Building construction risk intelligent identification and early warning method and system based on BIM model

Abstract

The invention provides a building construction risk intelligent identification and early warning method and system based on a BIM model, and relates to the technical field of construction risk early warning, and the method comprises the steps: obtaining a building information model and historical meteorological data of a target scene; constructing a numerical wind tunnel model, carrying out transient wind field simulation, carrying out region identification by combining a transient wind field simulation result, and obtaining a prior attention region set; according to the construction progress information of the target scene and the prior attention area set, defining a real-time attention area set of the target scene; collecting real-time wind power data, and traversing the real-time attention region set to perform dynamic risk assessment in combination with the real-time wind power data and the numerical wind tunnel model to obtain a real-time risk coefficient set; and performing visual marking in a visual interface of the building information model, and performing risk early warning in combination with a visual marking result and the real-time risk coefficient set. The technical problem that in the prior art, a traditional building construction risk early warning method is insufficient in early warning real-time performance is solved.

Description

Technical Field

[0001] The present invention relates to the field of construction risk early warning, and in particular to a method and system for intelligent identification and early warning of construction risks based on a BIM model. Background Art

[0002] Construction risk warning is the core link in ensuring construction safety. A timely and effective warning mechanism can identify safety hazards in advance and reduce the occurrence rate of safety accidents.

[0003] However, traditional construction risk warnings usually rely on manual inspections and experience-based judgments. The assessment of environmental risks such as wind-induced risks lacks a quantitative basis, and it is difficult to systematically integrate multi-source information, resulting in insufficient real-time risk warnings and the possibility of missing opportunities for action.

[0004] With the maturity of BIM technology, its advantages in three-dimensional modeling, full-factor information integration and visualization have gradually become prominent, which provides technical support for breaking through the limitations of traditional construction risk identification and early warning.

[0005] Therefore, there is an urgent need for an intelligent identification and early warning method for construction risks based on the BIM model, which can achieve accurate identification and timely early warning of construction risks by integrating BIM technology, real-time monitoring data and other multi-source information. Summary of the Invention

[0006] The present invention aims to solve the technical problem that the traditional construction risk early warning method in the prior art has insufficient early warning real-time performance, and provides a construction risk intelligent identification and early warning method and system based on the BIM model.

[0007] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a method for intelligent identification and early warning of construction risks based on a BIM model, comprising: Obtain the building information model and historical meteorological data of the target scene; Constructing a numerical wind tunnel model based on the building information model and historical meteorological data, performing transient wind field simulation, and performing region identification based on the transient wind field simulation results to obtain a priori focus region set; Defining a real-time focus area set of the target scene based on the construction progress information of the target scene and the prior focus area set; Collecting real-time wind data, and combining the real-time wind data with the numerical wind tunnel model, traversing the real-time focus area set to perform dynamic risk assessment and obtain a real-time risk coefficient set; In the visualization interface of the building information model, a plurality of real-time focus areas are visually marked according to the real-time risk coefficient set, and risk warning is performed in combination with the visualization marking result and the real-time risk coefficient set.

[0008] In a second aspect, the present invention provides an intelligent identification and early warning system for construction risks based on a BIM model, comprising: Data acquisition module, used to obtain the building information model and historical meteorological data of the target scene; A region identification module is used to construct a numerical wind tunnel model based on the building information model and historical meteorological data, perform transient wind field simulation, and perform region identification based on the transient wind field simulation results to obtain a priori focus region set; An area of ​​interest definition module, configured to define a real-time area of ​​interest set of the target scene based on the construction progress information of the target scene and the a priori area of ​​interest set; a risk assessment module, configured to collect real-time wind data, and combine the real-time wind data with the numerical wind tunnel model to traverse the real-time focus area set for dynamic risk assessment to obtain a real-time risk coefficient set; The risk warning module is used to visually mark multiple real-time focus areas according to the real-time risk coefficient set in the visualization interface of the building information model, and to perform risk warning in combination with the visualization marking results and the real-time risk coefficient set.

[0009] The beneficial effects of the present invention are: Compared to the existing technology, this application first obtains the building information model and historical meteorological data of the target scene, solving the dual limitations of fuzzy structural information and missing environmental parameters in traditional risk assessment, and providing a data foundation for subsequent dynamic risk simulation and early warning. Secondly, a numerical wind tunnel model is constructed based on the building information model and historical meteorological data, and transient wind field simulation is performed. The transient wind field simulation results are combined to identify regions and obtain a priori focus area set. Using historical data and simulation technology, a list of high-risk areas that may exist in each construction stage is determined, and potential risk points are locked in advance, providing reliable support for subsequent risk assessment and risk early warning. Thirdly, based on the construction progress information of the target scene and the prior focus area set, a real-time focus area set is defined for the target scene, ensuring that all high-risk points in the current stage are included in the monitoring scope, providing precise targeting for dynamic risk assessment. Furthermore, real-time wind data is collected and combined with the real-time wind data and the numerical wind tunnel model, and the real-time focus area set is traversed to perform dynamic risk assessment, obtaining a real-time risk coefficient set, realizing dynamic quantification of current construction risks, solving the problem of insufficient real-time performance of traditional methods, and providing a reliable quantitative basis for subsequent risk early warning. Finally, in the visualization interface of the building information model, multiple real-time areas of concern are visually marked according to the real-time risk coefficient set, and risk warnings are issued in combination with the visualization marking results and the real-time risk coefficient set. Through color visualization and multiple risk warning methods, risk information is ensured to reach the entire chain from the management end to the operation end, greatly shortening the risk response time.

[0010] Through the above technical solution, this application provides a structured three-dimensional structural benchmark and quantitative environmental parameter samples for risk assessment by obtaining the BIM model and historical meteorological data of the target scene, solving the problems of fuzzy structural information and missing environmental data in traditional methods; constructing a numerical wind tunnel model and conducting transient wind field simulation to identify a priori focus area sets, achieving accurate positioning of historical high-risk areas, overcoming the limitations of traditional assessments that rely on experience and lack systematicity; dynamically defining real-time focus area sets through construction progress information, allowing the risk monitoring range to be dynamically adjusted with the construction stage, avoiding the dislocation of monitoring dynamic scenes; combining real-time wind data with the numerical wind tunnel model for dynamic risk assessment, outputting a real-time risk coefficient set, achieving real-time quantitative update of risks, and solving the problem of early warning lag; finally, completing risk marking and multi-channel early warning in the BIM visual interface, achieving closed-loop management from risk identification to disposal. In this way, the efficiency of identifying construction risks and the timeliness of early warning are improved, providing intelligent technical support for construction safety management. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of the process of the intelligent identification and early warning method for construction risks based on the BIM model provided by the present invention; Figure 2 This is a structural diagram of the intelligent identification and early warning system for construction risks based on the BIM model provided by the present invention.

