A bim-based underground excavation engineering construction risk dynamic early warning method and system

Abstract

The application relates to the technical field of underground excavation engineering construction safety, in particular to a dynamic early warning method and system for underground excavation engineering construction risks based on BIM, which comprises the following steps: constructing a BIM risk integration model of underground excavation engineering according to engineering data of the underground excavation engineering and historical construction data of the same type of underground excavation engineering; binding a unique identifier to the collected risk monitoring data of the same group, preprocessing the risk monitoring data of the same identifier, and obtaining comprehensive risk indexes of corresponding model objects in the BIM risk integration model; inputting the risk comprehensive indexes of the model objects into the BIM risk integration model, analyzing the risk level and risk propagation influence of the model objects, generating a risk assessment report and triggering a hierarchical early warning; receiving feedback information of early warning closed-loop disposal executed according to the early warning information, and updating the BIM risk integration model according to the feedback information. The application can realize multi-dimensional risk identification, dynamic evaluation, accurate early warning and closed-loop disposal.

Description

Technical Field

[0001] This application relates to the field of construction safety technology for underground tunneling projects, and in particular to a BIM-based dynamic early warning method and system for construction risks in underground tunneling projects. Background Technology

[0002] With the rapid development of urban underground space development and utilization, the scale and complexity of underground excavation projects (such as subway tunnels, underground utility tunnels, and underground stations) are increasing daily. Because underground excavation projects are typically located in complex hydrogeological environments, they are prone to safety risks during construction, such as surface subsidence, tunnel collapse, sudden water inrush, and deformation of surrounding buildings. Inadequate risk control can not only delay construction and increase costs but also potentially lead to major safety accidents, causing casualties and property damage. Therefore, accurate and timely dynamic early warning of risks during the construction of underground excavation projects is crucial to ensuring their safe implementation.

[0003] Currently, risk early warning for underground excavation projects mainly relies on a combination of traditional manual monitoring and experience-based judgment. While some projects have introduced simple monitoring equipment and early warning systems, several technical shortcomings remain: First, risk identification is incomplete; traditional methods struggle to cover multi-dimensional risk factors such as geology, construction, environment, and structure, easily leading to risk omissions. Second, early warnings are often delayed, mostly reactive, making it difficult to predict and dynamically manage risks in advance, and hindering timely containment of accidents. Third, visualization is low; the location, level, and scope of impact of risks are difficult to present intuitively, hindering managers from quickly grasping the risk situation. Fourth, data utilization is low; monitoring data, construction information, and geological data are independent, preventing data fusion and analysis, resulting in insufficient accuracy in risk assessment. Fifth, early warning and response are disconnected; a robust response mechanism is lacking, making it difficult to quickly implement response measures and form a closed-loop management system after early warning information is released.

[0004] Therefore, how to achieve multi-dimensional risk identification, dynamic assessment, accurate early warning, and closed-loop handling in underground excavation projects is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This application provides a BIM-based method and system for dynamic early warning of construction risks in underground tunneling projects, so as to realize multi-dimensional risk identification, dynamic assessment, accurate early warning and closed-loop handling of underground tunneling projects.

[0006] To solve the above-mentioned technical problems, this application provides the following technical solution: A BIM-based dynamic early warning method for construction risks in underground tunneling projects includes the following steps: Step S110: Constructing an integrated BIM risk model for the underground tunneling project based on engineering data and historical construction data of similar underground tunneling projects; Step S120: Binding unique identifiers to risk monitoring data of the same group collected, preprocessing risk monitoring data with the same identifier, and obtaining comprehensive risk indicators for the corresponding model objects in the integrated BIM risk model; Step S130: Inputting the comprehensive risk indicators of the model objects into the integrated BIM risk model, analyzing the risk level and risk propagation impact of the model objects, generating a risk assessment report, and triggering graded early warnings; Step S140: Receiving feedback information on the early warning information and executing closed-loop early warning measures, and updating the integrated BIM risk model based on the feedback information.

[0007] The BIM-based dynamic early warning method for construction risks in underground tunneling projects, as described below, preferably includes the following sub-steps in step S110: Step S111: Obtain the construction design drawings, geological survey report, monitoring plan, construction plan, and historical construction data of similar underground tunneling projects; Step S112: Construct a BIM basic model of the underground tunneling project based on the construction design drawings; Step S113: Import the geological survey report and monitoring plan into the BIM basic model to generate a geological BIM sub-model and a monitoring BIM sub-model; Step S114: Perform spatial matching and temporal correlation between the construction plan and the BIM basic model, geological BIM sub-model, and monitoring BIM sub-model to generate an integrated BIM risk model; Step S115: Add risk factors to the model objects of the integrated BIM risk model based on the geological survey report, construction plan, and historical construction data of similar underground tunneling projects, thereby improving the integrated BIM risk model.

[0008] The above-described BIM-based dynamic early warning method for construction risks in underground excavation projects preferably includes the following sub-steps in the preprocessing: detecting outliers in risk monitoring data with the same identifier and removing outliers; normalizing the risk monitoring data with the same identifier to the [0,1] interval after removing outliers; and fusing the normalized values ​​of the risk monitoring data with the same identifier to transform them into a comprehensive risk index for the corresponding model object.

[0009] The BIM-based dynamic early warning method for underground excavation construction risks, as described above, preferably includes the following sub-steps in step S130: Step S131: Binding the comprehensive risk index of the model object to the corresponding model object in the BIM integrated risk model; Step S132: Comparing the comprehensive risk index with the risk level threshold range of the bound model object, and visually labeling the high / medium risk level of the model object; Step S133: Calculating the risk propagation influence coefficient of the labeled model object's risk level on its associated model object, to determine the chain risk comprehensive index of the associated model object; Step S134: Determining the overall comprehensive risk index of the underground excavation project based on the comprehensive risk index of the labeled model object and the chain risk comprehensive index of its associated model object; Step S135: Triggering a graded early warning based on the overall comprehensive risk index of the underground excavation project, the comprehensive risk index of the labeled model object, and the chain risk comprehensive index of its associated model object.

[0010] The BIM-based dynamic early warning method for underground excavation construction risks, as described above, preferably includes the following triggering mechanisms for tiered early warnings: a yellow warning is triggered if the comprehensive risk index of any model object is not less than the medium-risk threshold, or the overall comprehensive risk index of the underground excavation project is not less than the yellow warning threshold; a red warning is triggered if the comprehensive risk index of any model object is not less than the high-risk threshold, or the overall comprehensive risk index of the underground excavation project is not less than the red warning threshold, or the cascading comprehensive risk index of any model object is not less than the high-risk threshold.

