Real-time control method and system for grouting disaster control in underground engineering based on digital twin

Abstract

The present invention relates to the field of underground engineering grouting technology, and in particular to a real-time management and control method and system for underground engineering grouting disaster control based on digital twins, including: constructing a grouting engineering database; establishing an underground engineering disaster prediction model based on the grouting engineering database; constructing a grouting disaster control scheme optimization model based on the grouting engineering database; establishing a grouting disaster control monitoring model; establishing a digital twin model for grouting prevention and control of underground engineering disasters based on the underground engineering disaster prediction model, the grouting disaster control scheme optimization model and the grouting disaster control monitoring model; inputting the actual underground engineering preliminary data into the digital twin model for grouting prevention and control of underground engineering disasters, outputting the predicted disaster type and disaster degree, outputting the optimized grouting scheme, and visualizing the grouting disaster control process. Realize monitoring and visualization of the entire grouting process, realize real-time prediction of underground engineering disasters, and perform real-time optimization of grouting disaster control and monitoring schemes.

Description

Technical Field

[0001] The present invention relates to the technical field of underground engineering grouting, and in particular to a real-time management and control method and system for underground engineering grouting disaster control based on digital twins. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of the economy, the development and utilization of underground space has been increasingly strengthened. Compared with the above-ground space, different types of geological disasters are often encountered in the process of underground space development and utilization, including sudden water inrush, collapse, harmful gases, etc., which have extremely strong control and harm to underground engineering construction, causing large economic losses, adverse environmental and social impacts during the construction phase, and it is difficult to prevent and control. There are also major safety hazards during the operation phase. Grouting, as the most commonly used prevention and control method for various geological disasters, can effectively seal sudden water inrush, reinforce rock formations, and prevent and control disasters. However, the complexity of the rock medium of underground engineering makes the grouting process extremely concealed, and it is impossible to intuitively determine the specific process of slurry diffusion and the changes in various parameters during the grouting process. It is difficult to provide scientific and quantitative guidance for the grouting management of geological disasters in underground engineering construction.

[0004] The grouting process is a complex multiphase flow and phase change process. Existing research methods have preliminarily realized the monitoring and analysis of some data in the grouting process of underground engineering, but there is still a lack of monitoring and visualization of the grouting process under complex geological conditions. In addition, the existing research on the analysis of grouting disaster control effects has low prediction accuracy, unclear division and integration of various indicators, and no complete comprehensive system for the entire grouting process has been established, making it impossible to conduct real-time management and control of grouting disaster control in underground engineering in a more scientific, reasonable and efficient manner. Summary of the invention

[0005] In order to address the deficiencies of the prior art, the present invention provides a real-time management and control method and system for underground engineering grouting disaster control based on digital twins; realizes monitoring and visualization of the entire grouting process, realizes real-time prediction of underground engineering disasters, and performs real-time optimization of grouting disaster control and monitoring schemes.

[0006] On the one hand, a real-time management and control method for underground engineering grouting disaster control based on digital twin is provided, including:

[0007] Construct a grouting engineering database; based on the grouting engineering database, establish an underground engineering disaster prediction model;

[0008] Based on the grouting engineering database, an optimization model for grouting disaster control scheme is constructed;

[0009] Establish a grouting disaster control monitoring model;

[0010] Based on the underground engineering disaster prediction model, grouting disaster control scheme optimization model and grouting disaster control monitoring model, a digital twin model for grouting prevention and control of underground engineering disasters is established;

[0011] The actual preliminary data of underground projects is input into the digital twin model of underground project disaster grouting prevention and control, and the predicted disaster type and degree are output, as well as the optimized grouting plan, to visualize the grouting disaster control process.

[0012] On the other hand, a real-time management and control system for underground engineering grouting disaster control based on digital twins is provided, including:

[0013] The first establishment module is configured to: construct a grouting engineering database; and establish an underground engineering disaster prediction model based on the grouting engineering database;

[0014] The second building module is configured to: construct a grouting disaster control scheme optimization model based on a grouting engineering database;

[0015] The third establishment module is configured to: establish a grouting disaster control monitoring model;

[0016] A fourth building module is configured to: establish a digital twin model of grouting prevention and control of underground engineering disasters based on an underground engineering disaster prediction model, a grouting disaster control scheme optimization model, and a grouting disaster control monitoring model;

[0017] The output module is configured to: input the actual preliminary data of underground engineering into the digital twin model of underground engineering disaster grouting prevention and control, output the predicted disaster type and disaster degree, output the optimized grouting plan, and visualize the grouting disaster control process.

