Modular tunnel construction management and control method and system based on digital twinning

Abstract

The invention relates to the technical field of data processing, and discloses a modular tunnel construction management and control method and system based on digital twinning, and the method comprises the steps: collecting historical construction data, carrying out the preprocessing of the historical construction data, obtaining a structured data set, extracting a feature matrix from the structured data set, and applying an improved clustering algorithm to obtain a clustering result, mining an association rule, generating an association rule set, and converting the association rule set into an optimal parameter combination. According to the invention, through a digital twin model fusing a physical model and a professional model, full-process virtual construction and real-time mapping of a tunnel construction environment can be realized, and the digital integration level and the visualization level of construction management and control are greatly improved; by means of the AI-based visual analysis technology, construction safety production conditions and key operation parameters can be automatically recognized and confirmed, the recognition accuracy and response speed are remarkably improved, and the process conversion efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a modular tunnel construction management and control method and system based on digital twins. Background Technology

[0002] Tunnel construction management refers to the management activities conducted during tunnel engineering construction to ensure the smooth achievement of objectives such as construction safety, quality, schedule, and cost through systematic planning, organization, and supervision. Its core lies in coordinating human, material, and equipment resources through scientific methods and technical means, optimizing construction processes, and preventing potential risks. Management content includes developing construction plans, monitoring geological conditions, standardizing work procedures, implementing safety measures, and controlling material quality and costs.

[0003] However, traditional tunnel construction management faces several challenges. First, the management model is relatively simplistic, relying primarily on manual experience and static planning, lacking dynamic simulation and virtual construction capabilities based on real-time data, making it difficult to effectively predict construction risks and optimize parameters. Second, the automation level of process connections is low, with key construction conditions relying on manual judgment and lacking automation mechanisms, easily leading to construction delays and safety risks. Furthermore, the existing construction management systems suffer from insufficient digital integration; the physical construction environment and digital management models are independent, lacking effective fusion and failing to achieve full-process virtual construction and management supported by digital twin technology. Finally, intelligent identification capabilities are lacking; existing systems mainly rely on manual supervision, lacking AI-based monitoring and analysis and automatic identification of key operating conditions, making it difficult to achieve automated confirmation and early warning of construction safety. These problems limit the efficiency and safety of the construction process. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a modular tunnel construction management and control method and system based on digital twins.

[0005] To achieve the above objectives, the present invention provides the following technical solution: Firstly, a modular tunnel construction management and control method based on digital twins is provided, the method comprising: S101: Obtain three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to construct physical models and professional models respectively, define semantic mapping rules between physical models and professional models, establish dynamic mapping relationships between physical models and professional models, and obtain digital twin models. S102: Collect historical construction data, preprocess the historical construction data to obtain a structured dataset, extract the feature matrix from the structured dataset, apply the improved clustering algorithm to obtain the clustering results and mine the association rules, generate an association rule set, and transform the association rule set into the optimal parameter combination. S103: Based on the digital twin model, the optimal parameter combination is simulated and predicted using finite element analysis and discrete algorithms, and the optimal parameter combination is iteratively optimized. S104: Use a digital twin model to collect data streams in real time, combine AI to identify the data streams, obtain process identification results, judge the process identification results, and issue process conversion instructions.

[0006] Furthermore, the improved clustering algorithm is used to obtain clustering results, and the specific steps are as follows: Geological condition classification field is extracted from the feature matrix for each sample, geological difference matrix between each sample is calculated, and probability distribution strategy is used to obtain the selection probability of each sample. The sample pair with the largest address difference is selected first as the initial center point set. Each sample is assigned to the initial center point closest to it, resulting in multiple clusters. The center of the cluster is updated by calculating the mean of all samples in the cluster. The samples are then reassigned to the new cluster center, and the distance the center point moves is calculated each time. Set a movement threshold. If the movement distance of the center point is less than the movement threshold, stop clustering and obtain the clustering result.

