Cooperative monitoring method and system for whole process of shield tunnel construction based on digital twinning

Abstract

The invention discloses a shield tunnel construction whole process cooperative monitoring method and system based on digital twinning, and the method comprises the steps: firstly determining a monitoring element system, collecting multiple types of data, accessing the resources of each participant, carrying out the standardization processing to form a shared data resource pool, and constructing a total element digital twinning model based on the resource pool, synchronous mapping of a physical construction scene and a digital model is realized through dynamic updating of a linkage mechanism of data and the model; then, a multi-participant cooperative monitoring platform is built, space-time dimension coupling correlation analysis is carried out on multi-source data in the model by means of a platform fusion analysis module, dynamic causal correlation rules of structural response, environmental change and construction behaviors are extracted, and risk pre-judgment and early warning are completed in combination with a safety judgment standard and a real-time working condition correction threshold value; and finally, early warning information and disposal suggestions are pushed, a model simulation effect is utilized, feedback data are tracked to form closed-loop management, and risk controllability is ensured. And the intelligent level of construction safety management and control is improved.

Description

Technical Field

[0001] This invention belongs to the field of shield tunnel construction safety monitoring technology, and more specifically, relates to a collaborative monitoring method and system for the entire shield tunnel construction process based on digital twins. Background Technology

[0002] During shield tunnel construction, the engineering geological conditions are complex and variable, and the surrounding environment is densely populated with sensitive points. The entire construction process involves multi-dimensional dynamic changes in structural stress, ground deformation, and equipment operation, which places stringent demands on the comprehensiveness, real-time nature, and coordination of construction monitoring. Currently, shield tunnel construction monitoring mostly adopts a decentralized data acquisition mode. The data storage formats and acquisition standards of various participants are not uniform, forming data silos. This makes it difficult to effectively integrate multi-source data and provide comprehensive support for construction decision-making.

[0003] Traditional monitoring methods rely on single-dimensional data monitoring and fixed threshold judgments, lacking in-depth mining of the spatiotemporal correlation characteristics of data. This makes it difficult to accurately capture the dynamic causal relationship between construction behavior, environmental changes, and structural responses, easily leading to misjudgments of risks or delayed early warnings. Furthermore, existing monitoring systems lack a unified collaborative platform, resulting in poor information exchange among design, construction, and supervision stakeholders. This leads to delays in developing risk management plans, low collaboration efficiency, and difficulty in quickly responding to unexpected risks during construction.

[0004] Furthermore, traditional monitoring systems lack a digital mapping platform that precisely matches the physical construction scenario, making it impossible to intuitively present the dynamic evolution of the entire construction process. It also makes it difficult to simulate and predict the effectiveness of risk mitigation plans in advance, resulting in insufficiently targeted mitigation measures and potential resource waste or inadequate risk control. These problems not only affect construction efficiency but may also lead to safety accidents such as structural instability and damage to the surrounding environment, severely restricting the safety controllability and intelligent level of shield tunnel construction.

[0005] Therefore, building a monitoring method that can achieve efficient data integration, multi-party collaboration, accurate risk prediction, and closed-loop handling is of great practical significance for improving the safety management and control capabilities of the entire shield tunnel construction process, reducing construction risks, and ensuring project quality and the safety of the surrounding environment. It is also an urgent need for the current intelligent development of shield tunnel construction. Summary of the Invention

[0006] This invention aims to solve the problems of poor data integration, insufficient multi-entity collaboration, inaccurate risk prediction, and lack of closed-loop response in shield tunnel construction monitoring. By constructing a standardized data resource pool, a full-element digital twin model, and a collaborative monitoring platform, it deeply mines the spatiotemporal coupling correlation and dynamic causal rules of data, realizes real-time mapping, accurate early warning, and collaborative response throughout the construction process, improves the level of intelligent construction safety management, and ensures the safety of the project and the surrounding environment.

[0007] To address the aforementioned deficiencies or improvement needs of existing technologies, as a first aspect of this invention, the present invention provides a collaborative monitoring method for the entire process of shield tunnel construction based on digital twins, comprising: S1. Establish a monitoring element system for the entire shield tunnel construction process, collect various types of data corresponding to this system and access the data resources of all participating parties, and form a shared data resource pool through standardized processing; S2. Based on the shared data resource pool, construct a digital twin model of all elements of shield tunnel construction, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model; S3. Build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, perform spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extract dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combine preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. S4. Push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management until the risk is controllable.

[0008] Furthermore, the monitoring element system in S1 covers data related to engineering objects, construction processes, environmental conditions, and responses to external influences.