[0018] The above technical solution has the following advantages or beneficial effects:

[0019] Establish a grouting disaster control engineering database and carry out unified structural processing of various data, divide geological units to represent geological data and assign disaster attribute values ​​to construct a three-dimensional geological model, determine the degree of disaster impact corresponding to different score values, predict the type and degree of underground engineering disasters, and establish an underground engineering disaster prediction model.

[0020] Based on the prediction of underground engineering disasters, the grouting attribute value bonus criteria are obtained, and the multi-objective grouting construction plan is optimized by considering the optimal grouting disaster control effect, the lowest grouting construction cost, the most efficient construction process, etc. The numerical simulation and the actual grouting engineering data are fused by complementary data methods to obtain a more complete global grouting slurry flow situation, generalize the diffusion of grouting slurry in the time and space dimensions, and establish a grouting disaster control optimization model. Based on the attribute value disaster prediction and grouting plan optimization, the grouting disaster control monitoring plan is optimized and selected, and a grouting disaster control monitoring model is established.

[0021] Based on the underground engineering disaster prediction model, grouting disaster control scheme optimization model and grouting disaster control monitoring model, a digital twin system for underground engineering disaster grouting prevention and control is established. By inputting the actual underground engineering preliminary data to calculate the disaster attribute value, the type and degree of the disaster is predicted, the grouting scheme is optimized and the grouting disaster control monitoring system is optimized. Considering the time and space dimensions, the data update cycle of the grouting disaster control process is set, the on-site monitoring data is fed back in real time, the data in the digital twin system is dynamically adjusted, and the grouting scheme and monitoring scheme are optimized again, so as to carry out real-time early warning and emergency management of grouting construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0023] Figure 1 This is a flow chart of the method of embodiment 1. DETAILED DESCRIPTION

[0024] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0025] Embodiment 1

[0026] This embodiment provides a real-time management and control method for underground engineering grouting disaster control based on digital twins;

[0027] like Figure 1 As shown in the figure, the real-time control method of underground engineering grouting disaster control based on digital twin includes:

[0028] S101: Construct a grouting engineering database; based on the grouting engineering database, establish an underground engineering disaster prediction model;

[0029] S102: Based on the grouting engineering database, build a grouting disaster control scheme optimization model;

[0030] S103: Establish a grouting disaster control monitoring model;

[0031] S104: Based on the underground engineering disaster prediction model, the grouting disaster control scheme optimization model and the grouting disaster control monitoring model, a digital twin model for grouting prevention and control of underground engineering disasters is established;

[0032] S105: Input the preliminary data of the actual underground project into the digital twin model of the underground project disaster grouting prevention and control, output the predicted disaster type and disaster degree, output the optimized grouting plan, and visualize the grouting disaster control process.

[0033] Furthermore, the S101: constructing a grouting engineering database, specifically including: geological data, geophysical inversion data, drilling sensing data and in-situ test data;

[0034] Among them, geological data include: engineering construction areas, groundwater distribution, and stratigraphic boundaries obtained through geological surveys;

[0035] Geophysical inversion data, including: water-rich, harmful gas-containing, karst and fault data;

[0036] Data sensed while drilling, including: physical and mechanical parameters, geochemical characteristics and fracture data;

[0037] In-situ test data, including: geostress, permeability and water inflow.

[0038] Furthermore, the step S101: establishing an underground engineering disaster prediction model based on a grouting engineering database further includes:

[0039] S101-1: Divide the data in the grouting engineering database into structured data and unstructured data;

[0040] S101-2: Extract information from unstructured data in the grouting engineering database;

[0041] S101-3: The missing data in the grouting engineering database are supplemented by interpolation;

[0042] S101-4: Based on the grouting engineering database, the geological body is divided into several spatial unit bodies. Each spatial unit body is a cube of a set size. The center point of each cube is named in the form of three-dimensional coordinates and disaster attribute values ​​to obtain an underground engineering disaster prediction model.

[0043] Furthermore, the S101-1: divides the data in the grouting engineering database into structured data and unstructured data, wherein the structured data includes: three-dimensional spatial area data of the engineering construction area, grouting parameters and underground disaster data; wherein the grouting parameters include: arrangement of the number of grouting holes, selection of grouting slurry, grouting speed, grouting pressure, grouting time and grouting volume; underground disaster data includes: water gushing time, water gushing volume, surrounding rock collapse; grouting monitoring data includes: arrangement of the number of monitoring instruments, monitored grouting speed and pressure.