[0012] In the accompanying drawings, the components represented by the reference numerals are as follows: Data acquisition module 11, area identification module 12, focus area definition module 13, risk assessment module 14, risk warning module 15. DETAILED DESCRIPTION

[0013] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0014] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0015] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0016] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a method for intelligent identification and early warning of construction risks based on a BIM model, comprising: S10: Obtain a building information model and historical meteorological data of the target scene.

[0017] Traditional construction risk management relies heavily on two-dimensional drawings and empirical judgment, making it difficult to accurately depict the dynamic structural form of the construction area, such as changes in the spatial layout of scaffolding erection and component hoisting. Furthermore, there is a lack of quantitative evidence for predicting risks related to environmental factors such as wind. BIM models, on the other hand, provide three-dimensional structured data containing multi-dimensional information such as geometric dimensions, material properties, and construction phase markers, providing an accurate structural form benchmark for risk assessment. Historical meteorological data also provides a data foundation for identifying extreme wind conditions and is the core basis for determining the most adverse environmental convection information.

[0018] To address the above issues, this application obtains the building information model and historical meteorological data of the target scene.

[0019] The Building Information Model (BIM) refers to the BIM model of the target construction scenario, encompassing geometric information (e.g., structural dimensions, spatial layout, component location), physical information (e.g., material strength, fire resistance rating), and functional information (e.g., component connection methods, construction process requirements), throughout the building's entire lifecycle. For example, a native BIM model containing all-disciplinary data, including architectural, structural, and mechanical and electrical, can be obtained from the design firm and used as the BIM for the target scenario. This BIM provides a spatial benchmark for risk assessment. For example, wind speed data (e.g., average wind speed, maximum wind speed), wind direction data (e.g., wind direction frequency distribution, prevailing wind direction), and extreme weather records (e.g., typhoons and severe convective weather) can be obtained from meteorological authorities, meteorological databases, or third-party meteorological service platforms. Historical meteorological data provides an environmental benchmark for wind field simulations. Analyzing historical wind environment characteristics, especially extreme wind conditions, provides historical reference for subsequent identification of high-risk areas, ensuring that risk assessments encompass potential adverse meteorological conditions.

[0020] Furthermore, step S10 of the method, after obtaining the building information model and historical meteorological data of the target scene, includes: Obtain construction implementation planning information for the target scenario and extract multiple construction phases; Marking the building information model in stages according to the plurality of construction stages to obtain a model marking result, wherein the model marking result includes a plurality of groups of stage marks and associated stage model boundaries; The building information model and the model marking result are stored in association.

[0021] In an embodiment of the present application, the construction implementation planning information of the target scenario is first obtained, and multiple construction phases are extracted. The construction implementation planning information includes documents such as the overall construction schedule, the division of sub-items, and the process connection plan. For example, the construction implementation planning information of the target scenario is obtained, and multiple construction phases are extracted from the overall construction schedule and the division of sub-items. For example, the foundation phase lasts 1-30 days, the main structure phase lasts 31-180 days, and the decoration phase lasts 181-270 days. By extracting multiple construction phases, the dynamically changing construction process is broken down into static phase units, so that the risk assessment can adapt to the construction content of different phases.

[0022] Secondly, the building information model is marked in stages according to multiple construction stages to obtain model marking results, wherein the model marking results include multiple sets of stage marks and associated stage model boundaries. For example, based on the extracted construction stages, stage marking is performed in the BIM model to obtain model marking results including multiple sets of stage marks and associated stage model boundaries, wherein the stage marking refers to assigning a unique identifier to each construction stage, such as ST-01 for foundation, ST-02 for the third floor of the main structure, etc., and associating information such as the construction content, construction period, and responsible person of the stage; the stage model boundary refers to the use of three-dimensional boundary lines or color blocks in the BIM model to delineate the construction area of ​​the stage, such as the boundary of the third floor of the main structure construction stage includes the third floor slab, wall components and corresponding working surface, material storage area, etc.; the model marking results can reflect the specific construction information of the stage.

[0023] Finally, the building information model and the model marking results are stored in an associative manner. For example, database technology, such as the associative storage function of a BIM collaboration platform, can be used to bind the 3D geometric data of the BIM model with the model marking results. Later, when conducting wind farm simulations or risk assessments, the building information model of the current construction phase can be directly retrieved instead of the complete building model. This reduces the computational effort and prevents irrelevant areas from interfering with risk analysis.

[0024] In summary, compared to existing technologies, this application obtains a building information model and historical meteorological data for the target scene. This overcomes the dual limitations of traditional risk assessments: fuzzy structural information and missing environmental parameters, providing a data foundation for subsequent dynamic risk simulation and early warning.

[0025] S20: constructing a numerical wind tunnel model based on the building information model and historical meteorological data, performing transient wind field simulation, and performing region identification in combination with the transient wind field simulation results to obtain a priori focus region set.

[0026] Building construction is a dynamic process. The structural morphology at different construction stages varies significantly, causing the effects of wind loads to change dynamically with the construction stage. In addition, natural wind fields are transient, such as gusts and turbulence. It is impossible to accurately capture the impact of wind on the construction area through empirical judgment or simplified calculations alone, which may lead to delayed or distorted risk assessment.

[0027] To address the above issues, this application constructs a numerical wind tunnel model based on the building information model and historical meteorological data, performs transient wind field simulation, and combines the transient wind field simulation results to perform region identification and obtain a priori focus area set.

[0028] Specifically, step S20 in the method includes: Randomly selecting the building information model according to the model marking result to obtain a regional building information model; In combination with a preset time window constraint, the historical meteorological data is parsed to obtain the most unfavorable environmental convection information, wherein the most unfavorable environmental convection information includes at least a wind speed index and a wind direction index; Based on the regional building information model and the most unfavorable environmental convection information, construct the numerical wind tunnel model, and perform a transient wind field simulation accordingly to obtain the transient wind field simulation results, wherein the transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Traversing the model marking results, iteratively selecting the regional building information model to perform transient wind field simulation; Dual-target threshold screening is performed based on the multiple transient wind field simulation results to determine multiple prior regions of interest, obtain the prior region of interest set, and establish a region-stage association relationship between the prior region of interest set and the model labeling result.