[0011] A BIM-based dynamic early warning system for construction risks in underground tunneling projects includes: a model building module, a binding and preprocessing module, a risk analysis and early warning module, and a risk closed-loop management module. The model building module constructs an integrated BIM risk model for the underground tunneling project based on engineering data and historical construction data from similar underground tunneling projects. The binding and preprocessing module assigns unique identifiers to risk monitoring data from the same group, preprocesses the risk monitoring data with the same identifier, and obtains the comprehensive risk index of the corresponding model object in the integrated BIM risk model. The risk analysis and early warning module inputs the comprehensive risk index of the model object into the integrated BIM risk model, analyzes the risk level and risk propagation impact of the model object, and triggers tiered early warnings. The risk closed-loop management module receives feedback information on the early warning information and executes closed-loop management based on the early warning information, and updates the integrated BIM risk model according to the feedback information.

[0012] The BIM-based dynamic early warning system for construction risks in underground tunneling projects, as described above, preferably includes a model construction module comprising: a data acquisition submodule, a basic model construction submodule, a sub-model construction submodule, an integrated submodule, and a model improvement submodule. The data acquisition submodule acquires construction design drawings, geological survey reports, monitoring plans, construction plans, and historical construction data from similar underground tunneling projects. The basic model construction submodule constructs a BIM basic model of the underground tunneling project based on the construction design drawings. The sub-model construction submodule imports the geological survey report and monitoring plan into the BIM basic model, generating a geological BIM sub-model and a monitoring BIM sub-model, respectively. The integrated submodule spatially matches and temporally correlates the construction plan with the BIM basic model, the geological BIM sub-model, and the monitoring BIM sub-model to generate a BIM risk integrated model. The model improvement submodule adds risk factors to the model objects of the BIM risk integrated model based on the geological survey report, construction plan, and historical construction data from similar underground tunneling projects, thereby improving the BIM risk integrated model.

[0013] In the BIM-based underground excavation engineering construction risk dynamic early warning system described above, preferably, the binding preprocessing module performs outlier detection on the risk monitoring data with the same identifier and removes the outliers. After removing the outliers, the risk monitoring data with the same identifier is normalized to the [0,1] interval. The normalized values ​​of the risk monitoring data with the same identifier are then merged to transform them into a comprehensive risk index for the corresponding model object.

[0014] The BIM-based dynamic early warning system for underground excavation engineering construction risks, as described above, preferably includes a risk analysis and early warning module comprising: an indicator object binding submodule, a comparison and annotation submodule, a chain risk analysis submodule, an overall risk assessment submodule, and a tiered early warning triggering submodule. The indicator object binding submodule binds the comprehensive risk index of a model object to the corresponding model object within the BIM integrated risk model. The comparison and annotation submodule compares the comprehensive risk index with the risk level threshold range of the bound model object, and visually annotates the high / medium risk levels of the model object. The chain risk analysis submodule calculates the risk propagation influence coefficient of the annotated model object's risk level on its associated model objects to determine the chain risk comprehensive index of the associated model objects. The overall risk assessment submodule determines the overall comprehensive risk index of the underground excavation project based on the comprehensive risk index of the annotated model object and the chain risk comprehensive index of its associated model objects. The tiered early warning triggering submodule triggers tiered early warnings based on the overall comprehensive risk index of the underground excavation project, the comprehensive risk index of the annotated model object, and the chain risk comprehensive index of its associated model objects.

[0015] In the BIM-based dynamic early warning system for underground excavation projects described above, preferably, a yellow warning is triggered if the comprehensive risk index of any model object is not less than the medium-risk threshold, or the overall comprehensive risk index of the underground excavation project is not less than the yellow warning threshold; a red warning is triggered if the comprehensive risk index of any model object is not less than the high-risk threshold, or the overall comprehensive risk index of the underground excavation project is not less than the red warning threshold, or the cascading comprehensive risk index of any model object is not less than the high-risk threshold.

[0016] Compared to the aforementioned background technologies, the BIM-based dynamic early warning method and system for underground excavation engineering construction risks provided in this application can cover multiple risk factors such as geology, construction, environment, and structure, thereby comprehensively identifying risks in underground excavation engineering in advance and promptly curbing the development of accidents. Furthermore, this application enables the visualization of risk warnings, intuitively presenting information such as risk level and location, facilitating managers to quickly grasp the risk situation. In addition, this application can utilize data from different sources related to underground excavation engineering, as well as historical data from similar underground excavation projects, resulting in high data utilization and effectively improving the accuracy of risk assessment. Moreover, this application simultaneously includes early warning and response stages, forming a closed-loop management system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a flowchart of the BIM-based dynamic early warning method for construction risks in underground tunneling projects as described in this application; Figure 2 This is a flowchart of the BIM risk integration model for underground excavation projects as described in this application; Figure 3 This is a flowchart illustrating the risk status and risk propagation impact of the analytical model objects in this application; Figure 4 This is a schematic diagram of the BIM-based dynamic early warning system for construction risks in underground tunneling projects as described in this application. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0020] Example 1 like Figure 1 As shown, this application provides a BIM-based dynamic early warning method for construction risks in underground tunneling projects, including the following steps: Step S110: Based on the engineering data of the underground tunneling project and the historical construction data of similar underground tunneling projects, construct a BIM risk integration model for the underground tunneling project. BIM (Building Information Modeling) is a data-driven, simulated, and manageable "digital twin" of a building. It is a three-dimensional visual model in which each model object carries corresponding information and can be used throughout the entire lifecycle of a project (design → construction → operation and maintenance → demolition).

[0021] Among them, such as Figure 2 As shown, step S110 includes the following sub-steps: Step S111: Obtain the construction design drawings, geological survey report, monitoring plan, construction plan, and historical construction data of similar underground excavation projects for the underground tunneling project; Before commencing underground excavation, it is essential to obtain engineering data for the project and historical construction data for similar projects. This engineering data includes: construction design drawings, geological survey reports, monitoring plans, and construction plans.

[0022] The construction design drawings include technical elements of the underground excavation project, such as: excavation cross-section (3.5m wide, 4.0m high), initial support (shotcrete + steel arch frame), secondary lining, construction shaft, monitoring points, etc. The geological survey report shows geological conditions and geological risk factors for the underground excavation project, such as: stratum distribution (silty clay, sand), soil and rock parameters (cohesion, internal friction angle), groundwater level (depth 5.2m), location of fault fracture zones, etc., and geological risk factors such as: surrounding rock collapse, water inrush, mudslide, and surrounding rock instability. The monitoring plan specifies the monitoring types and corresponding monitoring point layout requirements for underground excavation projects. For example, monitoring types include: tunnel crown settlement, surface settlement, anchor bolt axial force, and lining concrete stress. Monitoring point layout requirements are: crown settlement monitoring points must be placed along the tunnel crown centerline; surface settlement monitoring points must be placed on the surface above the tunnel; anchor bolt axial force monitoring points must be placed at the anchor bolt ends; and lining concrete stress monitoring points must be placed at key sections of the lining. The construction plan outlines the construction procedures and risk factors for underground excavation projects.