[0044] Furthermore, the S101-1: divides the data in the grouting engineering database into structured data and unstructured data, wherein the unstructured data includes: formation characteristic data, formation environment data, and disaster property data, the formation characteristic data includes: physical and mechanical parameters, geochemical characteristics, and fractured karst media; the formation environment data includes: rich water, containing harmful gases, and broken zones; the disaster property data includes: water gushing volume, and harmful gas escape.

[0045] Furthermore, the S101-2: extracting information from the unstructured data in the grouting engineering database specifically includes:

[0046] Natural Language Processing (NLP) is used to extract information from unstructured data, represent the extracted information in structured text, and merge the structured text representation with the structured data. The extracted information includes: physical and mechanical parameters and geochemical characteristics, fractured karst media, fractured zone area and degree of fracture, water-rich and gushing amount range and size, harmful gas area, storage volume and escape range. A three-dimensional geological model is established based on structured data and structured text data extracted through NLP information.

[0047] Furthermore, the S101-4: based on the grouting engineering database, the geological body is divided into a number of space units, each of which is a cube of a set size, and the center point of each cube is named by a three-dimensional coordinate plus a disaster attribute value, to obtain an underground engineering disaster prediction model, including:

[0048] The divided spatial unit body is taken as the basic unit, and the data in the unit body is used as the characterization data of the entire unit body; the data in the unit body includes: water content, gas content, rock mass fragmentation degree, fracture karst medium, physical and mechanical parameters and geochemical characteristics;

[0049] The disaster attribute value A determined by each unit ix , expressed as:

[0050] A ix =A i -l i -x i ;

[0051] Among them, A i Indicates the type of disaster, such as flooding, landslide and rock burst, etc. i represents the disaster score, x i Indicates the corresponding scores added to the disaster attribute value score, such as water inflow A 1x =A 1 -l 1 -x1 , where A 1 Indicates the type of flood disaster, l 1 Indicates the disaster attribute value score, x 1 Indicates the additional score of each characterization data for the disaster attribute value; where x i The corresponding units and fractions can be calculated from known engineering data:

[0052] C(x) j,k =L k ;

[0053] Where C(x) j,k represents the disaster attribute value bonus matrix of the jth geological parameter in the kth known grouting project; L k is the disaster attribute value matrix of the kth known grouting project. An existing grouting project data can be selected to initialize a disaster attribute value according to its disaster situation as a reference standard. Other engineering disaster attribute values ​​are calculated accordingly. This reference standard value must remain unchanged in the same digital twin system.

[0054] Add up the disaster attribute values ​​determined by all the divided units:

[0055] For each disaster category, add up the values ​​of all units to get the disaster attribute value of each disaster category:

[0056]

[0057] Among them, A i Indicated as flood disaster category, Represents the comprehensive attribute value of the disaster in all cells. It represents the total added value of each characterization data to the disaster attribute value score, A ix represents the disaster attribute value of flood disaster, and k represents the number of units;

[0058] Then the sum of the disaster attribute values ​​on the entire three-dimensional geological model is obtained;

[0059] Analyze the proportion of each disaster attribute value, and cancel the disaster impact when it is lower than the set threshold;

[0060] According to the second item of the disaster attribute value Determine the scope and extent of the disaster for the current disaster category;

[0061] According to the third item of disaster attribute value Determine the specific form of the current disaster, including: rock burst signs before the landslide, rock burst frequency, water inrush disaster scope, water inrush volume, simultaneous water inrush and high temperature, and simultaneous water inrush and mud inrush;

[0062] Based on the training set, the convolutional neural network is trained to obtain an underground engineering disaster prediction model; the training set includes: disaster attribute values ​​of known disaster types.

[0063] It should be understood that the degree of disaster impact corresponding to different score values ​​is determined based on the data of each grouting project, and the type and degree of underground engineering disasters are predicted (for example, for a total score of 1000, if the water gushing disaster attribute value is 900, the landslide attribute value is 100, and the corresponding loose sandy soil stratum score item accounts for a higher score, then the type of disaster that occurs is: water gushing disaster brings mud and sand and the water gushing volume can reach a certain amount. If there is a high ground stress bonus, a rock burst may occur).

[0064] Furthermore, the S102: constructing a grouting disaster control scheme optimization model based on the grouting engineering database specifically includes:

[0065] S102-1: Evaluate and learn the various parameters of the grouting scheme in the grouting engineering database and the control of underground engineering disasters to obtain grouting attribute values; wherein the various parameters of the grouting scheme include: the number and arrangement of grouting holes, the size and shape of grouting holes, grouting pressure, grouting speed, slurry selection, grouting time, and grouting method;

[0066] S102-2: The grouting disaster control scheme optimization model is used to obtain an optimized disaster control scheme, wherein the grouting disaster control scheme optimization model is used to: based on the grouting attribute values, optimize the multi-objective grouting construction scheme while considering the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process.