[0029] In the embodiments of the present application, a building information model is first randomly selected based on the model labeling results to obtain a regional building information model. For example, a local regional model of a certain construction phase can be randomly selected from the building information model based on multiple sets of stage labels from the model labeling results. For example, the rebar binding area of ​​the second floor of the main structure can be used as the regional building information model. The regional building information model only contains the under-construction components, temporary facilities, and work spaces of that construction phase. Its data size is much smaller than that of the complete building model, which can significantly reduce the computational load of subsequent transient wind field simulations, improve analysis efficiency, and accurately focus on the risk characteristics unique to that construction phase.

[0030] Secondly, historical meteorological data is analyzed to obtain the most unfavorable environmental convection information, combined with a preset time window constraint. The preset time window constraint is a pre-set statistical range of historical meteorological data that matches the current construction phase. For example, if the construction phase corresponding to the current regional building information model is July, the preset time window constraint can be determined as meteorological data from July of the past five years. Those skilled in the art can dynamically adjust the selected meteorological data based on the climate characteristics of the construction area and the project cycle to ensure that the selected meteorological data is relevant to the wind environment characteristics during the actual construction period. The most unfavorable environmental convection information includes at least wind speed and wind direction indicators. Wind speed indicators include the historical maximum instantaneous wind speed, short-term gust peaks, and wind speeds exceeding construction safety limits within the preset time window constraint. Wind direction indicators include the wind direction directly facing the vertical impact of airflow on temporary enclosures or the facade of aerial work platforms within the preset time window constraint, as well as dangerous wind directions that have historically caused safety accidents. By defining the extreme wind environment parameters for the same period in history, extreme working condition inputs are provided for the construction of the numerical wind tunnel model, ensuring that the simulation scenario fully covers the wind environment likely to be encountered during the construction phase.

[0031] Next, based on the regional building information model and the most unfavorable environmental convection information, a numerical wind tunnel model was constructed, and transient wind field simulations were performed accordingly to obtain transient wind field simulation results. The numerical wind tunnel model is a virtual wind tunnel system built using computational fluid dynamics (CFD) technology. It can accurately simulate the fluid motion state of air flowing through the construction area. The input features of the numerical wind tunnel model include the regional building information model (reflecting the structural form details of the current construction stage) and the most unfavorable environmental convection information (including wind speed and direction in extreme wind conditions). Among them, transient wind field simulation refers to the simulation of the dynamic evolution process of the non-steady-state wind field, that is, the simulation of the irregular changes in wind speed and direction over time, such as the sudden increase in gusts and short-term changes in wind direction. The transient wind field simulation results include at least wind pressure distribution and turbulence distribution. Wind pressure distribution refers to the spatial distribution of pressure and unit area force generated by wind on various components in the regional building information model (such as scaffolding poles, formwork, temporarily stacked materials, etc.). High-pressure areas may cause deformation or even collapse of components. Turbulence distribution refers to the random pulsation intensity of the airflow velocity vector, such as the irregular fluctuation amplitude of wind speed and wind direction. High turbulence areas may cause scaffolding flutter, hanging components swing, and interfere with the balance stability of high-altitude workers. By performing transient wind field simulation under the most unfavorable environmental convection information, the dynamic interaction process between wind and construction area can be restored, providing quantifiable physical parameters for risk assessment.

[0032] Exemplarily, transient wind field simulation can be carried out through the following technical paths: based on computer fluid dynamics (CFD) technology, the regional building information model is divided into structured grids. The grid size can be dynamically set according to actual needs. The inlet boundary is set as the velocity inlet (matching the wind speed index and wind direction index), and the outlet boundary is set as the free flow outlet. The Reynolds time-averaged equations (RANS) combined with the k-ε turbulence model can be used for solution. Finally, the transient wind field simulation is completed through iterative calculation, and the wind pressure cloud map, turbulence intensity vector map and other results of the construction area are output.

[0033] Furthermore, the model marking results are traversed, and the regional building information model is iteratively selected from the building information model. The transient wind field simulation is performed according to the aforementioned method to ensure that the wind-induced risk simulation of all construction stages is covered.

[0034] Finally, dual-target threshold screening is performed based on multiple transient wind field simulation results to determine multiple prior regions of interest, obtain a priori region of interest set, and establish a region-stage correlation relationship between the prior region of interest set and the model labeling results. Among them, dual-target threshold screening is based on multiple transient wind field simulation results, combined with the characteristics of the construction stage, to set risk thresholds for wind pressure distribution and turbulence distribution respectively. For example, during the main structure construction stage, wind pressure is set to ≥0.5kPa and turbulence ≥0.2, while during the roof operation stage, wind pressure can be adjusted to ≥0.3kPa and turbulence ≥0.15 due to the special high-altitude environment. Areas that meet both risk thresholds are screened as prior focus areas. Prior focus areas have high wind pressure and turbulence under extreme wind conditions and are high-risk points that have been quantitatively verified in historical simulations. All the prior focus areas obtained are integrated to form a prior focus area set, such as the corner of the main three-story scaffolding and the edge of the roof material storage area. The prior focus area set is a list of high-risk areas based on historical data and transient wind field simulations, providing key monitoring targets for subsequent real-time risk monitoring during the construction stage. Exemplarily, based on the model labeling results, the prior area of ​​interest is associated with the construction stage in the model labeling results. For example, the corner of the main 3-layer scaffolding is associated with the 3rd layer construction stage of the main structure, and the edge of the roofing material storage area is associated with the roofing project construction stage, forming a region-stage association relationship between the prior area of ​​interest set and the model labeling results. Subsequently, according to the real-time construction stage, the prior area of ​​interest corresponding to the stage can be quickly retrieved to avoid inefficiency caused by indiscriminate monitoring.

[0035] In summary, compared to existing technologies, this application constructs a numerical wind tunnel model based on the building information model and historical meteorological data, performs transient wind field simulations, and combines these transient wind field simulation results to identify regions and obtain a priori sets of areas of concern. In this way, using historical data and simulation technology, a list of high-risk areas that may exist during each construction phase can be identified, potential risk points can be identified in advance, and provide reliable support for subsequent risk assessments and risk warnings.

[0036] S30: defining a real-time focus area set of the target scene according to the construction progress information of the target scene and the a priori focus area set.

[0037] Traditional risk management methods mostly rely on fixed risk lists or manual inspections. They are unable to dynamically adapt to real-time changes in the construction phase, and it is difficult to couple and analyze real risks with historical risk points. This may ultimately lead to delayed risk warnings and missed inspections of key points.

[0038] To address the above issues, this application defines a real-time focus area set for the target scene based on the construction progress information of the target scene and the prior focus area set.

