planning module for path planning to obtain pipe jacking machine pushing path information. The adjusting unit is configured to perform construction based on the pipe jacking machine pushing path information, until the machine heads of the two corresponding pipe jacking machines reach the preset closing section position, obtain the temperature distribution and thickness distribution of the closing section position, and adjust the preset historical freezing scheme based on the temperature distribution and thickness distribution of the closing section position to obtain a freezing construction scheme of the closing section position. The construction unit is configured to perform construction based on the freezing construction scheme of the closing section position, until the freezing wall of the closing section position reaches a preset condition, and construct a machine head construction scheme based on the first information, the machine head construction scheme including a machine head disassembly scheme, a machine head shell welding scheme, and a machine head waterproof construction scheme. The processing unit is configured to perform construction based on the machine head construction scheme, until the waterproof effect of the machine head shell reaches a preset threshold, construct a closing section reinforced concrete construction scheme and a pressure compensation grouting scheme based on the first information, and perform construction in sequence according to the closing section reinforced concrete construction scheme and the pressure compensation grouting scheme to obtain a final construction result. The adjusting unit includes: A first adjusting sub-unit configured to construct a finite element model based on preset geological data information of the closing section position, define the properties and boundary conditions of the geological data, and obtain the constructed finite element model. A second adjusting sub-unit configured to divide the constructed finite element model into grids, input the temperature information collected by each grid into the finite element model to solve the temperature field of the heat conduction equation, and obtain the temperature field distribution information of the closing section. A third adjusting sub-unit configured to select temperature nodes related to the freezing wall based on the temperature field distribution information of the closing section, and perform regression analysis based on preset historical freezing wall data to obtain a mathematical relationship between the temperature and the thickness of the freezing wall. A fourth adjusting sub-unit configured to predict the thickness of the freezing wall of each grid area based on the mathematical relationship, and determine an optimal freezing parameter combination based on the thickness of the freezing wall of each grid area and the temperature field distribution information, the freezing parameter combination including a combination of freezing medium temperature, freezing time, and freezing spacing. A fifth adjusting sub-unit configured to send the optimal freezing parameter combination and the historical freezing scheme to a preset neural network structure for learning and training, and predict the freezing construction scheme of the closing section position.

6. The intelligent control system for freeze through based on deep learning of claim 5, wherein, The planning unit includes: A first planning sub-unit configured to perform dimension reduction processing on pipe jacking machine image data and environment data obtained on the construction site to obtain machine head image data and machine head environment data of the pipe jacking machine. A second planning sub-unit configured to analyze the machine head image data and the machine head environment data of the pipe jacking machine based on a preset linear discriminant analysis method to obtain machine head position information of the pipe jacking machine. A third planning sub-unit configured to construct a three-dimensional model based on the machine head position information, the environment data, and the construction target data of the pipe jacking machine to obtain first sub-information, the first sub-information including machine head position three-dimensional coordinates, environment three-dimensional coordinates, and construction target three-dimensional coordinates. The fourth planning subunit is configured to map the first sub-information into a one-dimensional sequence corresponding to a preset Hilbert curve, and generate a pushing path of the pipe jacking machine through the mapping; The fifth planning subunit is configured to perform feasibility analysis on the pushing path of the pipe jacking machine based on a convex hull algorithm, and obtain a pushing path of the pipe jacking machine with the shortest pushing distance.

7. The intelligent control system for freeze through based on deep learning of claim 5, wherein, The construction unit comprises: The first construction subunit is configured to obtain head structure data information of parameter data of the pipe jacking machine in the first information, and decompose a preset head disassembly task into a multi-level structure using an analytic hierarchy process, and obtain a hierarchical model of disassembly steps; The second construction subunit is configured to calculate the weight of each disassembly task step based on the relative importance weight of the preset head disassembly task, and sort each head disassembly task according to the weight calculation result, wherein the sorting is performed in the order from large to small of the relative importance weight, and a head disassembly scheme is obtained; The third construction subunit is configured to obtain the geometric shape of the pipe jacking machine in the image data of the pipe jacking machine in the first information and preset welding requirement information; The fourth construction subunit is configured to construct a welding path model based on the geometric shape of the pipe jacking machine and the preset welding requirement information, wherein the starting point, the ending point and the constraint condition of the path are defined to obtain a plurality of welding path information; The fifth construction subunit is configured to perform optimization selection on all the welding path information based on a particle swarm optimization algorithm, wherein each particle represents a path selection, and each path is iteratively evaluated for fitness, and the optimization degree is calculated until the optimal welding path information is obtained, and the optimal welding path information is used as the head shell welding scheme; The sixth construction subunit is configured to select all preset waterproof materials based on the environmental data of the pipe jacking machine in the first information, wherein the waterproof material with the minimum cost and meeting the preset conditions is selected as the final selected waterproof material, and the preset construction scheme corresponding to the final selected waterproof material is used as the head waterproof construction scheme.

8. The intelligent control system for freeze through based on deep learning of claim 5, wherein, The processing unit comprises: The first processing subunit is configured to extract geological parameters, environmental parameters and construction target data of the jointing section position based on the environmental data; The second processing subunit is configured to construct a multi-objective optimization function based on the geological conditions, environmental parameters and construction target data of the jointing section position, wherein the multi-objective optimization model takes construction intensity, construction cost and construction time as optimization objectives; The third processing subunit is configured to solve the multi-objective optimization function to obtain an optimal jointing section reinforced concrete construction scheme.

Images