[0067] Furthermore, the S102-1: evaluating and learning the various parameters of the grouting scheme in the grouting engineering database and the control of underground engineering disasters to obtain grouting attribute values ​​specifically includes:

[0068] Assume the time-dependent function of each grouting parameter adding points to the grouting property value, solve the matrix equation considering the same disaster in different engineering cases, and calculate the addition of each parameter to the grouting property value:

[0069] B(t+n) j,k =B(t) j,k +K[l(t) k -l(t+n) k ],j=1,2...,N,k=1,2...,M;

[0070] Among them, B(t+n) j,k B(t) is the scoring matrix of the grouting attribute value of the j-th grouting parameter at time t+n and time t in the k-th known grouting engineering data. i,jis the grouting attribute value bonus matrix of the jth grouting parameter at time t+n and the kth known grouting engineering data, k is the loss coefficient matrix of the conversion between the disaster attribute value and the grouting attribute value, l(t+n) j is the disaster attribute value score matrix of the k-th known data grouting project at time t+n, l(t) j is the disaster attribute value score matrix of the kth known data grouting project at time t, where M ≥ N.

[0071] Each time the calculation steps are updated, the score value of each grouting parameter solved by the equation for the grouting attribute value is recorded until the final time is reached. In different disaster situations, the score value of each grouting parameter for the grouting attribute value is solved separately; for each disaster, different engineering data combinations are performed, and the above equations are repeatedly solved. The score values ​​at each time point are averaged to obtain the final score value, and the curve function of each parameter score value and time is fitted using various methods such as least squares method and polynomial interpolation.

[0072] Furthermore, the S102-2: based on the grouting attribute values, the optimization solution of the multi-objective grouting construction scheme is performed while considering the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process, specifically including:

[0073] S102-21: Initialize individual samples: take the average attribute value of each grouting parameter as the first sample, take the average attribute value of each parameter as the middle point, expand the value to both sides of the value, and perform permutations and combinations as other samples; the average attribute value of each grouting parameter is that under the current disaster situation, there are Q influencing grouting parameters, and the value is divided into Q parts;

[0074] For each grouting scheme, the initial population is set to (N, 3), where N is the population size, 3 is the three target variables required for decision making: the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process. For the jth individual y i :

[0075] y j ={y j,1 ,y j,2 ,y j,3},j=1,2,…,N

[0076] Among them, y j,1 Indicates that the grouting attribute values ​​corresponding to different grouting parameters satisfy the numerical calculation relationship with the disaster attribute value, y j,2 Represents the grouting construction cost under the current grouting parameters; y j,3 It indicates the grouting construction efficiency under the current grouting parameters.

[0077] yj,1 Indicates that the grouting attribute values ​​corresponding to different grouting parameters satisfy the calculation relationship of the loss matrix coefficient K with the calculated engineering disaster attribute value. For detailed calculation, see B(t+n) in 102-1. j,k The calculation formula for .

[0078]

[0079] Among them, n is the number of grouting parameters, k is the type of grouting parameters, a is the unit quantitative cost of the kth grouting parameter in the dimensionless case, and b is the unit quantitative number of the kth grouting parameter in the dimensionless case;

[0080]

[0081] Where n is the number of grouting parameters, k is the type of grouting parameters, c is the added effect of the kth grouting parameter on the grouting property value per unit time in the dimensionless case, α k is the weight score value of the kth grouting parameter, which can be determined according to the type of disaster prevention and control;

[0082] S102-22: After the population is initialized, the optimization individuals are selected for the three decision targets that need to be decided at present, and the adjacent sample differences are used to perform mutation operations according to the set scaling factors;

[0083] The attribute value score of the target grouting disaster control effect is used as a criterion within the attribute value score fluctuation range to determine whether the newly added sample is within the set area;

[0084] Crossover step: select the midpoint parameter value of the existing samples as the test individual, and always select the optimal option among the three target decisions in the difference, mutation and crossover operations of each cycle step, and enter the next cycle operation until the final cycle time step is reached;

[0085] In the process of selecting optimization options, the optimization degree of each parameter in different intervals of the longitudinal scale is compared. If the difference in optimization degree is greater than the interval of the set threshold, the interval with the minimum optimization degree is removed;

[0086] Set the cycle time step. The value must be divisible by the value of the equal-valued extension span from the middle point of the initial population setting.

[0087] S102-23: Determine the grouting model based on the established three-dimensional geological model and grouting scheme; set the initial conditions and boundary conditions of the grouting simulation, divide the three-dimensional geological model into grids, and obtain the grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram of the grouting model through numerical simulation; correct the grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram through actual grouting engineering data to obtain the corrected grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram.