[0007] Furthermore, the specific steps for obtaining the initial center point are as follows: A1: Construct a geological category vector based on the geological condition classification field, calculate the Euclidean distance for each pair of samples, and obtain the geological dissimilarity matrix; A2: Randomly select the first center point, calculate the minimum squared distance from each sample to the center point, and generate a selection probability table; A3: Randomly select secondary center points according to the selected probability distribution table, calculate the geological difference between the first center point and the secondary center points, and use them as the candidate center point set to verify whether the geological difference of the candidate center point set meets the basic threshold. If the conditions are met, the candidate centroid set is determined as the initial centroid set. If the conditions are not met, steps A2 to A3 are repeated until a sample pair that meets the conditions is found.

[0008] Furthermore, the specific steps for mining association rules include: The continuous eigenvalues ​​in the feature matrix are transformed into discrete levels, and key-value pairs are generated for each sample to generate a transaction dataset. The Apriori algorithm is used to scan the transaction dataset, generate frequent itemsets with support greater than or equal to N, extract rules with confidence greater than or equal to M from the frequent itemsets, and verify the validity of the rules to obtain the association rule set.

[0009] Furthermore, the specific steps for simulating and predicting the optimal parameter combination using finite element analysis and the discrete element method include: Finite element meshes are generated based on digital twin models, with ground stress and construction loads used as boundary conditions, displacement constraints are set, and the output is a finite element file. Nonlinear finite element analysis based on finite element files is used to simulate the plastic deformation of rock mass, evolve the stress field throughout the construction cycle, calculate key indicators, and obtain stress cloud diagrams and deformation thermograms. The rock mass is simultaneously discretized into multiple particles, and contact parameters between the particles are set. Based on the contact parameters, the cyclic load of the contact force is counted by the rain flow counting method, and the overall deformation of the support structure is calculated by using the displacement field of the particles, so as to obtain the displacement curve of the support structure and the distribution map of the rock mass fragmentation degree. The maximum displacement and stress values ​​of the stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are compared with the threshold values ​​of the construction specifications to obtain a safety assessment report. The stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are then superimposed onto the digital twin model to synthesize a simulation video.

[0010] Furthermore, the optimal parameter combination is iteratively optimized, and the specific steps include: A genetic algorithm is used to iteratively optimize the optimal parameter combination to obtain an optimized parameter combination. Based on the optimized parameter combination, simulation and prediction are performed again to determine whether the requirements are met. The optimized parameter combination after successful verification is simultaneously fed back to the structured dataset.

[0011] Furthermore, the specific steps of using a genetic algorithm to iteratively optimize the optimal parameter combination include: B1: Randomly generate multiple sets of parameter combinations as the initial population matrix, randomly select two initial population matrices to determine the intersection point position, and generate the second-generation population matrix. B2: For each second-generation population matrix, perform Gaussian mutation on each parameter with probability 'a' to obtain the mutated population matrix; B3: Based on the safety assessment report and construction efficiency design objective function, obtain the objective function value of each set of parameters, select the excellent parameter combination based on the objective function value, and use it as the selection population matrix; B4: Count the number of iterations. If the number of iterations reaches H, stop the iteration. Otherwise, repeat B1 to B3.

[0012] Secondly, a modular tunnel construction management and control system based on digital twins is provided, which is implemented based on the aforementioned modular tunnel construction management and control method based on digital twins. The system includes: The model building module acquires three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to build physical and professional models respectively, defines semantic mapping rules between the physical and professional models, establishes dynamic mapping relationships between the physical and professional models, and obtains a digital twin model. The construction parameter module collects historical construction data, preprocesses the historical construction data to obtain a structured dataset, extracts the feature matrix from the structured dataset, applies an improved clustering algorithm to obtain clustering results and mine association rules, generates an association rule set, and transforms the association rule set into the optimal parameter combination. The parameter optimization module, relying on the digital twin model, uses finite element analysis and discrete element method to simulate and predict the optimal parameter combination, and iteratively optimizes the optimal parameter combination. The process management module uses a digital twin model to collect data streams in real time, combines AI to identify the data streams, obtain process identification results, judge the process identification results, and issue process conversion instructions.

[0013] Thirdly, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the aforementioned modular tunnel construction control method based on digital twins.