[0039] Specifically, step S30 in the method includes: Determine the real-time construction stage based on the construction progress information and extract the corresponding stage engineering quantity information; Analyze the phased engineering quantity information to determine multiple phased focus areas of the target scene, wherein the phased focus areas at least include a material storage area and a personnel gathering area; Based on the real-time construction stage, the a priori focus area set is called in stages in combination with the area-stage association relationship to obtain a stage-a priori focus area set; A plurality of the stage regions of interest are combined with the stage-prior region of interest set to obtain the real-time region of interest set.

[0040] In the embodiments of the present application, the real-time construction phase is first determined based on the construction progress information, and the corresponding phase quantity information is extracted. For example, the real-time construction phase can be determined from the progress report, supervision log, and construction status update of the BIM model. For example, the real-time construction phase corresponding to the steel bar binding of the second layer of the main structure is the construction of the second layer of the main structure. Specific quantity information, such as the construction scope, work content, resource distribution, etc. of the real-time construction phase, is then extracted from the real-time construction phase as the phase quantity information.

[0041] Secondly, analyze the phase engineering quantity information to determine multiple phase focus areas of the target scene, where the phase focus areas include at least material storage areas and personnel gathering areas, which are areas with naturally high-risk attributes. For example, the three-dimensional coordinates of the material storage areas and personnel gathering areas of the target scene can be determined from the resource distribution in the phase engineering quantity information. For example, the area with an elevation of 12.5m, X=10-15m, and Y=20-25m on the second floor is the steel bar storage area, and the area with an elevation of 12.5m, X=25-28m, and Y=30-35m on the second floor is the workers' rest area. In this way, multiple phase focus areas with naturally high-risk attributes in the current construction phase are identified, and the phase focus areas must be included in the monitoring scope.

[0042] Thirdly, based on the real-time construction stage, the prior area of ​​interest set is called in stages in combination with the region-stage association relationship to obtain the stage-prior area of ​​interest set. For example, when the real-time construction stage is the construction of the second floor of the main structure, based on the region-stage association relationship, all the prior areas of interest associated with the real-time construction stage are retrieved from the prior area of ​​interest set. For example, the construction of the second floor of the main structure is associated with the corner of the external scaffolding on the west side of the second floor of the main structure in the prior area of ​​interest, and it is included in the stage-prior area of ​​interest set. The stage-prior area of ​​interest set is a high-risk area that is susceptible to extreme wind conditions at this stage, as verified by historical meteorological data and transient wind field simulation. Its risk characteristics are highly matched with the real-time construction stage, providing an accurate reference based on historical experience for risk monitoring in the current stage.

[0043] Finally, multiple phase focus areas are merged with the phase-prior focus area set to obtain a real-time focus area set. For example, multiple phase focus areas with high-risk attributes in the real-time construction phase are spatially merged with the phase-prior focus area set determined based on historical meteorological data and transient wind field simulation to form a real-time focus area set that includes both real-world risk areas and historical risk areas. For example, if multiple phase focus areas are the steel bar stacking area and the worker rest area, and the phase-prior focus area set is the west side external scaffolding on the second floor of the main structure, the real-time focus area set obtained after merging is: {steel bar stacking area, worker rest area, west side external scaffolding on the second floor of the main structure}. If multiple phase focus areas overlap with the phase-prior focus area set, they are merged into the same monitoring unit to avoid repeated monitoring. In this way, the risk points in the real-time construction phase and the risk areas prone to problems in this phase based on historical experience are covered, forming an accurate and comprehensive list of real-time monitoring objects.

[0044] In summary, compared to existing technologies, this application defines a real-time set of focus areas for the target scene based on the construction progress information of the target scene and the a priori set of focus areas. This ensures that all high-risk points at the current stage are included in the monitoring scope, providing precise targeting for dynamic risk assessment.

[0045] S40: collecting real-time wind data, and combining the real-time wind data with the numerical wind tunnel model, traversing the real-time focus area set to perform dynamic risk assessment and obtain a real-time risk coefficient set.

[0046] Construction risk assessment needs to respond to real-time scenarios such as changes in construction stages and dynamic changes in wind fields. However, traditional methods mostly rely on fixed risk lists or offline numerical simulations, resulting in a lag between risk assessment results and actual working conditions, and an inability to capture real-time risks in a timely manner.

[0047] To address the above issues, the present application collects real-time wind data, combines the real-time wind data with the numerical wind tunnel model, traverses the real-time focus area set to perform dynamic risk assessment, and obtains a real-time risk coefficient set.

[0048] Specifically, step S40 in the method includes: Based on the real-time construction stage, extracting the regional building information model accordingly; Analyzing the real-time wind data, extracting real-time wind speed and real-time wind direction, and updating the regional building information model according to the real-time wind speed and the real-time wind direction; Performing a transient wind field simulation based on the updated regional building information model to obtain real-time transient wind field simulation results, wherein the real-time transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Position matching is performed on the real-time focus area set and the real-time transient wind field simulation result, and risk assessment indicators of multiple real-time focus areas are extracted, wherein the risk assessment indicators include at least turbulence intensity, turbulence kinetic energy, pressure fluctuation rate and dynamic pressure; Dynamic risk assessment of the plurality of real-time focus areas is performed according to the risk assessment index, and a plurality of real-time risk coefficients are obtained and output as the real-time risk coefficient set.

[0049] In the present embodiment, a regional building information model is first extracted based on the real-time construction phase. For example, if the real-time construction phase is the second floor of the main structure, the regional building information model for the second floor of the main structure is extracted from the BIM model. The regional building information model includes parameters such as the components under construction and temporary facilities for the second floor of the main structure.

[0050] Secondly, analyze the real-time wind data, extract the real-time wind speed and real-time wind direction, and update the regional building information model based on the real-time wind speed and real-time wind direction. For example, real-time wind data can be collected by anemometers, wind direction sensors, and other equipment deployed at the construction site. The real-time wind speed (e.g., 8m / s) and real-time wind direction (e.g., southeast wind, 135°) can then be extracted from the data. The real-time wind speed and real-time wind direction are then used as new boundary conditions and input into the regional building information model, replacing the fixed parameters of the most unfavorable environmental convection information. The updated regional building information model is then linked in real time with the on-site wind conditions, ensuring that subsequent simulation results are closer to the real-time wind environment parameters.

[0051] Again, based on the updated regional building information model, a transient wind field simulation is performed in the same manner as in step S20 to obtain real-time transient wind field simulation results, wherein the real-time transient wind field simulation results include at least wind pressure distribution and turbulence distribution. Wind pressure distribution refers to the pressure distribution of real-time wind data on the surface of components in the region, such as the wind pressure in a certain area is 0.4 kPa. Turbulence distribution refers to the distribution of the degree of turbulence of the real-time airflow, such as the turbulence intensity in a certain personnel working area is 0.18. The real-time transient wind field simulation results are obtained based on real-time wind data and are closer to the real-time risk status.