[0088] Furthermore, the S103: grouting disaster control monitoring model is used to:

[0089] The solution space is established, and multiple initial solutions are set. The initial solution is set as the monitoring value of the existing grouting engineering monitoring instrument. In the calculation of different time steps, the value that can be divided by the length of the interval of two adjacent initial solutions is used as the neighboring solution update. The domain solution is optimized through different initial solutions simultaneously. The judgment method is expressed as:

[0090]

[0091] Among them, P is the optimal solution, S n and S e are the monitoring overlap range of the generated neighboring solution and the monitoring overlap range of the existing solution, respectively. n and C e are the cost of the neighboring solution monitoring instrument generated and the cost of the existing solution monitoring instrument, respectively, n and P e are the generated neighboring solutions and existing solutions respectively;

[0092] Set the sample library that meets the minimum area condition during the calculation process, and set the maximum sample capacity for the sample library that does not meet the minimum cost condition. After the sample library reaches the maximum capacity, optimize the selection within the sample library and remove sample data;

[0093] After the time step update is completed, the optimal solution is compared with the data in the sample library. If there is a sample with an area smaller than the optimal solution, and the difference between the cost condition and the optimal solution is within the set threshold, the sample and the calculated optimal solution are listed together as optional solutions.

[0094] Further, S104: establishing a digital twin model for grouting prevention and control of underground engineering disasters based on the underground engineering disaster prediction model, the grouting disaster control scheme optimization model and the grouting disaster control monitoring model;

[0095] The underground engineering disaster prediction model, the grouting disaster control scheme optimization model and the grouting disaster control monitoring model are merged to obtain a digital twin model for grouting prevention and control of underground engineering disasters.

[0096] Furthermore, the S105: inputting the actual underground engineering preliminary data into the digital twin model of underground engineering disaster grouting prevention and control, outputting the predicted disaster type and disaster degree, outputting the optimized grouting plan, and visualizing the grouting disaster control process, specifically includes:

[0097] By inputting the preliminary data of actual underground engineering, the underground engineering disaster prediction model calculates the disaster attribute values, the underground engineering disaster prediction model predicts the type and degree of disaster, the grouting disaster control scheme optimization model optimizes the selection of grouting schemes, and the grouting disaster control monitoring model optimizes the layout of grouting disaster control monitoring conditions.

[0098] Furthermore, the method further comprises:

[0099] Taking into account the time and space dimensions, a data update cycle for the grouting disaster control process is set. In each update cycle, the global grouting slurry flow situation generalized in the grouting disaster control scheme optimization model is used to determine the three-dimensional geological unit characterization data that needs to be removed. Through real-time feedback from on-site monitoring data, the data in the digital twin system is dynamically adjusted and the disaster attribute value parameters are recalculated to optimize the selection of grouting schemes and monitoring schemes. Protection criteria for various types of disasters are formulated based on existing engineering data, and an early warning mechanism is provided for on-site construction personnel based on periodic feedback from underground engineering disaster data, so that engineering protection schemes can be dynamically adjusted within different cycle times.

[0100] The data of the entire process of grouting prevention and control of this underground project is fully recorded and imported into the digital twin system as a source of data optimization.

[0101] Furthermore, the method further comprises:

[0102] During the actual grouting process, the grouting plan is continuously adjusted through the real-time transmission of on-site data, the changes in various parameters of the grouting process are continuously updated, and the grouting disaster control effect is analyzed in real time.

[0103] It can realize the real-time conversion of online data and dynamic cloud maps, and display the changes of various attribute parameters in the process of grouting disaster control, the diffusion form of grouting slurry in three-dimensional space, the real-time effect of grouting disaster control such as the amount of water inflow, etc., and update the grouting plan in real time. It can realize the enlarged display of local space, which can display the disaster problems caused by various parameter changes that may occur in the system calculation process, and on this basis, carry out real-time early warning and emergency management of grouting construction safety.

[0104] Embodiment 2

[0105] This embodiment provides a real-time management and control system for underground engineering grouting disaster control based on digital twins;

[0106] The real-time management and control system for underground engineering grouting disaster control based on digital twin includes:

[0107] The first establishment module is configured to: construct a grouting engineering database; and establish an underground engineering disaster prediction model based on the grouting engineering database;

[0108] The second building module is configured to: construct a grouting disaster control scheme optimization model based on a grouting engineering database;

[0109] The third establishment module is configured to: establish a grouting disaster control monitoring model;

[0110] A fourth building module is configured to: establish a digital twin model of grouting prevention and control of underground engineering disasters based on an underground engineering disaster prediction model, a grouting disaster control scheme optimization model, and a grouting disaster control monitoring model;

[0111] The output module is configured to: input the actual preliminary data of underground engineering into the digital twin model of underground engineering disaster grouting prevention and control, output the predicted disaster type and disaster degree, output the optimized grouting plan, and visualize the grouting disaster control process.