[0014] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed, implements the aforementioned modular tunnel construction control method based on digital twins.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This application discloses a modular tunnel construction management and control method and system based on digital twins, including: collecting historical construction data, preprocessing the historical construction data to obtain a structured dataset, extracting a feature matrix from the structured dataset, applying an improved clustering algorithm to obtain clustering results and mine association rules, generating an association rule set, and transforming the association rule set into an optimal parameter combination. This invention, through a digital twin model that integrates physical and professional models, can realize the virtual construction and real-time mapping of the entire tunnel construction environment, greatly improving the digital integration and visualization level of construction management and control. With the help of AI-based visual analysis technology, it can automatically identify and confirm construction safety production conditions and key operational parameters, significantly improving identification accuracy and response speed, and automatically prompting the next process based on process connection relationships, realizing automated parameter control within processes and semi-automated management of process transitions. Furthermore, by mining historical construction data and analyzing machine learning algorithms, a construction index optimization model was constructed, which improved construction efficiency and enhanced risk warning capabilities, thereby achieving precise and forward-looking management of the construction process. To improve the efficiency of transitions between procedures, an intelligent connection and automatic prompting mechanism was also established, effectively shortening the waiting time for procedure transitions. This invention significantly improves the level of intelligence in tunnel construction management, optimizes the construction management process, and promotes the refinement and efficiency of the entire construction process. Attached Figure Description

[0016] Figure 1 A flowchart of a modular tunnel construction control method based on digital twin provided by the present invention; Figure 2 A schematic diagram of the modular structure of a modular tunnel construction control system based on digital twin provided by the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided by the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 Please see Figure 1 As shown in the figure, this embodiment discloses a modular tunnel construction management and control method based on digital twins, the method comprising: S101: Obtain three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to construct physical models and professional models respectively, define semantic mapping rules between physical models and professional models, establish dynamic mapping relationships between physical models and professional models, and obtain digital twin models. Among them, the three-dimensional spatial data of the tunnel construction area can be obtained by using a three-dimensional laser scanner to scan the entire cross section of the tunnel body, while the construction parameter data is collected in real time by deploying an IoT sensor network (such as GPS locators, attitude sensors, and environmental monitoring sensors) to collect construction parameters such as the position of mechanical equipment, vibration status, temperature / humidity, etc.

[0019] Meanwhile, relying on technologies such as BIM modeling, laser scanning and IoT sensing, physical models are constructed and dynamically updated. The three-dimensional spatial data of the tunnel construction area serves as the basic geometric input for BIM modeling, and sensor data is bound to the equipment entities in the BIM model by ID to form a dynamic attribute mapping. Furthermore, the training of the physical model is generated by registering the three-dimensional spatial data of the tunnel construction area with the CAD model using the ICP (Iterative Closest Point) algorithm, which does not require supervised training and directly relies on the scanning accuracy.

[0020] The professional model covers business logic models such as construction management process, process condition library, and material distribution mechanism. Its training adopts supervised learning to pre-train baseline parameters, and then iteratively optimizes them through genetic algorithm. Semantic mapping rules are defined between physical model and professional model, such as the steel bar binding process corresponding to the steel bar entity in BIM model.

[0021] Furthermore, a relational connection between the physical model and the professional model is established through an SQL database. For example, the strength grade parameter of the concrete pouring process is bound to the concrete component attributes in the BIM model, and the bound data supports bidirectional queries.

[0022] Meanwhile, in order to achieve a two-way synchronization mechanism, this embodiment uses the Flink processing framework to clean, align and aggregate the three-dimensional spatial data and IoT sensor data of the laser scanning tunnel construction area in real time, so as to output a real-time data stream as input for business logic judgment.

[0023] Business decision instructions are mapped to execution devices in the physical environment through API interfaces, such as adjusting the posture of a robotic arm or updating the status of a BIM model.

[0024] S102: Collect historical construction data, preprocess the historical construction data to obtain a structured dataset, extract the feature matrix from the structured dataset, apply the improved clustering algorithm to obtain the clustering results and mine the association rules, generate an association rule set, and transform the association rule set into the optimal parameter combination. Historical construction data includes tunnel geometric deviations, mechanical condition parameters, ambient temperature and humidity, process aging, support parameters, rock hardness, etc., and its preprocessing includes: Linear interpolation was used for missing mechanical location data, and missing geological condition values ​​were filled in by nearby construction logs. For outliers, box plots were used to identify outliers, and construction logs were used to verify whether they were data entry errors, resulting in a cleaned dataset. Based on the process conditions, highly correlated parameters are selected from the cleaned dataset to generate a structured dataset. Then, the continuous parameters in the structured dataset are standardized, while the classification parameters are one-hot encoded to obtain the feature matrix.