[0023] Step S112: Construct the BIM basic model of the underground excavation project based on the construction design drawings; Based on the technical elements of the underground excavation project contained in the construction design drawings, a BIM basic model of the underground excavation project is constructed using Revit software (a professional BIM modeling software). The BIM basic model establishes a complete three-dimensional coordinate system and engineering components within the three-dimensional coordinate system.

[0024] Step S113: Import the geological survey report and monitoring plan into the BIM basic model respectively to generate the geological BIM sub-model and the monitoring BIM sub-model; The geological survey report is imported into the BIM base model. Based on the geological conditions information of the underground excavation project shown in the geological survey report, stratigraphic units are generated in the BIM base model. The stratigraphic units are then matched and integrated with all engineering components in the same three-dimensional coordinate system, thereby expanding the BIM base model to form a geological BIM sub-model specifically for expressing the geological environment.

[0025] The monitoring plan is imported into the BIM base model. Based on the monitoring type information and corresponding monitoring point layout requirements of the underground excavation project specified in the monitoring plan, corresponding monitoring points are added to the BIM base model. The monitoring points are in the same three-dimensional coordinate system as all engineering components. The monitoring type (arch crown settlement, ground surface settlement, anchor bolt axial force, lining concrete stress), monitoring frequency (once every 2 hours), and monitoring threshold (arch crown settlement ≤ 3 mm / d, ground surface settlement ≤ 2 mm / d) of each monitoring point are specified. The monitoring type, monitoring frequency, and monitoring threshold of each monitoring point are recorded in the additional information of the corresponding monitoring point. Thus, a monitoring BIM sub-model specifically for expressing safety monitoring is formed on the basis of the BIM base model.

[0026] Step S114: Spatially match and temporally correlate the construction plan with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model to generate an integrated BIM risk model; Since the geological BIM sub-model, monitoring BIM sub-model, and BIM base model share the same three-dimensional coordinate system, and since the construction procedures for underground excavation projects specified in the construction plan are themselves a complete set of information containing spatial and temporal information, the construction procedures are spatially matched and temporally associated with engineering components in the BIM base model, stratigraphic units in the geological BIM sub-model, and monitoring points in the monitoring BIM sub-model according to the project mileage, construction location, and time sequence. This achieves the integrated integration of construction procedures with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model. For example, the spatial integration of construction procedures such as excavation, support, and grouting is combined with the temporal integration of the construction procedures, such as their sequence, construction duration, and construction steps, thereby generating an integrated BIM risk model.

[0027] Step S115: Add risk factors to the model objects of the BIM risk integration model based on the geological survey report, construction plan and historical construction data of similar underground excavation projects, thereby improving the BIM risk integration model. The geological survey report revealed geological risk factors, such as: surrounding rock collapse, water inrush, mudslide, and surrounding rock instability; the construction plan identified construction risk factors for the underground excavation project, such as: support failure, excessive settlement, excessive arch settlement, and face instability; historical construction data of similar underground excavation projects included historical risk factors such as collapses, settlement, leakage, and support damage that had occurred in previous similar underground excavation projects.

[0028] Geological risk factors, construction risk factors, and historical risk factors are added to the supplementary information of model objects such as engineering components, stratigraphic units, monitoring points, and construction procedures in the BIM integrated risk model, thereby improving the BIM integrated risk model. For example: adding the following to the supplementary information of engineering components (arches, shotcrete, anchors): support failure risk and deformation exceeding limit risk; adding the following to the supplementary information of stratigraphic units (faults, water-rich layers, weak surrounding rock): water inrush, collapse, and surrounding rock instability risk; adding the following to the supplementary information of monitoring points (arch crown subsidence, surface settlement): early warning threshold and exceeding standard risk; adding the following to construction procedures (excavation, support, grouting): procedure delay risk and construction disturbance risk; and binding historical risk factors such as historical risk cases, historical monitoring data, and accident cases of similar underground excavation projects to the corresponding model objects.

[0029] Based on this, all engineering components (e.g., excavation section outline components, excavation step components, shotcrete layers, steel arch structure components, steel mesh components), geological units (e.g., faults, water-rich layers, weak surrounding rock), monitoring points (e.g., arch crown settlement, surface settlement), and construction procedures (e.g., excavation, support, grouting) in the BIM basic model are uniquely coded. This unique code is used to make the model objects identifiable, measurable, traceable, collaborative, and traceable, thereby significantly improving the accuracy and efficiency of underground excavation construction management.

[0030] Step S120: Bind a unique identifier to the risk monitoring data of the same group collected, preprocess the risk monitoring data with the same identifier, and obtain the comprehensive risk index of the corresponding model object in the BIM risk integration model. Multi-source data acquisition equipment is deployed at the construction site of the underground excavation project to collect multi-source risk monitoring data. For example, eight surrounding rock displacement sensors are deployed in fault fracture zones and soft soil strata areas to monitor displacement within the enclosure and crown settlement; twelve mechanical sensors are deployed on the support structure to monitor anchor bolt axial force, steel arch frame internal force, and lining concrete stress; four environmental sensors are deployed in the construction area to monitor harmful gas concentration, groundwater level changes, temperature, and humidity; six construction parameter acquisition terminals are installed on the construction machinery to monitor excavation progress, support time, grouting parameters, etc.; and two manual data acquisition terminals are also provided to supplement and input data that cannot be automatically collected, such as blasting parameters.

[0031] Assign a unique identifier to risk monitoring data belonging to the same group (e.g., a rock displacement sensor, an anchor bolt axial force) to indicate that the risk monitoring value of this group corresponds to a certain model object in the BIM risk integration model. For example, the monitoring point (e.g., the left arch waist of the fault fracture zone K3+200 section), the construction process (e.g., the 3rd cycle of the excavation process), the corresponding engineering component (e.g., the 5th steel arch), and the corresponding stratum unit (e.g., the fault at the 20th meter). And perform the following preprocessing on risk monitoring data with the same identifier to obtain the comprehensive risk index of the corresponding model object.

[0032] First, outlier detection is performed on the risk monitoring data with the same identifier, and outliers are removed to avoid them affecting subsequent risk identification, thereby improving the accuracy of subsequent risk identification.

[0033] Furthermore, the formula for outlier detection is as follows: , in, For the first The first model object The first in the risk monitoring data Individual risk monitoring data; For the first The first model object The mean of risk monitoring data, , For the first The first model object The first in the risk monitoring data Individual risk monitoring data Quantity; For the first The first model object Standard deviation of risk monitoring data .