[0009] Furthermore, the standardization process in S1 is specifically as follows: Let the collected multi-source raw data set be... ,in Represents data type, The total number of data types covers dimensions such as engineering structure status, construction process, geological and hydrological environment, and response of surrounding sensitive objects; This represents the number of samples collected under the same data type. This represents the total number of samples for the corresponding data type. Define a data validity function ,in For the first The reasonable lower limit for this type of data is determined based on monitoring and control values ​​and the range of geological and hydrological parameters in the material. For the first Reasonable upper limit for class data; when The data is then determined to be valid. Data is deemed invalid and removed. The valid data is normalized to obtain standardized data. ,in For the first The effective sample set of class data For the first The minimum value of valid data in a class. For the first The maximum value of valid data for a class; Data consistency verification formula , To and Associated data types; setting The consistency threshold is determined based on the accuracy requirements of the monitoring data in the material; when When the data is determined to be conflict-free, it is retained; when... At the same time, by combining data collection timestamps with construction conditions and performing correlation analysis, data with a higher degree of matching with actual working conditions is retained; the final shared data resource pool is formed. .

[0010] Furthermore, the digital twin model in S2 includes entity elements and associated logic related to the entire construction process.

[0011] Furthermore, the collaborative monitoring platform in S3 supports data sharing and collaborative operations among multiple entities by configuring the collaborative permissions and interaction processes of each participating entity.

[0012] Furthermore, the spatiotemporal dimension coupling correlation analysis process in S3 is as follows: First, the temporal evolution characteristics and spatial distribution characteristics of multi-source data in the digital twin model are extracted. Let the time series set be... , Let be the total number of time nodes; let the set of spatial coordinates be . , This represents the total number of spatial monitoring points. A time-domain correlation model is constructed based on time series datasets. The sliding time window method is used to divide the data into segments, and the coherence of the rate of change of different types of data within the same time segment is calculated. : , in, For time time Standardized values ​​of class data For time time Standardized values ​​of class data The width of the sliding time window. for The rate of change of the data over time for The rate of change of the data over time; By combining spatial coordinate sets, a data spatial domain correlation model is established, and the gradient correlation degree between data at different spatial points is calculated. : , in, for Class data in spatial locations The gradient vector at a given point represents the spatial distribution trend of the data. for Class data in spatial locations The gradient vector at that point; Constructing a spatiotemporal coupled correlation model ,in Coherence of time-varying rate Correlation with spatial gradient The angle between the two sources is used to determine the coupling relationship of multi-source data by fusing spatiotemporal correlation features through trigonometric functions. When it approaches 1, it indicates Class data and Data of this type exhibits strong coupling and correlation in the spatiotemporal dimensions.

[0013] Furthermore, the extraction process of the dynamic causal association rules in S3 is as follows: Coupling correlation values ​​obtained based on spatiotemporal dimension coupling correlation analysis Construct a causal correlation matrix of multi-source data The matrix rows correspond to data types such as construction behavior and environmental change, while the columns correspond to data types such as structural response and response of surrounding objects. The matrix elements are the spatiotemporal coupling correlation values ​​between various types of data, intuitively presenting the distribution of the correlation strength between data. We introduce the Granger causality test model to determine the causal directionality of datasets with non-zero coupling correlation values ​​in the correlation matrix. For the latent dependent variable dataset, For the dataset of potential outcome variables, construct a regression equation: , in for Time-of-fact variable data, For the result variable lagged Data from the period, For the dependent variable lagged Data from the period, For constant terms, , For regression coefficients, For random error term, The lag order indicates the order before use. Regression was performed on the data from the previous period; The significance of the lagged coefficients of the dependent variable in the regression equation is verified using the F-test. If the test result rejects the F-test, the significance is negligible. If the null hypothesis is true, then the judgment is... for The Granger causality was identified, clarifying the causal relationship between the data; and the spatiotemporal coupling correlation values ​​were combined. The magnitude of the causal relationship is used to quantify the strength of the causal relationship and form the basic relationship unit of "dependent variable type - causal direction - relationship strength - effect variable type"; Based on dynamic data updates throughout the construction process, time-series iterative optimization is performed on the foundation-related units, and the rate of change of causal correlation strength at different construction stages is calculated. ,in This represents the coupling correlation value at the current stage. This is the coupling correlation value from the previous stage. The applicable scenarios and strength thresholds of association rules are dynamically adjusted for each stage time interval. By integrating the above-mentioned causal directionality, correlation strength, and dynamic change characteristics, structured dynamic causal correlation rules are extracted and formed. The rules cover the influence logic of construction behavior, environmental changes on structural response, and response of surrounding objects, as well as the lag time and correlation strength quantification range. Moreover, the rules are dynamically updated with the construction conditions to ensure adaptability to complex construction scenarios.