[0025] As a specific implementation method, the improved clustering algorithm is used to obtain clustering results, and the specific steps are as follows: Geological condition classification field is extracted from the feature matrix for each sample, geological difference matrix between each sample is calculated, and probability distribution strategy is used to obtain the selection probability of each sample. The sample pair with the largest address difference is selected first as the initial center point set. For example, selecting the two furthest samples from the hard rock sample set, or choosing the boundary sample between hard and soft rock as the center point, can be used as a specific example. The first center point was selected from the hard rock sample set with an anchor spacing of 1.5 meters and a water-cement ratio of 0.45.

[0026] From the soft rock sample set, a sample with an anchor spacing of 0.8 meters and a water-cement ratio of 0.5 was selected as the second center point.

[0027] Each sample is assigned to the initial center point closest to it, resulting in multiple clusters. The center of the cluster is updated by calculating the mean of all samples in the cluster. The samples are then reassigned to the new cluster center, and the distance the center point moves is calculated each time. Set a movement threshold. If the movement distance of the center point is less than the movement threshold, stop clustering and obtain the clustering result.

[0028] It should be noted that the moving threshold is usually selected as 0.001 to 0.01, and it can be selected according to the size of the data and the range of features.

[0029] In addition, silhouette coefficient verification can be performed to measure the clustering effect, and the process is as follows: First, calculate the average distance from a sample to other samples in the same cluster, which is used as the intra-cluster compactness. Then, calculate the average distance from a sample to the nearest sample in another cluster, which is used as the inter-cluster separation. The silhouette coefficient is then obtained using the silhouette coefficient calculation formula, which is as follows: ; In the formula, Cluster compactness The silhouette coefficient measures the density of a sample within a cluster and the density of the sample with the nearest cluster. A higher value indicates a better clustering effect.

[0030] The specific steps for obtaining the initial center point are as follows: A1: Construct a geological category vector based on the geological condition classification field, calculate the Euclidean distance for each pair of samples, and obtain the geological dissimilarity matrix; For example, the Euclidean distance between sample A (hard rock = 1, soft rock = 0) and sample B (hard rock = 0, soft rock = 1) is √2. A2: Randomly select the first center point, calculate the minimum squared distance from each sample to the center point, and generate a selection probability table; A3: Randomly select secondary center points according to the selected probability distribution table, calculate the geological difference between the first center point and the secondary center points, and use them as the candidate center point set to verify whether the geological difference of the candidate center point set meets the basic threshold. If the conditions are met, the candidate centroid set is determined as the initial centroid set. If the conditions are not met, steps A2 to A3 are repeated until a sample pair that meets the conditions is found.

[0031] The specific steps for mining association rules include: The continuous eigenvalues ​​in the feature matrix are transformed into discrete levels, and key-value pairs are generated for each sample to generate a transaction dataset. The Apriori algorithm is used to scan the transaction dataset, generate frequent itemsets with support greater than or equal to N, extract rules with confidence greater than or equal to M from the frequent itemsets, and verify the validity of the rules by checking for redundant items (i.e., the rules contain extra conditions) or contradictory logic (such as causal conflict between antecedent and consequent), thereby obtaining the association rule set.

[0032] The Apriori algorithm is a classic association rule mining algorithm, primarily used to discover frequent itemsets in a database and generate association rules based on these frequent itemsets. This algorithm expands frequent itemsets incrementally; if an itemset is frequent, then all its subsets must also be frequent. The basic process of the Apriori algorithm includes: first, scanning the dataset to find frequent 1-itemsets; then, iteratively generating candidate k+1 itemsets from the frequent k-itemsets, and scanning the dataset again to calculate support, until no new frequent itemsets can be found.