[0034] like Then the first The first model object The first type of risk monitoring data Individual risk monitoring data If it is an outlier, Then the first The first model object The first type of risk monitoring data Individual risk monitoring data This is a normal value.

[0035] Then, after removing outliers, the risk monitoring data with the same identifier are normalized to the [0,1] interval. This eliminates the dimensional differences in the risk monitoring data, avoids significant numerical disparities that could distort the BIM risk integration model, and ensures that all indicators participate fairly in subsequent risk identification. For example, surrounding rock displacement is measured in mm (e.g., 5, 12, 20); harmful gas concentration is measured in ppm (e.g., hundreds or thousands); and anchor bolt axial force is measured in kN (e.g., thousands or tens of thousands). If these data are not normalized, the anchor bolt axial force values ​​of several thousand will completely dominate the calculation results, while smaller values ​​such as surrounding rock displacement and harmful gas concentration will be overlooked. The BIM risk integration model will mistakenly believe that "anchor bolt axial force is the only risk."

[0036] Furthermore, the normalization formula is as follows: , in, For the first The first model object The first type of risk monitoring data Individual risk monitoring data The normalized value; For the first The first model object Minimum value of risk monitoring data; For the first The first model object The maximum value of risk monitoring data.

[0037] Finally, the normalized values ​​of the risk monitoring data with the same identifier are merged to transform them into a comprehensive risk index for the corresponding model objects that can be identified and calculated by the BIM integrated risk model. This simplifies subsequent risk identification and ensures the comprehensiveness of the data. The essence of this fusion is to synthesize indicators with different physical meanings, such as displacement, stress, water level, gas, and construction parameters, into a unified "comprehensive risk index," enabling the BIM integrated risk model to directly calculate, classify, and determine the risk level.

[0038] Furthermore, the fusion formula is as follows: , in, For the first A comprehensive risk index for each model object; For the first Risk weights for risk monitoring data; This refers to the number of types of risk monitoring data.

[0039] Step S130: Input the comprehensive risk index of the model object into the BIM integrated risk model, analyze the risk level and risk propagation impact of the model object, generate a risk assessment report and trigger a graded early warning; like Figure 3 As shown, step S130 includes the following sub-steps: Step S131: In the BIM risk integration model, bind the comprehensive risk index of the model object to the corresponding model object; Since each set of risk monitoring data that generates a comprehensive risk indicator is bound to a unique identifier, and this identifier corresponds to the corresponding model object in the BIM integrated risk model, after the comprehensive risk indicators of the model object are input into the BIM integrated risk model, the comprehensive risk indicators of the model object are bound to the corresponding model object in the BIM integrated risk model. This ensures that each comprehensive risk indicator can be mapped to a specific location in the BIM integrated risk model, providing spatial support for the targeting of risk identification and the accuracy of assessment.

[0040] Step S132: Compare the risk comprehensive index with the risk level threshold range of the model object it is bound to, and visually label the risk level of the model object with a high / medium risk level. The additional information of model objects in the BIM integrated risk model pre-records the risk level threshold range for each model object. Therefore, after binding the comprehensive risk index of the model object to the corresponding model object in step S131, the comprehensive risk index of each model object is compared with the risk level threshold range recorded in its additional information. The risk level threshold range is a pre-defined range based on engineering specifications, historical experience, and expert knowledge. For example: , in, For the first Risk level of each model object; The threshold for medium risk. This is a high-risk threshold, typically set to a value of [value missing]. , .

[0041] Based on the comparison results, model objects in the BIM risk integration model with high and medium risk levels are visually labeled to achieve a spatial representation of the risk level of model objects. For example, model objects are highlighted in different colors in the BIM risk integration model (e.g., yellow for medium risk, red for high risk), and risk labels pop up next to the model objects with the corresponding risk levels. Simultaneously, the comprehensive risk index and trigger time (i.e., the time when a medium or higher risk level is detected) are recorded in the model object's supplementary information, forming a risk event log for subsequent review of the labeled model object's data.

[0042] Step S133: Calculate the risk level of the labeled model object on the risk propagation influence coefficient of the associated model object, so as to determine the comprehensive index of the chain risk of the associated model object. By utilizing the spatial topological relationships between engineering components, geological units, and monitoring points in the BIM risk integration model, and the temporal logical relationships of construction procedures, the risk level of the labeled model objects is analyzed to determine the risk propagation impact on the associated model objects. The corresponding risk propagation impact coefficient is calculated, and the chain risk comprehensive index of the model objects associated with the labeled model objects is determined based on the risk propagation impact coefficient, thereby realizing the risk propagation impact analysis of the model objects.

[0043] Furthermore, the formula for the risk transmission impact coefficient is as follows: , in, For the first The model object and the first Risk propagation impact coefficient among model objects; For the first The risk source intensity coefficient of each model object is used as a risk source, and its value is... ; For the first The model object and the first The relationship indicator variable between model objects, if the first... The model object and the first If there are spatial topological relationships (e.g., adjacency, support, coverage) or temporal logical relationships (e.g., preceding and subsequent processes) between model objects, then ,otherwise ; For the first The model object up to the 1st The propagation attenuation coefficient of each model object is determined according to the type of correlation. For example, the propagation attenuation coefficient between the support structure and the surrounding rock is 0.8, and the propagation attenuation coefficient between adjacent monitoring points is 0.5. For the first The model object and the first Normalized spatial distance between model objects; The attenuation factor controls the rate at which propagation decays with normalized spatial distance.

[0044] Furthermore, the formula for the comprehensive cascading risk index is as follows: , in, For the first A comprehensive index of cascading risks for each model object; This is the collection of all high-risk model objects; For the first A comprehensive risk index for each model object.

[0045] Step S134: Determine the overall risk comprehensive index of the underground excavation project based on the risk comprehensive index of the labeled model objects and the chain risk comprehensive index of the associated model objects. After determining the comprehensive risk index of all identifiable model objects in the BIM risk integration model, the overall comprehensive risk index of the underground excavation project is determined based on the comprehensive risk index of all model objects marked in the BIM risk integration model and the comprehensive risk index of their associated model objects.

[0046] Furthermore, the formula for the overall risk composite index is as follows: , in, It serves as a comprehensive risk indicator for underground excavation projects, taking into account both high / medium risk at a single point and the risk chain propagation effect, and is used for overall risk assessment. For the first The importance weight of the model object is based on the importance weight of the first model object. The engineering criticality of each model object is determined (e.g., tunnel arch and fault fracture zone have higher weights). For indicator functions, when the first Risk level of each model object When it is high risk Take 1, when the first Risk level of each model object When it is medium risk Take 0.7; Risk level The total number of high-risk and medium-risk model objects; In order to be with the first A collection of model objects associated with a model object; This is the propagation impact balance coefficient, used to balance the weights of direct risk and propagation risk in the overall assessment.