[0112] It should be noted that the first establishment module, the second establishment module, the third establishment module, the fourth establishment module and the output module correspond to steps S101 to S105 in Embodiment 1, and the examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A real-time control method for underground engineering grouting disaster control based on digital twins, characterized by: include: Construct grouting engineering database; Based on the grouting engineering database, establish an underground engineering disaster prediction model; Based on the grouting engineering database, an optimization model for grouting disaster control scheme is constructed; Establish a grouting disaster control monitoring model; Based on the underground engineering disaster prediction model, grouting disaster control scheme optimization model and grouting disaster control monitoring model, a digital twin model for grouting prevention and control of underground engineering disasters is established; Input the actual preliminary data of underground engineering into the digital twin model of underground engineering disaster grouting prevention and control, output the predicted disaster type and disaster degree, output the optimized grouting plan, and visualize the grouting disaster control process; Based on the grouting engineering database, an optimization model for grouting disaster control scheme is constructed, which includes: The various parameters of the grouting scheme in the grouting engineering database are evaluated and studied for the control of underground engineering disasters to obtain grouting attribute values; among them, the various parameters of the grouting scheme include: the number and arrangement of grouting holes, the size and shape of grouting holes, grouting pressure, grouting speed, slurry selection, grouting time, and grouting method; The grouting disaster control scheme optimization model is used to obtain the optimized disaster control scheme, wherein the grouting disaster control scheme optimization model is used to: based on the grouting attribute value, optimize the multi-objective grouting construction scheme while considering the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process; Grouting disaster control monitoring model is used for: The solution space is established, and multiple initial solutions are set. The initial solution is set as the monitoring value of the existing grouting engineering monitoring instrument. In the calculation of different time steps, the value that can be divided by the length of the interval of two adjacent initial solutions is used as the neighboring solution update. The domain solution is optimized through different initial solutions simultaneously. The judgment method is expressed as: Among them, P is the optimal solution, S n and S e are the monitoring overlap range of the generated neighboring solution and the monitoring overlap range of the existing solution, respectively. n and C e are the cost of the neighboring solution monitoring instrument generated and the cost of the existing solution monitoring instrument, respectively, n and P e are the generated neighboring solutions and existing solutions respectively; Set the sample library that meets the minimum area condition during the calculation process, and set the maximum sample capacity for the sample library that does not meet the minimum cost condition. After the sample library reaches the maximum capacity, optimize the selection within the sample library and remove sample data; After the time step update is completed, the optimal solution is compared with the data in the sample library. If there is a sample with an area smaller than the optimal solution, and the difference between the cost condition and the optimal solution is within the set threshold, the sample and the calculated optimal solution are listed together as optional solutions.

2. The real-time management and control method for underground engineering grouting disaster control based on digital twins as claimed in claim 1 is characterized in that: Based on the grouting engineering database, an underground engineering disaster prediction model is established, which also includes: Divide the data in the grouting engineering database into structured data and unstructured data; Extract information from unstructured data in the grouting engineering database; The missing data in the grouting engineering database is supplemented by interpolation; Based on the grouting engineering database, the geological body is divided into several spatial units. Each spatial unit is a cube of a set size. The center point of each cube is named using three-dimensional coordinates and disaster attribute values ​​to obtain an underground engineering disaster prediction model.