[0033] S103: Based on the digital twin model, the optimal parameter combination is simulated and predicted using finite element analysis and discrete algorithms, and the optimal parameter combination is iteratively optimized. The specific steps for simulating and predicting the optimal parameter combination using finite element analysis and discrete element method include: Finite element meshes are generated based on digital twin models, with ground stress and construction loads used as boundary conditions, displacement constraints are set, and the output is a finite element file. Nonlinear finite element analysis based on finite element files is used to simulate the plastic deformation of rock mass, evolve the stress field throughout the construction cycle, calculate key indicators, and obtain stress cloud diagrams and deformation thermograms. The rock mass is simultaneously discretized into multiple particles, and contact parameters between the particles are set. Based on the contact parameters, the cyclic load of the contact force is counted by the rain flow counting method, and the overall deformation of the support structure is calculated by using the displacement field of the particles, so as to obtain the displacement curve of the support structure and the distribution map of the rock mass fragmentation degree. It should be noted that the contact parameters between particles define their mechanical behavior during contact, including: Contact stiffness: This affects the intensity of the particle's reaction upon contact, that is, the resistance between two particles when they come into contact. Higher contact stiffness results in a greater rebound force from the collision between particles, while lower stiffness leads to a softer contact.

[0034] Coefficient of friction: This determines the resistance to particle sliding on the contact surface. Higher friction makes it more difficult for particles to slide, suitable for simulating adhesion or resistance between particles, such as in the simulation of bulk materials or soil. A lower coefficient of friction makes it easier for particles to slide.

[0035] Coefficient of restitution: This refers to the degree to which particles recover after a collision. A higher coefficient of restitution means less energy loss after the collision, and the particles are more likely to bounce back, which is suitable for materials with high elasticity; conversely, a lower coefficient of restitution means greater energy loss, and the particles bounce back at a lower speed and height.

[0036] Adhesion: The adhesion force between particles can simulate the bonding force between particles, affecting the degree of particle aggregation and the adhesion behavior between particles. This parameter is particularly important when processing powders, granular mixtures, or clay-like materials.

[0037] Furthermore, by setting these contact parameters, various particle behaviors can be simulated: Compression behavior: When the contact stiffness is high, the compressive deformation between particles is small, and vice versa. Different materials and different particle structures will exhibit different compressive characteristics.

[0038] Friction and slip: The coefficient of friction controls the slip behavior between particles. At a high coefficient of friction, particles tend to block each other and accumulate; at a low coefficient of friction, particles slide more easily, which is suitable for simulating fluid or loose materials.

[0039] Collision and Elastic Recovery: The coefficient of restitution affects the elastic recovery of particles after a collision, determining how the particles bounce back. This parameter has a direct impact on energy loss during the collision process and the particle's motion state.

[0040] The maximum displacement and stress values ​​of the stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are compared with the threshold values ​​of the construction specifications to obtain a safety assessment report. The stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are then superimposed onto the digital twin model to synthesize a simulation video.

[0041] The method involves comparing stress cloud maps, deformation heat maps, support structure displacement curves, and rock mass fragmentation distribution maps with construction specification thresholds. Combined with digital twin models for visualization and simulation video synthesis, this allows for real-time and intuitive assessment of project safety. It helps engineers more accurately analyze potential risks and optimize construction plans. This approach not only improves the efficiency and accuracy of safety management but also provides dynamic monitoring during construction, enabling timely detection and resolution of problems, reducing the risk of accidents, and providing data support for later project maintenance and management.

[0042] For example, the iterative optimization of the optimal parameter combination includes the following specific steps: A genetic algorithm is used to iteratively optimize the optimal parameter combination to obtain an optimized parameter combination. Based on the optimized parameter combination, simulation and prediction are performed again to determine whether the requirements are met. The optimized parameter combination after successful verification is simultaneously fed back to the structured dataset.

[0043] The specific steps of using a genetic algorithm to iteratively optimize the optimal parameter combination include: B1: Randomly generate multiple sets of parameter combinations as the initial population matrix, randomly select two initial population matrices to determine the intersection point position, and generate the second-generation population matrix. B2: For each second-generation population matrix, Gaussian mutation is performed on each parameter with probability 'a' to obtain the mutated population matrix. In this embodiment, 'a' is set to 5%. The 5% Gaussian mutation probability is based on a balanced design, aiming to introduce appropriate randomness through a small number of mutations to avoid the algorithm getting trapped in local optima while preserving the desirable characteristics of most offspring. This probability ensures the algorithm remains stable during optimization.