[0047] Step S135: Trigger a graded early warning based on the overall risk comprehensive index of the underground excavation project, the risk comprehensive index of the labeled model objects, and the chain risk comprehensive index of the associated model objects; The warning triggering rules are as follows: If the comprehensive risk index of any model object is not less than the medium-risk threshold, that is: Or, the overall risk index of the underground excavation project is not less than the yellow warning threshold, that is: If so, a yellow alert will be triggered; If the comprehensive risk index of any model object is not less than the high-risk threshold, that is: Or, the overall risk index of the underground excavation project is not less than the red warning threshold, that is: Or, there exists a cascading risk composite index for any model object that is not less than the high-risk threshold, i.e.: If this occurs, a red alert will be triggered.

[0048] Step S140: Receive feedback information on the execution of closed-loop handling of early warning information, and update the BIM risk integration model according to the feedback information; When a tiered warning is triggered, the warning information is pushed to relevant engineering personnel so that they can initiate the corresponding handling process according to the warning level (e.g., yellow, red). After the relevant engineering personnel implement risk control measures, they will report the feedback information of the closed-loop handling of the warning. Based on the feedback information of the closed-loop handling of the warning (e.g., information of the model object after handling), the closed-loop management of "warning - response - handling - feedback" is realized.

[0049] Example 2 like Figure 4 As shown, this application provides a BIM-based dynamic early warning system 400 for construction risks in underground excavation projects, including: a model building module 410, a binding preprocessing module 420, a risk analysis and early warning module 430, and a risk closed-loop handling module 440.

[0050] The model building module 410 constructs an integrated BIM risk model for underground excavation projects based on engineering data and historical construction data of similar underground excavation projects.

[0051] BIM (Building Information Modeling) is a data-driven, simulated, and manageable "digital twin" of a building. It is a three-dimensional visual model in which each model object carries corresponding information and can be used throughout the entire lifecycle of a project (design → construction → operation and maintenance → demolition).

[0052] The model building module 410 includes: a data acquisition submodule 411, a basic model building submodule 412, a sub-model building submodule 413, an integrated submodule 414, and a model improvement submodule 415.

[0053] The data acquisition submodule 411 acquires the construction design drawings, geological survey reports, monitoring plans, construction plans, and historical construction data of similar underground excavation projects.

[0054] Before commencing underground excavation, it is essential to obtain engineering data for the project and historical construction data for similar projects. This engineering data includes: construction design drawings, geological survey reports, monitoring plans, and construction plans.

[0055] The construction design drawings include technical elements of the underground excavation project, such as: excavation cross-section (3.5m wide, 4.0m high), initial support (shotcrete + steel arch frame), secondary lining, construction shaft, monitoring points, etc. The geological survey report shows geological conditions and geological risk factors for the underground excavation project, such as: stratum distribution (silty clay, sand), soil and rock parameters (cohesion, internal friction angle), groundwater level (depth 5.2m), location of fault fracture zones, etc., and geological risk factors such as: surrounding rock collapse, water inrush, mudslide, and surrounding rock instability. The monitoring plan specifies the monitoring types and corresponding monitoring point layout requirements for underground excavation projects. For example, monitoring types include: tunnel crown settlement, surface settlement, anchor bolt axial force, and lining concrete stress. Monitoring point layout requirements are: crown settlement monitoring points must be placed along the tunnel crown centerline; surface settlement monitoring points must be placed on the surface above the tunnel; anchor bolt axial force monitoring points must be placed at the anchor bolt ends; and lining concrete stress monitoring points must be placed at key sections of the lining. The construction plan outlines the construction procedures and risk factors for underground excavation projects.

[0056] The basic model construction submodule 412 constructs the BIM basic model of the underground excavation project based on the construction design drawings.

[0057] Based on the technical elements of the underground excavation project contained in the construction design drawings, a BIM basic model of the underground excavation project is constructed using Revit software (a professional BIM modeling software). The BIM basic model establishes a complete three-dimensional coordinate system and engineering components within the three-dimensional coordinate system.

[0058] Sub-model construction sub-module 413 imports the geological survey report and monitoring plan into the BIM basic model to generate a geological BIM sub-model and a monitoring BIM sub-model.

[0059] The geological survey report is imported into the BIM base model. Based on the geological conditions information of the underground excavation project shown in the geological survey report, stratigraphic units are generated in the BIM base model. The stratigraphic units are then matched and integrated with all engineering components in the same three-dimensional coordinate system, thereby expanding the BIM base model to form a geological BIM sub-model specifically for expressing the geological environment.

[0060] The monitoring plan is imported into the BIM base model. Based on the monitoring type information and corresponding monitoring point layout requirements of the underground excavation project specified in the monitoring plan, corresponding monitoring points are added to the BIM base model. The monitoring points are in the same three-dimensional coordinate system as all engineering components. The monitoring type (arch crown settlement, ground surface settlement, anchor bolt axial force, lining concrete stress), monitoring frequency (once every 2 hours), and monitoring threshold (arch crown settlement ≤ 3 mm / d, ground surface settlement ≤ 2 mm / d) of each monitoring point are specified. The monitoring type, monitoring frequency, and monitoring threshold of each monitoring point are recorded in the additional information of the corresponding monitoring point. Thus, a monitoring BIM sub-model specifically for expressing safety monitoring is formed on the basis of the BIM base model.

[0061] The integrated sub-module 414 spatially matches and temporally correlates the construction plan with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model to generate an integrated BIM risk model.

[0062] Since the geological BIM sub-model, monitoring BIM sub-model, and BIM base model share the same three-dimensional coordinate system, and since the construction procedures for underground excavation projects specified in the construction plan are themselves a complete set of information containing spatial and temporal information, the construction procedures are spatially matched and temporally associated with engineering components in the BIM base model, stratigraphic units in the geological BIM sub-model, and monitoring points in the monitoring BIM sub-model according to the project mileage, construction location, and time sequence. This achieves the integrated integration of construction procedures with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model. For example, the spatial integration of construction procedures such as excavation, support, and grouting is combined with the temporal integration of the construction procedures, such as their sequence, construction duration, and construction steps, thereby generating an integrated BIM risk model.

[0063] The model improvement submodule 415 adds risk factors to the model objects of the BIM risk integration model based on geological survey reports, construction plans, and historical construction data of similar underground excavation projects, thereby improving the BIM risk integration model.

[0064] The geological survey report revealed geological risk factors, such as: surrounding rock collapse, water inrush, mudslide, and surrounding rock instability; the construction plan identified construction risk factors for the underground excavation project, such as: support failure, excessive settlement, excessive arch settlement, and face instability; historical construction data of similar underground excavation projects included historical risk factors such as collapses, settlement, leakage, and support damage that had occurred in previous similar underground excavation projects.