[0014] Furthermore, the process of forming a closed-loop management system by tracking feedback data in S4 is as follows: Once the early warning and response plan is determined, based on the real-time mapping characteristics of the digital twin model, multi-source feedback data after the implementation of the response measures are collected simultaneously. All feedback data are processed according to standardized logic to achieve format unification and validity verification, forming a set of response feedback data. Construct a quantitative model of treatment effect ,in This is the original set of abnormal data when the warning is triggered. This is a set of feedback data after the intervention; the model quantifies the risk mitigation effect of the intervention measures by calculating the relative changes in data before and after the intervention. When the value is negative and its absolute value increases over time, it indicates that the measures taken have effectively reduced the abnormal data value, and the risk shows a convergence trend; when When the value is positive or the absolute value remains unchanged, it indicates that the measures taken have not achieved the expected results. Based on the extracted dynamic causal association rules, analyze the core reasons why the treatment effect did not meet expectations; if the treatment plan did not cover key causal factors, supplement and adjust measures based on the association rules; if the adaptability of the original plan decreased due to dynamic changes in construction conditions, use a digital twin model to re-simulate and optimize the plan, generate targeted adjustment suggestions and push them to the collaborative monitoring platform. Establish an iterative response mechanism, re-implement the response according to the optimized plan, and repeat the data collection and effect quantitative analysis process until... The data must meet the normal data range corresponding to the preset safety assessment criteria.

[0015] As a second aspect of the present invention, a collaborative monitoring system for the entire process of shield tunnel construction based on digital twins is also provided, comprising: The shared data resource pool construction unit is used to determine the monitoring element system for the entire shield tunnel construction process, collect the various types of data corresponding to this system and access the data resources of each participant, and form a shared data resource pool through standardized processing; The digital twin model construction and mapping unit is used to construct a digital twin model of all elements of shield tunnel construction based on the shared data resource pool, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model. The spatiotemporal coupling analysis unit is used to build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, it performs spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extracts dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combines preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. The collaborative handling closed-loop management unit is used to push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management, until the risk is controllable.

[0016] As a third aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, which is executed by a processor, according to any one of the claims, a collaborative monitoring method for the entire process of shield tunnel construction based on digital twins.

[0017] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: 1. The present invention provides a collaborative monitoring method for the entire shield tunnel construction process based on digital twins. This method establishes a monitoring element system for the entire shield tunnel construction process, collects multiple types of data, and integrates data resources from various participants. After standardization, a shared data resource pool is formed. Data standardization uses a validity judgment function to filter valid data, normalization to unify data format and accuracy, and consistency verification formulas to eliminate data conflicts, ensuring data integrity, accuracy, and relevance. This technical feature breaks down data barriers among multiple participants in traditional construction, solves the problems of data dispersion and inconsistent formats, provides a high-quality data foundation for subsequent digital twin modeling and collaborative analysis, ensures the reliability of analysis results in each subsequent stage, and provides solid data support for the entire monitoring process.

[0018] 2. The present invention provides a collaborative monitoring method for the entire shield tunnel construction process based on digital twins. This method constructs a full-element digital twin model based on a shared data resource pool, utilizes a data-model linkage mechanism to achieve synchronous mapping between the physical construction scene and the digital model, and then conducts spatiotemporal coupling correlation analysis through the fusion analysis module of the collaborative monitoring platform. The time-domain correlation model quantifies the synergy of data change rates, the spatial domain correlation model calculates the gradient correlation, and the spatiotemporal coupling model integrates the features of both to achieve threshold-free quantification, thereby extracting dynamic causal correlation rules. This technical feature constructs a precise digital mirror of the physical scene, uncovers deep spatiotemporal coupling relationships and causal logic between data, breaks through the limitations of traditional single data threshold judgment, improves the accuracy and foresight of risk prediction, and provides a scientific logical basis for risk identification.

[0019] 3. The present invention provides a collaborative monitoring method for the entire shield tunnel construction process based on digital twins. This method pushes early warning information and handling suggestions through a collaborative monitoring platform, simulates the effects of solutions using a digital twin model, tracks feedback data and constructs a quantitative model of handling effects, analyzes the reasons for unsatisfactory results using dynamic causal association rules, and establishes an iterative handling mechanism to form a closed-loop management system. This technical feature achieves collaborative, visualized, and closed-loop risk handling, solving the problems of inefficient multi-entity collaboration and difficulty in predicting solution effects in traditional handling methods. It ensures the pertinence and effectiveness of handling measures, while updating the closed-loop handling archive to an optimized knowledge base, providing a reference for similar projects, promoting the development of shield tunnel construction monitoring towards intelligence and efficiency, and ensuring the safety and controllability of the entire construction process. Attached Figure Description

[0020] Figure 1 This is a flowchart of the collaborative monitoring method for the entire shield tunnel construction process based on digital twins, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of underwater topographic measurement according to an embodiment of the present invention; Figure 3This is a schematic diagram of the layout of groundwater level measuring points according to an embodiment of the present invention; Figure 4 This is a system unit diagram of an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0022] Example 1 Please refer to Figure 1 This embodiment 1 provides a collaborative monitoring method for the entire process of shield tunnel construction based on digital twins, including: S1. Establish a monitoring element system for the entire shield tunnel construction process, collect various types of data corresponding to this system and access the data resources of all participating parties, and form a shared data resource pool through standardized processing; S2. Based on the shared data resource pool, construct a digital twin model of all elements of shield tunnel construction, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model; S3. Build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, perform spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extract dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combine preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. S4. Push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management until the risk is controllable.