[0044] B3: Based on the safety assessment report and construction efficiency design objective function, obtain the objective function value of each set of parameters, select the excellent parameter combination based on the objective function value, and use it as the selection population matrix; Its objective function is: ; In this embodiment, The value is 0.6, is 0.4. Its value is based on the "Safety Code for Tunnel Engineering Construction". Safety risk control is the primary constraint for optimizing construction parameters, and it is stipulated that the structural safety weight shall not be less than 60%.

[0045] B4: Count the number of iteration times. When it reaches H times, stop the iteration; otherwise, repeat B1 to B3. And the value of H is 100, which is usually determined according to algorithm convergence, computing resources, time efficiency, and engineering experience. 100 times is sufficient to ensure that the algorithm can effectively optimize the solution and avoid excessive computational redundancy.

[0046] S104: Use the digital twin model to collect data streams in real time, combine with AI to identify the data streams, obtain the process identification results, judge the process identification results, and issue process conversion instructions; It should be noted that the specific steps for combining AI to identify the data streams include: Use the YOLOv5 algorithm to perform object detection on the data streams, identify the gestures of construction workers and the operating states of machinery (such as the position of the concrete vibrating rod), and classify them to judge the type of current construction process (such as steel bar binding / concrete pouring).

[0047] Meanwhile, before judging the process identification results, key confirmation conditions for each construction module such as full-section steel bar binding, concrete pouring, concrete vibrating with a group of rods, and standard concrete curing need to be defined.

[0048] Among them, the key confirmation conditions include: for full-section steel bar binding, the process is confirmed to be qualified when the steel bar specifications and quantities are consistent with the BIM model, the spacing and cover thickness are qualified through machine vision detection, and the binding of key nodes is qualified; for concrete pouring, the process is confirmed to be qualified when its mix ratio and strength grade are consistent with the design requirements, and on-site measurement ensures that there are no quality problems such as concrete segregation and initial setting; for concrete pouring, the process is confirmed to be qualified when the vibrating track coverage rate is 100% and the density meets the standard; for standard curing, the process is confirmed to be qualified when the actual curing duration of the concrete meets the standard.

[0049] If the current process is judged unqualified, the construction parameters can be adjusted through the PLC. For example, if the steel bar binding spacing exceeds the tolerance, automatically adjust the positioning accuracy of the robotic arm to the design value; if the concrete vibrating coverage rate is insufficient, extend the vibrating time to the qualified value.

[0050] If the current process is judged qualified, trigger the next process prompt through the industrial control system and output the process conversion instruction.

[0051] In addition, to ensure the coordinated operation of modules such as model building, construction parameters, parameter optimization, and process management, a unified data bus and message queue mechanism is used to realize real-time data transmission and status synchronization between modules, ensuring the integrated operation and real-time response of the entire system, and establishing data interaction interfaces and coordination mechanisms between various functional modules.

[0052] This mechanism includes: using Kafka to build a high-throughput message middleware, supporting TOPIC partitioning and consumer group load balancing to ensure the real-time distribution of tens of millions of messages per day, defining RESTful API + JSON Schema data format specifications, and unifying the interface protocols for model building, construction parameters, parameter optimization, and process management.

[0053] Each module publishes data through TOPIC, while other modules subscribe to relevant TOPICs through consumer groups. Data traceability is achieved using timestamps and model version numbers to ensure data version consistency during state synchronization.

[0054] Meanwhile, for scenarios requiring low latency, the MQTT protocol is used to ensure that construction instructions can be transmitted quickly and reliably; for scenarios requiring high throughput, Kafka is used to process large-scale data streams (such as three-dimensional spatial data of laser scanning tunnel construction areas), and network bandwidth usage is optimized through batch processing and data compression.

[0055] Each module periodically sends heartbeat packets to check if it is operating normally. If no response is received within a timeout period, a fault alarm is triggered and the module is switched to a backup module. Incremental transmission is used for data updates to reduce bandwidth pressure.