[0052] Furthermore, the real-time focus area set is positionally matched with the real-time transient wind field simulation results, and risk assessment indicators for multiple real-time focus areas are extracted from the matched real-time transient wind field simulation results. These risk assessment indicators include at least turbulence intensity, turbulent kinetic energy, pressure fluctuation rate, and dynamic pressure: Turbulence intensity refers to the degree of fluctuation in airflow velocity. Higher turbulence intensity indicates more turbulent airflow, which increases the likelihood of exacerbating safety risks. Turbulent kinetic energy refers to the energy of turbulent motion. Greater turbulent kinetic energy increases the impact energy of airflow on building structures, which increases the likelihood of exacerbating safety risks. Pressure fluctuation rate refers to the amplitude of wind pressure fluctuations over time. Greater pressure fluctuation rate increases the likelihood of fatigue damage to components. Dynamic pressure refers to the amount of kinetic energy converted from airflow into pressure, directly reflecting the force of wind on objects. Thus, the risk assessment indicators characterize the impact mechanism of the wind environment on the construction area from four dimensions: airflow stability (turbulence intensity), energy destructiveness (turbulent kinetic energy), load fluctuation (pressure fluctuation rate), and direct force (dynamic pressure), providing a reliable physical parameter basis for subsequent risk quantification assessment.

[0053] Finally, a dynamic risk assessment is performed on multiple real-time focus areas based on the risk assessment indicators, and multiple real-time risk coefficients are obtained and output as a real-time risk coefficient set. For example, the real-time risk coefficients can be calculated through weighted summation, and differentiated weights (the sum of the weights is 1) are assigned to the risk assessment indicators based on the risk characteristics of different real-time focus areas. For example, for the personnel operation area, since personnel safety is directly related to airflow stability, the weights assigned to turbulence intensity, turbulent kinetic energy, pressure fluctuation rate, and dynamic pressure are 0.4, 0.1, 0.2, and 0.3, respectively. For the material storage area, since storage stability depends on wind thrust and load fluctuations, the weights assigned to turbulence intensity, turbulent kinetic energy, pressure fluctuation rate, and dynamic pressure are 0.1, 0.2, 0.3, and 0.4, respectively. Secondly, the safety threshold of the risk assessment index can be pre-set based on prior data such as construction safety specifications and historical data. By calculating the ratio of the risk assessment index to the safety threshold of the risk assessment index, the risk assessment index can be converted into a risk contribution value between 0 and 1. The higher the risk contribution value, the closer the risk is to the threshold, that is, the greater the risk. For example, the safety threshold of the risk assessment index can be set as turbulence intensity ≤ 0.2 and turbulence kinetic energy ≤ 2.0m based on prior data such as construction safety specifications and historical data. 2 / s 2 , pressure fluctuation rate ≤ 0.15kPa / s, dynamic pressure ≤ 0.6kPa, when the turbulence intensity of a real-time focus area is 0.15, and the turbulence kinetic energy is 1.8m 2 / s 2When the pressure fluctuation rate is 0.12 kPa / s and the dynamic pressure is 0.3 kPa, the risk contribution values ​​of the risk assessment indicators can be calculated as follows: 0.15 / 0.2=0.75, 1.8 / 2.0=0.9, 0.12 / 0.15=0.8, 0.3 / 0.6=0.5. If the weights of the turbulence intensity, turbulent kinetic energy, pressure fluctuation rate and dynamic pressure in the real-time focus area are 0.4, 0.1, 0.2 and 0.3 respectively, the real-time risk coefficient of the real-time focus area is calculated by weighted summation = 0.4×0.75+0.1×0.9+0.2×0.8+0.3×0.5=0.7. In this way, multiple real-time risk coefficients are calculated according to the same method and output as a real-time risk coefficient set.

[0054] Furthermore, after the step of “collecting real-time wind data, combining the real-time wind data with the numerical wind tunnel model, traversing the real-time focus area set to perform dynamic risk assessment, and obtaining a real-time risk coefficient set”, the following steps are further included: Obtaining the most unfavorable environmental convection information and the transient wind field simulation results corresponding to the real-time construction phase, and performing a priori risk assessment accordingly; Acquire a plurality of the real-time risk coefficient sets and a plurality of the real-time wind force data corresponding to the real-time construction stage; Iteratively calculating a plurality of difference wind data between the real-time wind data and the most unfavorable environmental convection information to obtain a difference wind data set; Iteratively calculating risk coefficient difference data between a plurality of the real-time risk coefficient sets and a priori risk assessment results to obtain a plurality of risk difference data sets; Taking the differential wind dataset as input and multiple risk differential datasets as supervision, a migration assessment model is constructed and trained, and dynamic risk assessment is performed according to the migration assessment model.

[0055] In an embodiment of the present application, the most unfavorable environmental convection information and transient wind field simulation results corresponding to the real-time construction stage are first obtained, and a priori risk assessment is performed accordingly. For example, based on step S20, the most unfavorable environmental convection information and transient wind field simulation results corresponding to the real-time construction stage are obtained, and the turbulence intensity, turbulent kinetic energy, pressure fluctuation rate and dynamic pressure are obtained from the transient wind field simulation results in the same manner as the aforementioned steps, and a priori risk assessment is performed to obtain a priori risk assessment result. The priori risk assessment result is a risk coefficient assuming that the real-time construction stage is in the most unfavorable environmental convection information, which can be used as a benchmark reference for comparing the difference between real-time risk and historical extreme risk.

[0056] Secondly, based on the above steps, multiple real-time wind data of the real-time construction phase are obtained, and multiple real-time risk coefficient sets are correspondingly obtained.

[0057] Next, the difference wind data between the multiple real-time wind data and the most unfavorable environmental convection information is iteratively calculated to obtain a difference wind data set. For example, if the real-time wind data has a wind speed of 8 m / s and a wind direction of 135°, and the most unfavorable environmental convection information has a historical maximum wind speed of 12 m / s and a historical dangerous wind direction of 90°, then the wind speed difference is -4 m / s and the wind direction difference is 45°, forming quantified difference wind data. In this way, the difference between the multiple real-time wind data and the most unfavorable environmental convection information is iteratively calculated to obtain a difference wind data set. The difference wind data set can reflect the deviation between the real-time wind data and the historical most unfavorable environmental convection information.