[0023] This embodiment 1 further elaborates on the above steps.

[0024] (1) Construction of shared data resource pool In shield tunnel construction, due to the involvement of multiple parties and the decentralized management of multi-dimensional data, problems such as inconsistent data formats and information barriers often occur in the traditional model. Therefore, it is necessary to first clarify the scope of elements that need to be monitored throughout the construction process, and then standardize and integrate the data.

[0025] First, a monitoring element system covering the project object, construction process, environmental conditions, and external impact response must be established. This system includes data related to the structural status of the project and the construction process, as well as geological and hydrological environment and responses from surrounding sensitive objects. Subsequently, various types of data corresponding to this system are collected and integrated with existing data resources from all participating parties, including construction, supervision, and design.

[0026] like Figure 2 The process of the equipment detecting underwater topography and generating a visualized topographic model corresponds to the data collection stage of the "environmental conditions" category in the monitoring element system. This topographic data belongs to the core information of the geological environment dimension and will be included in the category of multi-source data to be collected. The visualized presentation of the topographic model in the figure also reflects the preliminary structured processing results after data collection, providing a foundation for subsequent standardized integration. Figure 3 The groundwater level monitoring points distributed in the figure are specific monitoring carriers of the "environmental conditions" dimension in the monitoring element system. The corresponding groundwater level data is one of the key environmental data that needs to be collected. At the same time, the tunnels, existing ground and other elements in the figure also correspond to the monitoring content of the "engineering objects" category. The data associated with these elements will be synchronously connected to the data resources to jointly support the construction of the subsequent shared data resource pool.

[0027] Next, these multi-source raw data will be standardized: Let the collected multi-source raw data set be... ,in Represents data type, The total number of data types covers dimensions such as engineering structure status, construction process, geological and hydrological environment, and response of surrounding sensitive objects; This represents the number of samples collected under the same data type. This represents the total number of samples for the corresponding data type. Define a data validity function ,in For the first The reasonable lower limit for this type of data is determined based on monitoring and control values ​​and the range of geological and hydrological parameters in the material. For the first Reasonable upper limit for class data; when The data is then determined to be valid. Data is deemed invalid and removed. The valid data is normalized to obtain standardized data. ,in For the first The effective sample set of class data For the first The minimum value of valid data in a class. For the first The maximum value of valid data for a class; Data consistency verification formula , To and Related data types, such as construction tunneling parameters and segment deformation data; settings The consistency threshold is determined based on the accuracy requirements of the monitoring data in the material; when When the data is determined to be conflict-free, it is retained; when... At the same time, by combining data collection timestamps with construction conditions and performing correlation analysis, data with a higher degree of matching with actual working conditions is retained; the final shared data resource pool is formed. All data meet the data accuracy requirements for digital twin modeling and the logical correlation of multi-source data coupling analysis.

[0028] After this series of processes, the final shared data resource pool has all data in the same format and with consistent precision, and the logical relationships between the data meet the needs of subsequent digital twin modeling and multi-source data coupling analysis.

[0029] (2) Digital twin model construction mapping In shield tunnel construction, traditional monitoring lacks a visual and dynamic carrier that corresponds to the actual scene, making it difficult to intuitively present the changes throughout the construction process. Therefore, it is necessary to rely on the integrated shared data resource pool to build a digital model that can map the real construction scene.

[0030] First, based on a shared data resource pool, a digital twin model of all elements of shield tunnel construction is built. This model covers all physical elements involved in the entire construction process, including tunnel structure, shield equipment, geological strata, surrounding sensitive buildings and pipelines, etc. It also includes the relationship logic between these physical elements, such as the installation relationship between shield excavation and segment lining, the impact logic of stratum deformation on surrounding buildings, etc., to ensure that the model can completely reproduce the various relationships in actual construction.

[0031] A data-model linkage mechanism was then established: a real-time data transmission link was built between the shared data resource pool and the digital twin model, and a corresponding "data reception-status update" module was configured for each entity element in the model. When new real-time collected data (such as shield cutterhead torque, segment circumferential stress, and stratum pore water pressure) is added to the shared data resource pool, the data is synchronously pushed to the corresponding entity element module through the link. The module matches the entity's attribute fields according to the data type (e.g., matching the cutterhead torque data to the "cutterhead operating parameters" field of the shield equipment model), and then automatically adjusts the status parameters of the corresponding elements in the model according to the physical characteristics and correlation logic of the entity elements (e.g., when the cutterhead torque increases, the "tunneling resistance" attribute of the shield equipment in the model is synchronously increased). At the same time, the model refreshes the status of each element sequentially according to the construction timeline based on the data update timestamp, ensuring that the evolution rhythm of the model is consistent with the progress of the physical construction scenario. Ultimately, the synchronous mapping between the physical construction scenario and the digital twin model is achieved, allowing the model to accurately reflect the current status and evolution trend of the actual construction, providing a realistic digital carrier for subsequent data analysis and risk prediction.