[0056] Example 2 Please see Figure 2 As shown, based on the same inventive concept, this embodiment discloses a modular tunnel construction management and control system based on digital twins. For details not covered in this embodiment, please refer to the relevant sections of Embodiment 1. The system includes: The model building module acquires three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to build physical and professional models respectively, defines semantic mapping rules between the physical and professional models, establishes dynamic mapping relationships between the physical and professional models, and obtains a digital twin model. The construction parameter module collects historical construction data, preprocesses the historical construction data to obtain a structured dataset, extracts the feature matrix from the structured dataset, applies an improved clustering algorithm to obtain clustering results and mine association rules, generates an association rule set, and transforms the association rule set into the optimal parameter combination. The parameter optimization module, relying on the digital twin model, uses finite element analysis and discrete element method to simulate and predict the optimal parameter combination, and iteratively optimizes the optimal parameter combination. The process management module uses a digital twin model to collect data streams in real time, combines AI to identify the data streams, obtain process identification results, judge the process identification results, and issue process conversion instructions.

[0057] Example 3 Please see Figure 3 As shown, this embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the modular tunnel construction control method based on digital twins provided by the above methods.

[0058] Since the electronic device described in this embodiment is the electronic device used to implement the modular tunnel construction management and control method based on digital twins in this application embodiment, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the modular tunnel construction management and control method based on digital twins in this application embodiment. Therefore, how the electronic device implements the method in this application embodiment will not be described in detail here. Any electronic device used by those skilled in the art to implement the modular tunnel construction management and control method based on digital twins in this application embodiment falls within the scope of protection of this application.

[0059] Example 4 This embodiment discloses a computer-readable storage medium, including a memory, a processor, and a computer program stored on the memory and running on the processor. When the processor executes the computer program, it implements the modular tunnel construction control method based on digital twins provided by the above methods.

[0060] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters, weights, and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0061] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired or wireless network. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0062] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0063] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0064] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0065] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0066] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0067] 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.

[0068] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A modular tunnel construction control method based on digital twins, characterized in that, The method includes: S101: Obtain three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to construct physical models and professional models respectively, define semantic mapping rules between physical models and professional models, establish dynamic mapping relationships between physical models and professional models, and obtain digital twin models. S102: Collect historical construction data, preprocess the historical construction data to obtain a structured dataset, extract the feature matrix from the structured dataset, apply the improved clustering algorithm to obtain clustering results and mine association rules, generate an association rule set, and transform the association rule set into the optimal parameter combination. S103: Based on the digital twin model, the optimal parameter combination is simulated and predicted using finite element analysis and discrete algorithms, and the optimal parameter combination is iteratively optimized. S104: Use a digital twin model to collect data streams in real time, combine AI to identify the data streams, obtain process identification results, judge the process identification results, and issue process conversion instructions.

2. The modular tunnel construction control method based on digital twins according to claim 1, characterized in that, The improved clustering algorithm is used to obtain clustering results. The specific steps are as follows: Geological condition classification field is extracted from the feature matrix for each sample, geological difference matrix between each sample is calculated, and probability distribution strategy is used to obtain the selection probability of each sample. The sample pair with the largest address difference is selected first as the initial center point set. Each sample is assigned to the initial center point closest to it, resulting in multiple clusters. The center of the cluster is updated by calculating the mean of all samples in the cluster. The samples are then reassigned to the new cluster center, and the distance the center point moves is calculated each time. Set a movement threshold. If the movement distance of the center point is less than the movement threshold, stop clustering and obtain the clustering result.

3. The modular tunnel construction control method based on digital twins according to claim 2, characterized in that, The specific steps for obtaining the initial center point are as follows: A1: Construct a geological category vector based on the geological condition classification field, calculate the Euclidean distance for each pair of samples, and obtain the geological dissimilarity matrix; A2: Randomly select the first center point, calculate the minimum squared distance from each sample to the center point, and generate a selection probability table; A3: Randomly select secondary center points according to the selected probability distribution table, calculate the geological difference between the first center point and the secondary center points, and use them as the candidate center point set to verify whether the geological difference of the candidate center point set meets the basic threshold. If the conditions are met, the candidate centroid set is determined as the initial centroid set. If the conditions are not met, steps A2 to A3 are repeated until a sample pair that meets the conditions is found.