[0058] Furthermore, risk coefficient difference data between multiple real-time risk coefficient sets and prior risk assessment results are iteratively calculated to obtain multiple risk difference data sets. For example, a difference calculation is performed between the real-time risk coefficient set and the prior risk assessment result. For example, if the real-time risk coefficient is 0.7 and the prior risk assessment result is 0.9, then the difference between the real-time risk coefficient set and the prior risk assessment result = 0.7-0.9 = -0.2, which serves as the risk coefficient difference data, indicating that the real-time risk is 20% lower than the historical extreme risk. In this way, the risk coefficient difference data between multiple real-time risk coefficient sets and the prior risk assessment results are iteratively calculated to obtain multiple risk difference data sets, and quantify the deviation between the actual risk and the historical extreme risk.

[0059] Finally, with the differential wind dataset as input and multiple risk difference datasets as supervision, a migration assessment model is constructed and trained, and dynamic risk assessment is performed based on the migration assessment model. Exemplarily, the migration assessment model can be trained through the following technical paths: 1. Data preparation: Divide the differential wind dataset and multiple risk difference datasets into training set, validation set, and test set according to the ratio of 7:1.5:1.5. 2. Model construction: The model can be constructed based on the transfer learning architecture, which mainly consists of a pre-trained feature extractor, a domain adaptation layer, and a task prediction layer. The pre-trained feature extractor uses a CNN-LSTM hybrid network pre-trained on a large-scale public wind farm database (such as the building wind engineering open source dataset). The CNN module (containing 3 convolutional layers with convolution kernel sizes of 3×3, 5×5, and 3×3 respectively) is used to extract the spatial features of wind speed difference and wind direction difference. The LSTM layer (2-layer bidirectional LSTM) is used to capture the temporal correlation of wind field differences. The weight is frozen by 60% to preserve the original image quality. The general wind farm feature extraction capability is retained; the domain adaptation layer contains two fully connected layers (with 128 and 64 nodes respectively) and a domain adversarial loss module. By minimizing the distribution difference between the source domain (public dataset) and the target domain (current engineering dataset), the pre-trained features are converted into a feature space suitable for the current construction scenario. The domain adversarial loss is implemented using a gradient reversal layer (GRL), forcing the feature extractor to learn domain-invariant features; the task prediction layer consists of one fully connected layer (with 32 nodes) and an output layer. The output layer uses a linear activation function to directly predict the output risk coefficient difference data. 3. Model Training: The model uses differential wind speed data from the training set as input and differential data from multiple risk factors as supervision. The Adam optimizer is used, with an initial learning rate of 0.001 and a cosine annealing strategy for dynamic adjustment (decaying by 10% every five rounds). The loss function uses a weighted combination of mean squared error (MSE) and domain adversarial loss (weight ratio 7:3). MSE ensures that the predicted value is close to the actual risk, while domain adversarial loss improves the model's adaptability across scenarios. Training is repeated for 50 rounds, with the loss calculated on the validation set after each round. Early stopping is triggered when the validation set loss decreases by less than 1e-4 for five consecutive rounds, indicating model convergence. Finally, model performance is evaluated on the test set. Training is considered complete if the mean absolute error (MAE) is ≤0.05, and a transfer evaluation model is output. In this way, transfer learning reuses general wind farm knowledge and combines it with a domain adaptation mechanism to adapt to specific project scenarios. This improves the model's prediction accuracy for construction site-specific wind conditions while maintaining training efficiency.

[0060] In summary, compared to existing technologies, this application collects real-time wind data, combines it with the numerical wind tunnel model, and traverses the real-time set of areas of interest to perform dynamic risk assessment, thereby obtaining a set of real-time risk coefficients. This achieves dynamic quantification of current construction risks, addresses the lack of real-time performance of traditional methods, and provides a reliable quantitative basis for subsequent risk warnings.

[0061] S50: In the visualization interface of the building information model, visually mark a plurality of real-time focus areas according to the real-time risk coefficient set, and perform risk warning in combination with the visualization marking result and the real-time risk coefficient set.

[0062] The above steps obtain a real-time risk coefficient set, which provides a quantitative basis for risk warning. Based on this risk warning, a closed-loop warning mechanism from data quantification to multi-dimensional response can be formed to ensure the speed of risk response.

[0063] To address the above issues, the present application visually marks multiple real-time areas of concern according to the real-time risk coefficient set in the visualization interface of the building information model, and performs risk warnings in combination with the visualization marking results and the real-time risk coefficient set.

[0064] Specifically, step S50 in the method includes: Traversing the plurality of real-time focus areas, combining the real-time risk coefficient with a preset color mapping rule, obtaining color marking information, and rendering the corresponding information to the visualization interface to obtain the visualization marking result; Continuously monitoring the real-time risk coefficient set, and generating and sending a risk warning signal when it is determined that any of the real-time risk coefficients exceeds a preset risk threshold; The risk warning signal is used to trigger at least one of the following operations: Highlighting and flashing the corresponding real-time focus area in the visual interface; Pushing a warning message including the visual marking result and the real-time risk coefficient to a predetermined terminal device; Activate the sound and light alarm devices and electronic fences deployed in the target scene.

[0065] In an embodiment of the present application, multiple real-time areas of interest are first traversed, and the real-time risk coefficient is combined with the preset color mapping rule to obtain color marking information, and the corresponding information is rendered to the visualization interface to obtain a visualization marking result. Exemplarily, multiple real-time areas of interest are traversed, and the real-time risk coefficient is associated with the preset color mapping rule. For example, low risk (such as real-time risk coefficient ≤ 0.6) is pre-set as a green mark, medium risk (such as 0.6 < real-time risk coefficient ≤ 0.8) is pre-set as a yellow mark, and high risk (such as real-time risk coefficient > 0.8) is pre-set as a red mark. According to the color mapping rule, color marking information of multiple real-time areas of interest is generated, and it is accurately rendered into the visualization interface of the BIM model to form a visualization marking result. In this way, the risk level can be intuitively distinguished by color, which improves the speed of locating high-risk areas.

[0066] Secondly, the real-time risk factor set is continuously monitored. When any real-time risk factor exceeds a preset risk threshold, a risk warning signal is generated and sent. The preset risk threshold can be differentiated based on construction specifications and regional characteristics. For example, a threshold of 0.7 (more stringent) can be set for areas with large gatherings of people, while a threshold of 0.8 can be set for areas with material storage, ensuring targeted warnings. The risk warning signal is used to trigger at least one of the following actions: highlighting the corresponding real-time area of ​​concern in the visual interface; pushing a warning message containing the visual marking results and the real-time risk factor to a predetermined terminal device; and activating the audio and visual alarm devices and electronic fences deployed in the target scene.