[0032] (3) Spatiotemporal coupling analysis In traditional shield tunnel construction monitoring, the lack of a unified interaction platform often leads to inefficient collaboration among multiple stakeholders, including design, construction, and supervision. Furthermore, single-data monitoring struggles to uncover deep correlations, making risk prediction prone to errors. Therefore, it is necessary to build a multi-stakeholder collaborative monitoring platform. This involves configuring the collaborative permissions and interaction processes for each participant, clarifying the data access, operation, and feedback permissions for different entities, standardizing information transmission processes, and ensuring smooth data sharing and collaborative operations among all parties.

[0033] The platform integrates a fusion analysis module, which is used to conduct spatiotemporal coupling correlation analysis on multi-source data in the digital twin model. Specifically, firstly, it extracts the temporal evolution characteristics and spatial distribution characteristics of the multi-source data in the digital twin model, assuming the time series set is... , Let be the total number of time nodes, where each time node corresponds to multiple types of data collected at the same moment, including the engineering structure status, construction activities, and environmental changes; let the spatial coordinate set be . , This refers to the total number of spatial monitoring points, covering key locations such as monitoring sections along the tunnel and monitoring points for surrounding sensitive objects. A time-domain correlation model is constructed based on time series datasets. The sliding time window method is used to divide the data into segments, and the coherence of the rate of change of different types of data within the same time segment is calculated. : , in, For time time Standardized values ​​of class data For time time Standardized values ​​of class data The width of the sliding time window. for The rate of change of the data over time for The time change rate of the data; this degree of coordination directly reflects the time synchronization and lag correlation logic between construction behavior and structural response and environmental changes, and can quantify the coupling strength of the time dimension without the need for preset judgment criteria; By combining spatial coordinate sets, a data spatial domain correlation model is established, and the gradient correlation degree between data at different spatial points is calculated. : , in, for Class data in spatial locations The gradient vector at a given point represents the spatial distribution trend of the data. for Class data in spatial locations The gradient vector at a given location; the ratio of the dot product of the gradient vectors to the product of their magnitudes forms a spatial correlation metric, which intuitively reflects the spatial synergy and propagation patterns of data changes in different regions; Constructing a spatiotemporal coupled correlation model ,in Coherence of time-varying rate Correlation with spatial gradient The angle between the two sources is used to determine the coupling relationship of multi-source data by fusing spatiotemporal correlation features through trigonometric functions. When it approaches 1, it indicates Class data and The data exhibits strong coupling relationships in the spatiotemporal dimensions. Based on this, dynamic causal relationship rules between structural response, environmental changes, and construction behavior during the construction process can be extracted, providing accurate logical support for subsequent risk prediction.

[0034] Furthermore, the coupling correlation values ​​obtained based on spatiotemporal dimension coupling correlation analysis Construct a causal correlation matrix of multi-source data The matrix rows correspond to data types such as construction behavior and environmental change, while the columns correspond to data types such as structural response and response of surrounding objects. The matrix elements are the spatiotemporal coupling correlation values ​​between various types of data, intuitively presenting the distribution of the correlation strength between data. We introduce the Granger causality test model to determine the causal directionality of datasets with non-zero coupling correlation values ​​in the correlation matrix. For potential dependent variable datasets (such as construction tunneling parameters, groundwater level change data). For datasets of potential outcome variables (such as segment deformation and building settlement data), construct regression equations: , in for Time-of-fact variable data, For the result variable lagged Data from the period, For the dependent variable lagged Data from the period, For constant terms, , For regression coefficients, For random error term, The lag order indicates the order before use. Regression was performed on the data from the previous period; The significance of the lagged coefficients of the dependent variable in the regression equation is verified using the F-test. If the test result rejects the F-test, the significance is negligible. If the null hypothesis is true, then the judgment is... for The Granger causality was identified, clarifying the causal relationship between the data; and the spatiotemporal coupling correlation values ​​were combined. The magnitude of the causal relationship is used to quantify the strength of the causal relationship and form the basic relationship unit of "dependent variable type - causal direction - relationship strength - effect variable type"; Based on dynamic data updates throughout the construction process, time-series iterative optimization is performed on the foundation-related units, and the rate of change of causal correlation strength at different construction stages is calculated. ,in This represents the coupling correlation value at the current stage. This is the coupling correlation value from the previous stage. The applicable scenarios and strength thresholds of association rules are dynamically adjusted for each stage time interval. By integrating the above-mentioned causal directionality, correlation strength, and dynamic change characteristics, structured dynamic causal correlation rules are extracted and formed. The rules cover the influence logic of construction behavior, environmental changes on structural response, and response of surrounding objects, as well as the lag time and correlation strength quantification range. Moreover, the rules are dynamically updated with the construction conditions to ensure adaptability to complex construction scenarios.