4. The modular tunnel construction control method based on digital twins according to claim 1, characterized in that, The specific steps for mining association rules include: The continuous eigenvalues ​​in the feature matrix are transformed into discrete levels, and key-value pairs are generated for each sample to generate a transaction dataset. The Apriori algorithm is used to scan the transaction dataset, generate frequent itemsets with support greater than or equal to N, extract rules with confidence greater than or equal to M from the frequent itemsets, and verify the validity of the rules to obtain the association rule set.

5. The modular tunnel construction control method based on digital twins according to claim 1, characterized in that, The specific steps for simulating and predicting the optimal parameter combination using finite element analysis and discrete element method include: Finite element meshes are generated based on digital twin models, with ground stress and construction loads used as boundary conditions, displacement constraints are set, and the output is a finite element file. Nonlinear finite element analysis based on finite element files is used to simulate the plastic deformation of rock mass, evolve the stress field throughout the construction cycle, calculate key indicators, and obtain stress cloud diagrams and deformation thermograms. The rock mass is simultaneously discretized into multiple particles, and contact parameters between the particles are set. Based on the contact parameters, the cyclic load of the contact force is counted by the rain flow counting method, and the overall deformation of the support structure is calculated by using the displacement field of the particles, so as to obtain the displacement curve of the support structure and the distribution map of the rock mass fragmentation degree. The maximum displacement and stress values ​​of the stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are compared with the threshold values ​​of the construction specifications to obtain a safety assessment report. The stress cloud map, deformation heat map, support structure displacement curve, and rock mass fracture distribution map are then superimposed onto the digital twin model to synthesize a simulation video.

6. The modular tunnel construction control method based on digital twins according to claim 1, characterized in that, The optimal parameter combination is iteratively optimized, and the specific steps include: A genetic algorithm is used to iteratively optimize the optimal parameter combination to obtain an optimized parameter combination. Based on the optimized parameter combination, simulation and prediction are performed again to determine whether the requirements are met. The optimized parameter combination after successful verification is simultaneously fed back to the structured dataset.

7. The modular tunnel construction control method based on digital twins according to claim 6, characterized in that, The specific steps of using a genetic algorithm to iteratively optimize the optimal parameter combination include: B1: Randomly generate multiple sets of parameter combinations as the initial population matrix, randomly select two initial population matrices to determine the intersection point position, and generate the second-generation population matrix. B2: For each second-generation population matrix, perform Gaussian mutation on each parameter with probability 'a' to obtain the mutated population matrix; B3: Based on the safety assessment report and construction efficiency design objective function, obtain the objective function value of each set of parameters, select the excellent parameter combination based on the objective function value, and use it as the selection population matrix; B4: Count the number of iterations. If the number of iterations reaches H, stop the iteration. Otherwise, repeat B1 to B3.

8. A modular tunnel construction control system based on digital twins, characterized in that, It is implemented based on the modular tunnel construction management and control method based on digital twins as described in any one of claims 1 to 7, wherein the system comprises: The model building module acquires three-dimensional spatial data, construction parameter data, and business logic data of the tunnel construction area to build physical and professional models respectively, defines semantic mapping rules between the physical and professional models, establishes dynamic mapping relationships between the physical and professional models, and obtains a digital twin model. The construction parameter module collects historical construction data, preprocesses the historical construction data to obtain a structured dataset, extracts the feature matrix from the structured dataset, applies an improved clustering algorithm to obtain clustering results and mine association rules, generates an association rule set, and transforms the association rule set into the optimal parameter combination. The parameter optimization module, relying on the digital twin model, uses finite element analysis and discrete element method to simulate and predict the optimal parameter combination, and iteratively optimizes the optimal parameter combination. The process management module uses a digital twin model to collect data streams in real time, combines AI to identify the data streams, obtain process identification results, judge the process identification results, and issue process conversion instructions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the modular tunnel construction control method based on digital twins as described in claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed, implements the modular tunnel construction control method based on digital twins as described in claims 1-7.

Images