[0065] Geological risk factors, construction risk factors, and historical risk factors are added to the supplementary information of model objects such as engineering components, stratigraphic units, monitoring points, and construction procedures in the BIM integrated risk model, thereby improving the BIM integrated risk model. For example: adding the following to the supplementary information of engineering components (arches, shotcrete, anchors): support failure risk and deformation exceeding limit risk; adding the following to the supplementary information of stratigraphic units (faults, water-rich layers, weak surrounding rock): water inrush, collapse, and surrounding rock instability risk; adding the following to the supplementary information of monitoring points (arch crown subsidence, surface settlement): early warning threshold and exceeding standard risk; adding the following to construction procedures (excavation, support, grouting): procedure delay risk and construction disturbance risk; and binding historical risk factors such as historical risk cases, historical monitoring data, and accident cases of similar underground excavation projects to the corresponding model objects.

[0066] Based on this, all engineering components (e.g., excavation section outline components, excavation step components, shotcrete layers, steel arch structure components, steel mesh components), geological units (e.g., faults, water-rich layers, weak surrounding rock), monitoring points (e.g., arch crown settlement, surface settlement), and construction procedures (e.g., excavation, support, grouting) in the BIM basic model are uniquely coded. This unique code is used to make the model objects identifiable, measurable, traceable, collaborative, and traceable, thereby significantly improving the accuracy and efficiency of underground excavation construction management.

[0067] The binding preprocessing module 420 binds a unique identifier to the risk monitoring data of the same group collected, and preprocesses the risk monitoring data with the same identifier to obtain the comprehensive risk index of the corresponding model object in the BIM risk integration model.

[0068] Multi-source data acquisition equipment is deployed at the construction site of the underground excavation project to collect multi-source risk monitoring data. For example, eight surrounding rock displacement sensors are deployed in fault fracture zones and soft soil strata areas to monitor displacement within the enclosure and crown settlement; twelve mechanical sensors are deployed on the support structure to monitor anchor bolt axial force, steel arch frame internal force, and lining concrete stress; four environmental sensors are deployed in the construction area to monitor harmful gas concentration, groundwater level changes, temperature, and humidity; six construction parameter acquisition terminals are installed on the construction machinery to monitor excavation progress, support time, grouting parameters, etc.; and two manual data acquisition terminals are also provided to supplement and input data that cannot be automatically collected, such as blasting parameters.

[0069] Assign a unique identifier to risk monitoring data belonging to the same group (e.g., a rock displacement sensor, an anchor bolt axial force) to indicate that the risk monitoring value of this group corresponds to a certain model object in the BIM risk integration model. For example, the monitoring point (e.g., the left arch waist of the fault fracture zone K3+200 section), the construction process (e.g., the 3rd cycle of the excavation process), the corresponding engineering component (e.g., the 5th steel arch), and the corresponding stratum unit (e.g., the fault at the 20th meter). And perform the following preprocessing on risk monitoring data with the same identifier to obtain the comprehensive risk index of the corresponding model object.

[0070] First, outlier detection is performed on the risk monitoring data with the same identifier, and outliers are removed to avoid them affecting subsequent risk identification, thereby improving the accuracy of subsequent risk identification.

[0071] Furthermore, the formula for outlier detection is as follows: , in, For the first The first model object The first in the risk monitoring data Individual risk monitoring data; For the first The first model object The mean of risk monitoring data, , For the first The first model object The first in the risk monitoring data Individual risk monitoring data Quantity; For the first The first model object Standard deviation of risk monitoring data .

[0072] like Then the first The first model object The first type of risk monitoring data Individual risk monitoring data If it is an outlier, Then the first The first model object The first type of risk monitoring data Individual risk monitoring data This is a normal value.

[0073] Then, after removing outliers, the risk monitoring data with the same identifier are normalized to the [0,1] interval. This eliminates the dimensional differences in the risk monitoring data, avoids significant numerical disparities that could distort the BIM risk integration model, and ensures that all indicators participate fairly in subsequent risk identification. For example, surrounding rock displacement is measured in mm (e.g., 5, 12, 20); harmful gas concentration is measured in ppm (e.g., hundreds or thousands); and anchor bolt axial force is measured in kN (e.g., thousands or tens of thousands). If these data are not normalized, the anchor bolt axial force values ​​of several thousand will completely dominate the calculation results, while smaller values ​​such as surrounding rock displacement and harmful gas concentration will be overlooked. The BIM risk integration model will mistakenly believe that "anchor bolt axial force is the only risk."

[0074] Furthermore, the normalization formula is as follows: , in, For the first The first model object The first type of risk monitoring data Individual risk monitoring data The normalized value; For the first The first model object Minimum value of risk monitoring data; For the first The first model object The maximum value of risk monitoring data.

[0075] Finally, the normalized values ​​of the risk monitoring data with the same identifier are merged to transform them into a comprehensive risk index for the corresponding model objects that can be identified and calculated by the BIM integrated risk model. This simplifies subsequent risk identification and ensures the comprehensiveness of the data. The essence of this fusion is to synthesize indicators with different physical meanings, such as displacement, stress, water level, gas, and construction parameters, into a unified "comprehensive risk index," enabling the BIM integrated risk model to directly calculate, classify, and determine the risk level.

[0076] Furthermore, the fusion formula is as follows: , in, For the first A comprehensive risk index for each model object; For the first Risk weights for risk monitoring data; This refers to the number of types of risk monitoring data.

[0077] The risk analysis and early warning module 430 inputs the comprehensive risk index of the model object into the BIM risk integration model, analyzes the risk level and risk propagation impact of the model object, and triggers graded early warnings.

[0078] The risk analysis and early warning module 430 includes: indicator object binding submodule 431, comparison and labeling submodule 432, chain risk analysis submodule 433, overall risk assessment submodule 344, and graded early warning triggering submodule 345.

[0079] The indicator object binding submodule 431 binds the comprehensive risk indicators of model objects to the corresponding model objects in the BIM risk integration model.

[0080] Since each set of risk monitoring data that generates a comprehensive risk indicator is bound to a unique identifier, and this identifier corresponds to the corresponding model object in the BIM integrated risk model, after the comprehensive risk indicators of the model object are input into the BIM integrated risk model, the comprehensive risk indicators of the model object are bound to the corresponding model object in the BIM integrated risk model. This ensures that each comprehensive risk indicator can be mapped to a specific location in the BIM integrated risk model, providing spatial support for the targeting of risk identification and the accuracy of assessment.