[0067] For example, the dynamic changes of the real-time risk coefficient set are continuously tracked, such as updating the data every 10 seconds. When the real-time risk coefficient of any area exceeds the preset risk threshold (such as 0.8), a risk warning signal is generated and sent. The warning signal is responded to through multiple channels to ensure that the risk information reaches relevant personnel efficiently, including: in the BIM visualization interface, the area that triggers the warning is flashing red and the boundary is enlarged; the warning message is pushed to the mobile phone APP, walkie-talkie and other terminal devices of the construction manager, safety officer, etc.; and the sound and light alarm devices around the high-risk area are triggered to sound an alarm.

[0068] In summary, compared to existing technologies, this application visually marks multiple real-time areas of concern based on the real-time risk coefficient set in the visualization interface of the building information model, and then combines the visual marking results with the real-time risk coefficient set to issue risk warnings. This ensures that risk information reaches the entire chain from management to operation through color visualization and multiple risk warning methods, significantly shortening risk response time.

[0069] In summary, the embodiments of the present application have at least the following technical effects: Compared to existing technologies, this application first obtains the building information model and historical meteorological data of the target scene. This overcomes the dual limitations of traditional risk assessments: fuzzy structural information and missing environmental parameters, providing a data foundation for subsequent dynamic risk simulation and early warning.

[0070] Secondly, this application constructs a numerical wind tunnel model based on the building information model and historical meteorological data, performs transient wind field simulations, and combines these transient wind field simulation results to identify regions and obtain a priori sets of areas of concern. In this way, using historical data and simulation technology, a list of high-risk areas that may exist during each construction phase is identified, potential risk points are identified in advance, and provide reliable support for subsequent risk assessments and risk warnings.

[0071] Thirdly, this application defines a real-time set of focus areas for the target scenario based on the target scenario's construction progress information and the aforementioned set of focus areas. This ensures that all high-risk points at the current stage are included in the monitoring scope, providing precise targeting for dynamic risk assessment.

[0072] Furthermore, this application collects real-time wind data and combines it with the numerical wind tunnel model to perform dynamic risk assessment across the real-time set of areas of interest, obtaining a set of real-time risk coefficients. This enables dynamic quantification of current construction risks, addresses the lack of real-time performance of traditional methods, and provides a reliable quantitative basis for subsequent risk warnings.

[0073] Finally, within the visualization interface of the building information model, this application visually marks multiple real-time areas of concern based on the real-time risk coefficient set, and then uses this visual marking and the real-time risk coefficient set to issue risk warnings. This ensures that risk information reaches the entire chain from management to operations through color visualization and multiple risk warning methods, significantly shortening risk response time.

[0074] Through the above technical solution, this application provides a structured three-dimensional structural benchmark and quantitative environmental parameter samples for risk assessment by obtaining the BIM model and historical meteorological data of the target scene, solving the problems of fuzzy structural information and missing environmental data in traditional methods; constructing a numerical wind tunnel model and conducting transient wind field simulation to identify a priori focus area sets, achieving accurate positioning of historical high-risk areas, overcoming the limitations of traditional assessments that rely on experience and lack systematicity; dynamically defining real-time focus area sets through construction progress information, allowing the risk monitoring range to be dynamically adjusted with the construction stage, avoiding the dislocation of monitoring dynamic scenes; combining real-time wind data with the numerical wind tunnel model for dynamic risk assessment, outputting a real-time risk coefficient set, achieving real-time quantitative update of risks, and solving the problem of early warning lag; finally, completing risk marking and multi-channel early warning in the BIM visual interface, achieving closed-loop management from risk identification to disposal. In this way, the efficiency of identifying construction risks and the timeliness of early warning are improved, providing intelligent technical support for construction safety management.

[0075] Example 2, as Figure 2 As shown, based on the same inventive concept as the method for intelligent identification and early warning of construction risks based on a BIM model provided in Example 1, an embodiment of the present invention also provides an intelligent identification and early warning system for construction risks based on a BIM model, comprising: Data acquisition module 11, used to obtain the building information model and historical meteorological data of the target scene; A region identification module 12 is configured to construct a numerical wind tunnel model based on the building information model and historical meteorological data, perform transient wind field simulation, and perform region identification based on the transient wind field simulation results to obtain a priori focus region set; The focus area definition module 13 is used to define a real-time focus area set of the target scene based on the construction progress information of the target scene and the prior focus area set; The risk assessment module 14 is configured to collect real-time wind data, and combine the real-time wind data with the numerical wind tunnel model to traverse the real-time focus area set for dynamic risk assessment to obtain a real-time risk coefficient set; The risk warning module 15 is used to visually mark multiple real-time focus areas according to the real-time risk coefficient set in the visualization interface of the building information model, and to perform risk warning in combination with the visual marking results and the real-time risk coefficient set.

[0076] The data acquisition module 11 is specifically configured to: Obtain construction implementation planning information for the target scenario and extract multiple construction phases; Marking the building information model in stages according to the plurality of construction stages to obtain a model marking result, wherein the model marking result includes a plurality of groups of stage marks and associated stage model boundaries; The building information model and the model marking result are stored in association.

[0077] The region identification module 12 is specifically configured to: Randomly selecting the building information model according to the model marking result to obtain a regional building information model; In combination with a preset time window constraint, the historical meteorological data is parsed to obtain the most unfavorable environmental convection information, wherein the most unfavorable environmental convection information includes at least a wind speed index and a wind direction index; Based on the regional building information model and the most unfavorable environmental convection information, construct the numerical wind tunnel model, and perform a transient wind field simulation accordingly to obtain the transient wind field simulation results, wherein the transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Traversing the model marking results, iteratively selecting the regional building information model to perform transient wind field simulation; Dual-target threshold screening is performed based on the multiple transient wind field simulation results to determine multiple prior regions of interest, obtain the prior region of interest set, and establish a region-stage association relationship between the prior region of interest set and the model labeling result.

[0078] The region of interest definition module 13 is specifically configured to: Determine the real-time construction stage based on the construction progress information and extract the corresponding stage engineering quantity information; Analyze the phased engineering quantity information to determine multiple phased focus areas of the target scene, wherein the phased focus areas at least include a material storage area and a personnel gathering area; Based on the real-time construction stage, the a priori focus area set is called in stages in combination with the area-stage association relationship to obtain a stage-a priori focus area set; A plurality of the stage regions of interest are combined with the stage-prior region of interest set to obtain the real-time region of interest set.