3. The real-time management and control method for underground engineering grouting disaster control based on digital twin as claimed in claim 2 is characterized in that: Based on the grouting engineering database, the geological body is divided into several spatial units. Each spatial unit is a cube of a set size. The center point of each cube is named by three-dimensional coordinates plus disaster attribute values. The underground engineering disaster prediction model is obtained, including: The divided spatial unit body is taken as the basic unit, and the data in the unit body is used as the characterization data of the entire unit body; the data in the unit body includes: water content, gas content, rock mass fragmentation degree, fracture karst medium, physical and mechanical parameters and geochemical characteristics; The disaster attribute value A determined by each unit ix , expressed as: A ix =A i -l i -x i ; Among them, A i Indicates the disaster category, which includes flooding, landslide and rock burst. i represents the disaster score, x i Indicates the corresponding scores added to the disaster attribute value score, such as water inflow A 1x =A1-l1-x1, where A1 represents the type of flood disaster, l1 represents the score of the disaster attribute value, and x1 represents the score added by each characterization data to the disaster attribute value; i The corresponding units and fractions can be calculated from known engineering data: C(x) j,k =L k ; Where C(x) j,k represents the disaster attribute value bonus matrix of the jth geological parameter in the kth known grouting project; L k is the disaster attribute value matrix of the kth known grouting project. An existing grouting project data can be selected to initialize a disaster attribute value according to its disaster situation as a reference standard. Other engineering disaster attribute values ​​are calculated accordingly. This reference standard value must remain unchanged in the same digital twin system. Add up the disaster attribute values ​​determined by all the divided units: For each disaster category, add up the values ​​of all units to get the disaster attribute value of each disaster category: Among them, A i Indicated as flood disaster category, Represents the comprehensive attribute value of the disaster in all cells. It represents the total added value of each characterization data to the disaster attribute value score, A ix represents the disaster attribute value of flood disaster, and k represents the number of units; Then the sum of the disaster attribute values ​​on the entire three-dimensional geological model is obtained; Analyze the proportion of each disaster attribute value, and cancel the disaster impact when it is lower than the set threshold; According to the second item of the disaster attribute value Determine the scope and extent of the disaster for the current disaster category; According to the third item of disaster attribute value Determine the specific form of the current disaster, including: rock burst signs before the landslide, rock burst frequency, water inrush disaster scope, water inrush volume, simultaneous water inrush and high temperature, and simultaneous water inrush and mud inrush; Based on the training set, the convolutional neural network is trained to obtain an underground engineering disaster prediction model; the training set includes: disaster attribute values ​​of known disaster types.

4. The real-time management and control method for underground engineering grouting disaster control based on digital twins as claimed in claim 1 is characterized in that: The various parameters of the grouting scheme in the grouting engineering database are evaluated and studied for the control of underground engineering disasters to obtain the grouting attribute values, including: Assume the time-dependent function of each grouting parameter adding points to the grouting property value, solve the matrix equation considering the same disaster in different engineering cases, and calculate the addition of each parameter to the grouting property value: B(t+n) j,k =B(t) j,k +K[l(t) k -l(t+n) k ],j=1,2...,N,k=1,2...,M; Among them, B(t+n) j,k B(t) is the scoring matrix of the grouting attribute value of the j-th grouting parameter at time t+n and time t in the k-th known grouting engineering data. j,k is the grouting attribute value bonus matrix of the jth grouting parameter at time t+n and the kth known grouting engineering data at time t, k is the loss coefficient matrix of the conversion between disaster attribute value and grouting attribute value, l(t+n) k is the disaster attribute value score matrix of the k-th known data grouting project at time t+n, l(t) k is the disaster attribute value score matrix of the kth known data grouting project at time t, where M ≥ N.

5. The real-time management and control method for underground engineering grouting disaster control based on digital twins as claimed in claim 1 is characterized in that: Based on the grouting property values, the multi-objective grouting construction scheme is optimized to consider the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process, including: Initialize the individual samples: take the average attribute value of each grouting parameter as the first sample, take the average attribute value of each parameter as the midpoint, expand it to both sides of the value with equal value, and make permutations and combinations as other samples; the average attribute value of each grouting parameter is that under the current disaster situation, there are Q influencing grouting parameters, and the value is divided into Q parts; For each grouting scheme, the initial population is set to (N, 3), where N is the population size, 3 is the three target variables required for decision making: the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process. For the jth individual y j : and j ={and j,1 ,and j,2 ,and j,3 },j=1,2,…,N Among them, y j,1 Indicates that the grouting attribute values ​​corresponding to different grouting parameters satisfy the numerical calculation relationship with the disaster attribute value, y j,2 Represents the grouting construction cost under the current grouting parameters; y j,3 Indicates the grouting construction efficiency under the current grouting parameters; After the population is initialized, the optimization individuals of the three decision-making targets that need to be decided are selected, and the adjacent sample differences are used to perform mutation operations according to the set scaling factor; the attribute value score of the target grouting disaster control effect is used as the judgment criterion within the attribute value score fluctuation range to determine whether the newly added samples are within the set area; crossover step: select the midpoint parameter value of the existing samples as the test individual, and always select the preferred option among the three target decisions in the difference, mutation and crossover operations of each cycle step, and enter the next cycle operation until the final cycle time step is reached; in the process of selecting the optimization options, the optimization degree of each parameter in different intervals of the longitudinal scale is compared. If the optimization degree difference is greater than the interval of the set threshold, the interval with the minimum optimization degree is removed; set the cycle time step, and the value must be divisible by the value of the equal-value extension span of the middle point set by the initialized population; Based on the established three-dimensional geological model and grouting scheme, the grouting model is determined; the initial conditions and boundary conditions of the grouting simulation are set, the three-dimensional geological model is gridded, and the grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram of the grouting model is obtained through numerical simulation. The grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram is corrected through actual grouting engineering data to obtain the corrected grouting slurry diffusion time and velocity, pressure and multi-phase distribution dynamic diagram.