[0081] The comparison and annotation submodule 432 compares the comprehensive risk index with the risk level threshold range of the model object it is bound to, and performs visual risk level annotation on the model object with a high / medium risk level.

[0082] The additional information of model objects in the BIM integrated risk model pre-records the risk level threshold range for each model object. Therefore, after binding the comprehensive risk index of the model object to the corresponding model object in step S131, the comprehensive risk index of each model object is compared with the risk level threshold range recorded in its additional information. The risk level threshold range is a pre-defined range based on engineering specifications, historical experience, and expert knowledge. For example: , in, For the first Risk level of each model object; The threshold for medium risk. This is a high-risk threshold, typically set to a value of [value missing]. , .

[0083] Based on the comparison results, model objects in the BIM risk integration model with high and medium risk levels are visually labeled to achieve a spatial representation of the risk level of model objects. For example, model objects are highlighted in different colors in the BIM risk integration model (e.g., yellow for medium risk, red for high risk), and risk labels pop up next to the model objects with the corresponding risk levels. Simultaneously, the comprehensive risk index and trigger time (i.e., the time when a medium or higher risk level is detected) are recorded in the model object's supplementary information, forming a risk event log for subsequent review of the labeled model object's data.

[0084] The chain risk analysis submodule 433 calculates the risk level of the labeled model object and the risk propagation influence coefficient of the associated model object, so as to determine the comprehensive chain risk index of the associated model object.

[0085] By utilizing the spatial topological relationships between engineering components, geological units, and monitoring points in the BIM risk integration model, and the temporal logical relationships of construction procedures, the risk level of the labeled model objects is analyzed to determine the risk propagation impact on the associated model objects. The corresponding risk propagation impact coefficient is calculated, and the chain risk comprehensive index of the model objects associated with the labeled model objects is determined based on the risk propagation impact coefficient, thereby realizing the risk propagation impact analysis of the model objects.

[0086] Furthermore, the formula for the risk transmission impact coefficient is as follows: , in, For the first The model object and the first Risk propagation impact coefficient among model objects; For the first The risk source intensity coefficient of each model object is used as a risk source, and its value is... ; For the first The model object and the first The relationship indicator variable between model objects, if the first... The model object and the first If there are spatial topological relationships (e.g., adjacency, support, coverage) or temporal logical relationships (e.g., preceding and subsequent processes) between model objects, then ,otherwise ; For the first The model object up to the 1st The propagation attenuation coefficient of each model object is determined according to the type of correlation. For example, the propagation attenuation coefficient between the support structure and the surrounding rock is 0.8, and the propagation attenuation coefficient between adjacent monitoring points is 0.5. For the first The model object and the first Normalized spatial distance between model objects; The attenuation factor controls the rate at which propagation decays with normalized spatial distance.

[0087] Furthermore, the formula for the comprehensive cascading risk index is as follows: , in, For the first A comprehensive index of cascading risks for each model object; This is the collection of all high-risk model objects; For the first A comprehensive risk index for each model object.

[0088] The overall risk assessment submodule 344 determines the overall risk comprehensive index of the underground excavation project based on the risk comprehensive index of the labeled model objects and the chain risk comprehensive index of the associated model objects.

[0089] After determining the comprehensive risk index of all identifiable model objects in the BIM risk integration model, the overall comprehensive risk index of the underground excavation project is determined based on the comprehensive risk index of all model objects marked in the BIM risk integration model and the comprehensive risk index of their associated model objects.

[0090] Furthermore, the formula for the overall risk composite index is as follows: , in, It serves as a comprehensive risk indicator for underground excavation projects, taking into account both high / medium risk at a single point and the risk chain propagation effect, and is used for overall risk assessment. For the first The importance weight of the model object is based on the importance weight of the first model object. The engineering criticality of each model object is determined (e.g., tunnel arch and fault fracture zone have higher weights). For indicator functions, when the first Risk level of each model object When it is high risk Take 1, when the first Risk level of each model object When it is medium risk Take 0.7; Risk level The total number of high-risk and medium-risk model objects; In order to be with the first A collection of model objects associated with a model object; This is the propagation impact balance coefficient, used to balance the weights of direct risk and propagation risk in the overall assessment.

[0091] The graded early warning triggering submodule 345 triggers graded early warnings based on the overall risk comprehensive index of the underground excavation project, the risk comprehensive index of the labeled model objects, and the chain risk comprehensive index of the associated model objects.

[0092] The warning triggering rules are as follows: If the comprehensive risk index of any model object is not less than the medium-risk threshold, that is: Or, the overall risk index of the underground excavation project is not less than the yellow warning threshold, that is: If so, a yellow alert will be triggered; If the comprehensive risk index of any model object is not less than the high-risk threshold, that is: Or, the overall risk index of the underground excavation project is not less than the red warning threshold, that is: Or, there exists a cascading risk composite index for any model object that is not less than the high-risk threshold, i.e.: If this occurs, a red alert will be triggered.

[0093] The risk closed-loop handling module 440 receives feedback information on the early warning information and performs early warning closed-loop handling accordingly, and updates the BIM risk integration model based on the feedback information.

[0094] When a tiered warning is triggered, the warning information is pushed to relevant engineering personnel so that they can initiate the corresponding handling process according to the warning level (e.g., yellow, red). After the relevant engineering personnel implement risk control measures, they will report the feedback information of the closed-loop handling of the warning. Based on the feedback information of the closed-loop handling of the warning (e.g., information of the model object after handling), the closed-loop management of "warning - response - handling - feedback" is realized.

[0095] This application covers multiple risk factors, including geology, construction, environment, and structure, enabling comprehensive and early identification of risks in underground excavation projects and timely containment of accidents. Furthermore, it provides visualized risk warnings, intuitively presenting risk levels and locations, facilitating rapid risk assessment for management personnel. Additionally, it utilizes data from various sources related to underground excavation projects, including historical data from similar projects, resulting in high data utilization and significantly improved risk assessment accuracy. Moreover, it incorporates both early warning and response mechanisms, forming a closed-loop management system.

[0096] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0097] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A BIM-based dynamic early warning method for construction risks in underground tunneling projects, characterized in that, Includes the following steps: Step S110: Based on the engineering data of the underground tunneling project and the historical construction data of similar underground tunneling projects, construct a BIM risk integration model for the underground tunneling project. Step S120: Bind a unique identifier to the risk monitoring data of the same group collected, preprocess the risk monitoring data with the same identifier, and obtain the comprehensive risk index of the corresponding model object in the BIM risk integration model. Step S130: Input the comprehensive risk index of the model object into the BIM integrated risk model, analyze the risk level and risk propagation impact of the model object, generate a risk assessment report and trigger a graded early warning; Step S140: Receive feedback information on the early warning closed-loop handling based on the early warning information, and update the BIM risk integration model according to the feedback information.