[0035] After extracting the dynamic causal relationship rules, the process moves to risk prediction and early warning. First, a preset safety assessment standard is introduced. This standard is based on industry standards, similar engineering cases, and geological survey reports for specific projects, covering the initial risk reference range for key indicators such as segment deformation, ground settlement, and the tilt of surrounding buildings. Then, the risk prediction threshold is adaptively adjusted based on real-time operating conditions. Combining the characteristics of the lag time and correlation strength between construction behavior and structural response in the dynamic causal relationship rules, the differences between the current construction parameters (such as tunnel boring speed and soil chamber pressure) and geological conditions (such as changes in surrounding rock grade) and the initial standard are analyzed. If tunneling in hard rock formations leads to an increase in cutterhead torque, the risk prediction threshold for segment stress is appropriately lowered based on the strong correlation between "increased torque and increased segment stress" characteristic in the relationship rules.

[0036] Then, risk level prediction is completed through association rules and revised thresholds. Real-time monitoring data is substituted into association rules to match corresponding causal relationships, determine whether the data exceeds the revised threshold range, and classify risk levels as minor, moderate, or severe. Finally, corresponding early warnings are triggered and pushed to the collaborative monitoring platform. The platform pushes early warning information in a targeted manner according to the permissions of each participating party, such as pushing operation adjustment suggestions to the construction party and verification requirements to the supervision party, ensuring that the early warnings are accurately adapted to the real-time construction scenario and providing a clear basis for multi-party collaborative handling.

[0037] (4) Collaborative closed-loop management After a risk warning is triggered during shield tunnel construction, relying solely on a single response measure is often insufficient to ensure that the risk is completely under control. Furthermore, traditional response methods lack a systematic evaluation and dynamic optimization mechanism for their effectiveness. Therefore, it is necessary to rely on a collaborative monitoring platform and a digital twin model to construct a closed-loop management process that extends from early warning push and scheme simulation to effect feedback.

[0038] Once an early warning is triggered, the collaborative monitoring platform will, based on the warning level and dynamic causal relationship rules, push the warning information and preliminary handling suggestions to relevant stakeholders such as construction, supervision, and design. This includes clearly identifying the risk location, type, and core influencing factors, along with reference handling plans adapted to the current working conditions, such as adjusting the tunnel boring machine's speed, optimizing soil chamber pressure parameters, or implementing ground reinforcement measures. To avoid the risks of blindly implementing handling measures, a digital twin model is first used to simulate the effects of the preliminary handling suggestions. The proposed handling parameters are input into the model, and based on the synchronous mapping characteristics between the model and the physical scene, the evolution trends of the engineering structure state and surrounding environment changes after the handling are deduced, predicting the feasibility and potential effects of the plan, and providing a basis for determining the final handling plan.

[0039] Once the early warning and response plan is determined, based on the real-time mapping characteristics of the digital twin model, multi-source feedback data after the implementation of the response measures are collected simultaneously, including engineering structure status feedback data, construction condition adjustment data, environmental change response data, and status feedback data of surrounding sensitive objects. All feedback data are processed according to standardized logic to complete format unification and validity verification, forming a set of response feedback data. Construct a quantitative model of treatment effect ,in This is the original abnormal data set (such as segment deformation and groundwater level anomalies) when the early warning is triggered. This is a set of feedback data after the intervention; the model quantifies the risk mitigation effect of the intervention measures by calculating the relative changes in data before and after the intervention. When the value is negative and its absolute value increases over time, it indicates that the measures taken have effectively reduced the abnormal data value, and the risk shows a convergence trend; when When the value is positive or the absolute value remains unchanged, it indicates that the measures taken have not achieved the expected results. Based on the extracted dynamic causal association rules, the core reasons why the treatment effect did not meet expectations are analyzed; if the treatment plan does not cover key causal factors, supplementary adjustment measures are made based on the association rules; if the adaptability of the original plan decreases due to dynamic changes in construction conditions, the plan is re-simulated and optimized using a digital twin model, and targeted adjustment suggestions are generated and pushed to the collaborative monitoring platform.

[0040] Establish an iterative response mechanism, re-implement the response according to the optimized plan, and repeat the data collection and effect quantitative analysis process until... The data must meet the normal data range corresponding to the preset safety assessment standards. During this process, each round of handling plan, feedback data, effect quantification results and adjustment logic are recorded to form a closed-loop handling file, which is then updated to the construction parameter optimization knowledge base to provide accurate reference for subsequent similar risk handling, thereby achieving closed-loop management and continuous optimization of risk handling.

[0041] This embodiment is particularly applicable to shield tunnel projects in urban core areas, crossing rivers, lakes, seas, and surrounding sensitive building complexes. In such projects, geological conditions are highly variable, environmental constraints are stringent, and the collaboration requirements of multiple stakeholders are high. This method, through multi-source data fusion, full-element digital twin mapping, and dynamic causal correlation analysis, can accurately capture the evolution patterns of construction risks, predict potential hazards such as segment deformation and ground settlement in advance, and quickly control risks by combining a closed-loop disposal mechanism. This can effectively reduce the incidence of safety accidents, ensure the safety of the main structure of the project and the surrounding environment, and provide reliable technical support for the construction of highly challenging shield tunnels.