6. The real-time management and control method for underground engineering grouting disaster control based on digital twins as claimed in claim 1 is characterized in that: The actual preliminary data of underground engineering is input into the digital twin model of underground engineering disaster grouting prevention and control, and the predicted disaster type and degree are output, as well as the optimized grouting plan, to visualize the grouting disaster control process, including: By inputting the preliminary data of actual underground engineering, the underground engineering disaster prediction model calculates the disaster attribute values, the underground engineering disaster prediction model predicts the type and degree of disaster, the grouting disaster control scheme optimization model optimizes the selection of grouting schemes, and the grouting disaster control monitoring model optimizes the layout of grouting disaster control monitoring conditions.

7. The real-time management and control method for underground engineering grouting disaster control based on digital twins as claimed in claim 1 is characterized in that: The method further comprises: Taking into account the time and space dimensions, a data update cycle for the grouting disaster control process is set. In each update cycle, the global grouting slurry flow situation generalized in the grouting disaster control scheme optimization model is used to determine the three-dimensional geological unit characterization data that needs to be removed. Through real-time feedback from on-site monitoring data, the data in the digital twin system is dynamically adjusted and the disaster attribute value parameters are recalculated to optimize the selection of grouting schemes and monitoring schemes. Protection criteria for various types of disasters are formulated based on existing engineering data, and an early warning mechanism is provided for on-site construction personnel based on periodic feedback from underground engineering disaster data, so that engineering protection schemes can be dynamically adjusted within different cycle times.

8. The real-time management and control system for underground engineering grouting disaster control based on digital twin is characterized by: include: The first establishment module is configured to: construct a grouting engineering database; Based on the grouting engineering database, establish an underground engineering disaster prediction model; The second building module is configured to: construct a grouting disaster control scheme optimization model based on the grouting engineering database; specifically includes: The various parameters of the grouting scheme in the grouting engineering database are evaluated and studied for the control of underground engineering disasters to obtain grouting attribute values; among them, the various parameters of the grouting scheme include: the number and arrangement of grouting holes, the size and shape of grouting holes, grouting pressure, grouting speed, slurry selection, grouting time, and grouting method; The grouting disaster control scheme optimization model is used to obtain the optimized disaster control scheme, wherein the grouting disaster control scheme optimization model is used to: based on the grouting attribute value, optimize the multi-objective grouting construction scheme while considering the optimal grouting disaster control effect, the lowest grouting construction cost, and the most efficient construction process; The third establishment module is configured to: establish a grouting disaster control monitoring model; the grouting disaster control monitoring model is used to: The solution space is established, and multiple initial solutions are set. The initial solution is set as the monitoring value of the existing grouting engineering monitoring instrument. In the calculation of different time steps, the value that can be divided by the length of the interval of two adjacent initial solutions is used as the neighboring solution update. The domain solution is optimized through different initial solutions simultaneously. The judgment method is expressed as: Among them, P is the optimal solution, S n and S e are the monitoring overlap range of the generated neighboring solution and the monitoring overlap range of the existing solution, respectively. n and C e are the cost of the neighboring solution monitoring instrument generated and the cost of the existing solution monitoring instrument, respectively, n and P e are the generated neighboring solutions and existing solutions respectively; Set the sample library that meets the minimum area condition during the calculation process, and set the maximum sample capacity for the sample library that does not meet the minimum cost condition. After the sample library reaches the maximum capacity, optimize the selection within the sample library and remove sample data; After the time step update is completed, the optimal solution is compared with the data in the sample library. If there is a sample with a smaller area than the optimal solution, and the difference between the cost condition and the optimal solution is within the set threshold, the sample and the calculated optimal solution are listed together as optional solutions; A fourth building module is configured to: establish a digital twin model of grouting prevention and control of underground engineering disasters based on an underground engineering disaster prediction model, a grouting disaster control scheme optimization model, and a grouting disaster control monitoring model; The output module is configured to: input the actual preliminary data of underground engineering into the digital twin model of underground engineering disaster grouting prevention and control, output the predicted disaster type and disaster degree, output the optimized grouting plan, and visualize the grouting disaster control process.

Images