2. The BIM-based dynamic early warning method for construction risks in underground tunneling projects according to claim 1, characterized in that, Step S110 includes the following sub-steps: Step S111: Obtain the construction design drawings, geological survey report, monitoring plan, construction plan, and historical construction data of similar underground excavation projects for the underground tunneling project; Step S112: Construct the BIM basic model of the underground excavation project based on the construction design drawings; Step S113: Import the geological survey report and monitoring plan into the BIM basic model respectively to generate the geological BIM sub-model and the monitoring BIM sub-model; Step S114: Spatially match and temporally correlate the construction plan with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model to generate an integrated BIM risk model; Step S115: Based on the geological survey report, construction plan, and historical construction data of similar underground excavation projects, add risk factors to the model objects of the BIM risk integration model to improve the BIM risk integration model.

3. The BIM-based dynamic early warning method for underground tunneling construction risks according to claim 1 or 2, characterized in that, Preprocessing includes the following sub-steps: Perform outlier detection on risk monitoring data with the same identifier and remove outliers; After removing outliers, the risk monitoring data with the same identifier are normalized to the [0,1] interval; The normalized values ​​of risk monitoring data with the same identifier are merged to transform them into a comprehensive risk indicator for the corresponding model object.

4. The BIM-based dynamic early warning method for underground tunneling construction risks according to claim 1 or 2, characterized in that, Step S130 includes the following sub-steps: Step S131: In the BIM risk integration model, bind the comprehensive risk index of the model object to the corresponding model object; Step S132: Compare the risk comprehensive index with the risk level threshold range of the model object it is bound to, and visually label the risk level of the model object with a high / medium risk level. Step S133: Calculate the risk level of the labeled model object on the risk propagation influence coefficient of the associated model object, so as to determine the comprehensive index of the chain risk of the associated model object. Step S134: Determine the overall risk comprehensive index of the underground excavation project based on the risk comprehensive index of the labeled model objects and the chain risk comprehensive index of the associated model objects. Step S135: Trigger a graded early warning based on the overall risk comprehensive index of the underground excavation project, the risk comprehensive index of the labeled model objects, and the chain risk comprehensive index of the associated model objects.

5. The BIM-based dynamic early warning method for construction risks in underground tunneling projects according to claim 4, characterized in that, The triggers for tiered early warnings include: If the comprehensive risk index of any model object is not less than the medium risk threshold, or the comprehensive risk index of the underground excavation project is not less than the yellow warning threshold, then a yellow warning will be triggered. If the comprehensive risk index of any model object is not less than the high-risk threshold, or the comprehensive risk index of the underground excavation project is not less than the red warning threshold, or the comprehensive risk index of any model object is not less than the high-risk threshold, then a red warning is triggered.

6. A BIM-based dynamic early warning system for construction risks in underground tunneling projects, characterized in that, include: The module includes a model building module, a binding preprocessing module, a risk analysis and early warning module, and a risk closed-loop handling module. The model building module constructs an integrated BIM risk model for underground tunneling projects based on engineering data and historical construction data of similar underground tunneling projects. The binding preprocessing module binds a unique identifier to the risk monitoring data of the same group collected, and preprocesses the risk monitoring data with the same identifier to obtain the comprehensive risk index of the corresponding model object in the BIM risk integration model. The risk analysis and early warning module inputs the comprehensive risk indicators of the model objects into the BIM integrated risk model, analyzes the risk level and risk propagation impact of the model objects, and triggers tiered early warnings. The risk closed-loop handling module receives feedback information on the early warning information and performs early warning closed-loop handling accordingly, and updates the BIM risk integration model based on the feedback information.

7. The BIM-based dynamic early warning system for construction risks in underground tunneling projects according to claim 6, characterized in that, The model building module includes: a data acquisition submodule, a basic model building submodule, a sub-model building submodule, an integrated submodule, and a model improvement submodule; The data acquisition submodule acquires construction design drawings, geological survey reports, monitoring plans, construction plans, and historical construction data of similar underground excavation projects. The basic model construction submodule constructs the BIM basic model of the underground excavation project based on the construction design drawings; The sub-model construction sub-module imports the geological survey report and monitoring plan into the BIM base model, respectively, to generate a geological BIM sub-model and a monitoring BIM sub-model. The integrated sub-module spatially matches and temporally correlates the construction plan with the BIM base model, geological BIM sub-model, and monitoring BIM sub-model to generate an integrated BIM risk model. The model improvement submodule adds risk factors to the model objects of the BIM risk integration model based on geological survey reports, construction plans, and historical construction data of similar underground excavation projects, thereby improving the BIM risk integration model.

8. The BIM-based dynamic early warning system for construction risks in underground tunneling projects according to claim 6 or 7, characterized in that, The binding preprocessing module performs outlier detection on the risk monitoring data with the same identifier and removes the outliers. After removing the outliers, the risk monitoring data with the same identifier is normalized to the [0,1] interval. The normalized values ​​of the risk monitoring data with the same identifier are then merged to transform them into a comprehensive risk indicator for the corresponding model object.

9. The BIM-based dynamic early warning system for construction risks in underground tunneling projects according to claim 6 or 7, characterized in that, The risk analysis and early warning module includes: indicator object binding submodule, comparison and labeling submodule, chain risk analysis submodule, overall risk assessment submodule, and graded early warning triggering submodule; The indicator object binding submodule binds the comprehensive risk indicators of model objects to the corresponding model objects in the BIM integrated risk model; The comparison and labeling submodule compares the comprehensive risk index with the risk level threshold range of the model object it is bound to, and performs visual risk level labeling on the model object with a high / medium risk level. The chain risk analysis submodule calculates the risk level of the labeled model object and its associated model object risk propagation influence coefficient to determine the chain risk comprehensive index of the associated model object; The overall risk assessment submodule determines the overall risk comprehensive index of the underground excavation project based on the comprehensive risk index of the labeled model objects and the comprehensive risk index of the chain of model objects associated with them. The tiered early warning triggering submodule triggers tiered early warnings based on the overall risk comprehensive index of the underground excavation project, the risk comprehensive index of the labeled model objects, and the chain risk comprehensive index of the associated model objects.

10. The BIM-based dynamic early warning system for construction risks in underground tunneling projects according to claim 9, characterized in that, If the comprehensive risk index of any model object is not less than the medium risk threshold, or the comprehensive risk index of the underground excavation project is not less than the yellow warning threshold, then a yellow warning will be triggered. If the comprehensive risk index of any model object is not less than the high-risk threshold, or the comprehensive risk index of the underground excavation project is not less than the red warning threshold, or the comprehensive risk index of any model object is not less than the high-risk threshold, then a red warning is triggered.

Images