[0042] At the level of intelligent development in the industry, the collaborative monitoring system and knowledge accumulation mechanism constructed in this embodiment can promote the transformation of shield tunnel construction from "experience-driven" to "data-driven". The resulting construction parameter optimization knowledge base can realize the reuse of risk management experience in similar projects, improving the overall risk management efficiency of the industry; the multi-participant collaborative platform breaks down information barriers in traditional management, promoting efficient linkage among design, construction, supervision and other entities. With the further popularization of digital twin and data mining technologies, this method can be extended to shield tunnel construction with different diameters and different geological conditions, and can also be linked with intelligent construction equipment to achieve automated handling, helping the intelligent upgrade of the shield tunnel construction field.

[0043] Example 2 Please refer to Figure 4 This embodiment 2 provides a collaborative monitoring system for the entire shield tunnel construction process based on digital twins, including: The shared data resource pool construction unit is used to determine the monitoring element system for the entire shield tunnel construction process, collect the various types of data corresponding to this system and access the data resources of each participant, and form a shared data resource pool through standardized processing; The digital twin model construction and mapping unit is used to construct a digital twin model of all elements of shield tunnel construction based on the shared data resource pool, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model. The spatiotemporal coupling analysis unit is used to build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, it performs spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extracts dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combines preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. The collaborative handling closed-loop management unit is used to push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management, until the risk is controllable.

[0044] Example 3 This embodiment 3 also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can realize any step of a collaborative monitoring method for the entire process of shield tunnel construction based on digital twins.

[0045] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0046] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0047] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A collaborative monitoring method for the entire process of shield tunnel construction based on digital twins, characterized in that, include: S1. Establish a monitoring element system for the entire shield tunnel construction process, collect various types of data corresponding to this system and access the data resources of all participating parties, and form a shared data resource pool through standardized processing; S2. Based on the shared data resource pool, construct a digital twin model of all elements of shield tunnel construction, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model; S3. Build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, perform spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extract dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combine preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. S4. Push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management until the risk is controllable.

2. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The monitoring element system in S1 covers data related to engineering objects, construction processes, environmental conditions, and responses to external influences.

3. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The standardization process in S1 is specifically as follows: Let the collected multi-source raw data set be... ,in Represents data type, The total number of data types covers dimensions such as engineering structure status, construction process, geological and hydrological environment, and response of surrounding sensitive objects; This represents the number of samples collected under the same data type. This represents the total number of samples for the corresponding data type. Define a data validity function ,in For the first The reasonable lower limit for this type of data is determined based on monitoring and control values ​​and the range of geological and hydrological parameters in the material. For the first Reasonable upper limit for class data; when The data is then determined to be valid. Data is deemed invalid and removed. The valid data is normalized to obtain standardized data. ,in For the first The effective sample set of class data For the first The minimum value of valid data in a class. For the first The maximum value of valid data for a class; Data consistency verification formula , To and Associated data types; setting The consistency threshold is determined based on the accuracy requirements of the monitoring data in the material; when When the data is determined to be conflict-free, it is retained. when At the same time, by combining data collection timestamps with construction conditions and performing correlation analysis, data with a higher degree of matching with actual working conditions is retained; the final shared data resource pool is formed. .

4. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The digital twin model in S2 includes entity elements and associated logic related to the entire construction process.

5. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The collaborative monitoring platform in S3 supports data sharing and collaborative operations among multiple entities by configuring the collaborative permissions and interaction processes of each participating entity.

6. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The process of spatiotemporal dimension coupling correlation analysis in S3 is as follows: First, the temporal evolution characteristics and spatial distribution characteristics of multi-source data in the digital twin model are extracted. Let the time series set be... , Let be the total number of time nodes; let the set of spatial coordinates be . , This represents the total number of spatial monitoring points. A time-domain correlation model is constructed based on time series datasets. The sliding time window method is used to divide the data into segments, and the coherence of the rate of change of different types of data within the same time segment is calculated. : ， in, For time time Standardized values ​​of class data For time time Standardized values ​​of class data The width of the sliding time window. for The rate of change of the data over time for The rate of change of the data over time; By combining spatial coordinate sets, a data spatial domain correlation model is established, and the gradient correlation degree between data at different spatial points is calculated. : ， in, for Class data in spatial locations The gradient vector at a given point represents the spatial distribution trend of the data. for Class data in spatial locations The gradient vector at that point; Constructing a spatiotemporal coupled correlation model ,in Coherence of time-varying rate Correlation with spatial gradient The angle between the two sources is used to determine the coupling relationship of multi-source data by fusing spatiotemporal correlation features through trigonometric functions. When it approaches 1, it indicates Class data and Data of this type exhibits strong coupling and correlation in the spatiotemporal dimensions.

7. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The extraction process of the dynamic causal association rules in S3 is as follows: Coupling correlation values ​​obtained based on spatiotemporal dimension coupling correlation analysis Construct a causal correlation matrix of multi-source data The matrix rows correspond to data types such as construction behavior and environmental change, while the columns correspond to data types such as structural response and response of surrounding objects. The matrix elements are the spatiotemporal coupling correlation values ​​between various types of data, intuitively presenting the distribution of the correlation strength between data. We introduce the Granger causality test model to determine the causal directionality of datasets with non-zero coupling correlation values ​​in the correlation matrix. For the latent dependent variable dataset, For the dataset of potential outcome variables, construct a regression equation: ， in for Time-of-fact variable data, For the result variable lagged Data from the period, For the dependent variable lagged Data from the period, For constant terms, , For regression coefficients, For random error term, The lag order indicates the order before use. Regression was performed on the data from the previous period; The significance of the lagged coefficients of the dependent variable in the regression equation is verified using the F-test. If the test result rejects the hypothesis, the regression equation is considered significant. If the null hypothesis is true, then the judgment is... for The Granger causality was identified, clarifying the causal relationship between the data; and the spatiotemporal coupling correlation values ​​were combined. The magnitude of the causal relationship is used to quantify the strength of the causal relationship and form the basic relationship unit of "dependent variable type - causal direction - relationship strength - effect variable type"; Based on dynamic data updates throughout the construction process, time-series iterative optimization is performed on the foundation-related units, and the rate of change of causal correlation strength at different construction stages is calculated. ,in This represents the coupling correlation value at the current stage. This is the coupling correlation value from the previous stage. The applicable scenarios and strength thresholds of association rules are dynamically adjusted for each stage time interval. By integrating the above-mentioned causal directionality, correlation strength, and dynamic change characteristics, structured dynamic causal correlation rules are extracted and formed. The rules cover the influence logic of construction behavior, environmental changes on structural response, and response of surrounding objects, as well as the lag time and correlation strength quantification range. Moreover, the rules are dynamically updated with the construction conditions to ensure adaptability to complex construction scenarios.

8. The collaborative monitoring method for the entire shield tunnel construction process based on digital twins according to claim 1, characterized in that, The process of forming a closed-loop management system from the tracking and feedback data in S4 is as follows: Once the early warning and response plan is determined, based on the real-time mapping characteristics of the digital twin model, multi-source feedback data after the implementation of the response measures are collected simultaneously. All feedback data are processed according to standardized logic to achieve format unification and validity verification, forming a set of response feedback data. Construct a quantitative model of treatment effect ,in This is the original set of abnormal data when the warning is triggered. This is a set of feedback data after the intervention; the model quantifies the risk mitigation effect of the intervention measures by calculating the relative changes in data before and after the intervention. When the value is negative and its absolute value increases over time, it indicates that the measures taken have effectively reduced the abnormal data value, and the risk shows a convergence trend; when When the value is positive or the absolute value remains unchanged, it indicates that the measures taken have not achieved the expected results. Based on the extracted dynamic causal relationship rules, we analyze the core reasons why the treatment effect did not meet expectations; If the disposal plan does not cover key causal factors, supplementary adjustment measures shall be taken based on the association rules. If the adaptability of the original plan decreases due to dynamic changes in construction conditions, the plan is re-simulated and optimized using a digital twin model, and targeted adjustment suggestions are generated and pushed to the collaborative monitoring platform. Establish an iterative response mechanism, re-implement the response according to the optimized plan, and repeat the data collection and effect quantitative analysis process until... The data must meet the normal data range corresponding to the preset safety assessment criteria.

9. A collaborative monitoring system for the entire process of shield tunnel construction based on digital twins, characterized in that, include: The shared data resource pool construction unit is used to determine the monitoring element system for the entire shield tunnel construction process, collect the various types of data corresponding to this system and access the data resources of each participant, and form a shared data resource pool through standardized processing; The digital twin model construction and mapping unit is used to construct a digital twin model of all elements of shield tunnel construction based on the shared data resource pool, and update the model through the linkage mechanism between data and model to complete the synchronous mapping between the physical construction scene and the digital twin model. The spatiotemporal coupling analysis unit is used to build a multi-participant collaborative monitoring platform. Through the integrated fusion analysis module of the platform, it performs spatiotemporal dimension coupling correlation analysis on multi-source data in the digital twin model, extracts dynamic causal correlation rules of structural response, environmental changes and construction behavior during construction, and combines preset safety assessment standards and real-time working conditions to adaptively correct risk prediction thresholds, complete risk level prediction and trigger corresponding early warnings. The collaborative handling closed-loop management unit is used to push early warning information and handling suggestions through the collaborative monitoring platform, simulate the effect of the solution using a digital twin model, track feedback data to form closed-loop management, until the risk is controllable.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor as described in any one of claims 1-8: a collaborative monitoring method for the entire process of shield tunnel construction based on digital twins.

Images