[0079] The risk assessment module 14 is specifically configured to: Based on the real-time construction stage, extracting the regional building information model accordingly; Analyzing the real-time wind data, extracting real-time wind speed and real-time wind direction, and updating the regional building information model according to the real-time wind speed and the real-time wind direction; Performing a transient wind field simulation based on the updated regional building information model to obtain real-time transient wind field simulation results, wherein the real-time transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Position matching is performed on the real-time focus area set and the real-time transient wind field simulation result, and risk assessment indicators of multiple real-time focus areas are extracted, wherein the risk assessment indicators include at least turbulence intensity, turbulence kinetic energy, pressure fluctuation rate and dynamic pressure; Dynamic risk assessment of the plurality of real-time focus areas is performed according to the risk assessment index, and a plurality of real-time risk coefficients are obtained and output as the real-time risk coefficient set.

[0080] Furthermore, a real-time risk coefficient set is obtained, and then: Obtaining the most unfavorable environmental convection information and the transient wind field simulation results corresponding to the real-time construction phase, and performing a priori risk assessment accordingly; Acquire a plurality of the real-time risk coefficient sets and a plurality of the real-time wind force data corresponding to the real-time construction stage; Iteratively calculating a plurality of difference wind data between the real-time wind data and the most unfavorable environmental convection information to obtain a difference wind data set; Iteratively calculating risk coefficient difference data between a plurality of the real-time risk coefficient sets and a priori risk assessment results to obtain a plurality of risk difference data sets; Taking the differential wind dataset as input and multiple risk differential datasets as supervision, a migration assessment model is constructed and trained, and dynamic risk assessment is performed according to the migration assessment model.

[0081] The risk warning module 15 is specifically used to: Traversing the plurality of real-time focus areas, combining the real-time risk coefficient with a preset color mapping rule, obtaining color marking information, and rendering the corresponding information to the visualization interface to obtain the visualization marking result; Continuously monitoring the real-time risk coefficient set, and generating and sending a risk warning signal when it is determined that any of the real-time risk coefficients exceeds a preset risk threshold; The risk warning signal is used to trigger at least one of the following operations: Highlighting and flashing the corresponding real-time focus area in the visual interface; Pushing a warning message including the visual marking result and the real-time risk coefficient to a predetermined terminal device; Activate the sound and light alarm devices and electronic fences deployed in the target scene.

[0082] In summary, the embodiments of the present application have at least the following technical effects: Compared with the existing technology, this application first obtains the building information model and historical meteorological data of the target scene through the data acquisition module, which solves the dual limitations of fuzzy structural information and missing environmental parameters in traditional risk assessment, and provides a data basis for subsequent dynamic risk simulation and early warning. Secondly, through the area identification module, a numerical wind tunnel model is constructed based on the building information model and historical meteorological data to perform transient wind field simulation, and regional identification is performed in combination with the transient wind field simulation results to obtain a priori focus area set. Using historical data and simulation technology, a list of high-risk areas that may exist in each construction stage is determined, and potential risk points are locked in advance, providing reliable support for subsequent risk assessment and risk early warning. Thirdly, through the focus area definition module, the real-time focus area set of the target scene is defined based on the construction progress information of the target scene and the prior focus area set, ensuring that high-risk points in the current stage are included in the monitoring range, providing precise targeting for dynamic risk assessment. Furthermore, through the risk assessment module, real-time wind data is collected, and combined with the real-time wind data and the numerical wind tunnel model, the real-time focus area set is traversed for dynamic risk assessment, and the real-time risk coefficient set is obtained, which realizes the dynamic quantification of the current construction risk, solves the problem of insufficient real-time performance of traditional methods, and provides a reliable quantitative basis for subsequent risk warnings. Finally, through the risk warning module, in the visualization interface of the building information model, multiple real-time focus areas are visually marked according to the real-time risk coefficient set, and risk warnings are carried out in combination with the visualization marking results and the real-time risk coefficient set. Through color visualization and a variety of risk warning methods, it is ensured that risk information reaches the entire chain from the management end to the operation end, greatly shortening the risk response time. In this way, the efficiency of identifying construction risks and the timeliness of warnings are improved, providing intelligent technical support for construction safety management.

[0083] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0084] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0088] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0089] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The intelligent identification and early warning method for construction risks based on BIM model is characterized by: include: Obtain the building information model and historical meteorological data of the target scene; Constructing a numerical wind tunnel model based on the building information model and historical meteorological data, performing transient wind field simulation, and performing region identification based on the transient wind field simulation results to obtain a priori focus region set; Defining a real-time focus area set of the target scene based on the construction progress information of the target scene and the prior focus area set; Collecting real-time wind data, and combining the real-time wind data with the numerical wind tunnel model, traversing the real-time focus area set to perform dynamic risk assessment and obtain a real-time risk coefficient set; In the visualization interface of the building information model, a plurality of real-time focus areas are visually marked according to the real-time risk coefficient set, and risk warning is performed in combination with the visualization marking result and the real-time risk coefficient set.

2. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 1, characterized in that: Obtain the building information model and historical meteorological data of the target scene, and then: Obtain construction implementation planning information for the target scenario and extract multiple construction phases; Marking the building information model in stages according to the plurality of construction stages to obtain a model marking result, wherein the model marking result includes a plurality of groups of stage marks and associated stage model boundaries; The building information model and the model marking result are stored in association.

3. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 2, characterized in that: A numerical wind tunnel model is constructed based on the building information model and historical meteorological data to perform transient wind field simulation. The transient wind field simulation results are combined to perform region identification and obtain a priori focus region set, including: Randomly selecting the building information model according to the model marking result to obtain a regional building information model; In combination with a preset time window constraint, the historical meteorological data is parsed to obtain the most unfavorable environmental convection information, wherein the most unfavorable environmental convection information includes at least a wind speed index and a wind direction index; Based on the regional building information model and the most unfavorable environmental convection information, construct the numerical wind tunnel model, and perform a transient wind field simulation accordingly to obtain the transient wind field simulation results, wherein the transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Traversing the model marking results, iteratively selecting the regional building information model to perform transient wind field simulation; Dual-target threshold screening is performed based on the multiple transient wind field simulation results to determine multiple prior regions of interest, obtain the prior region of interest set, and establish a region-stage association relationship between the prior region of interest set and the model labeling result.

4. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 3 is characterized in that: Based on the construction progress information of the target scene and the a priori set of focus areas, a real-time focus area set of the target scene is defined, including: Determine the real-time construction stage based on the construction progress information and extract the corresponding stage engineering quantity information; Analyze the phased engineering quantity information to determine multiple phased focus areas of the target scene, wherein the phased focus areas at least include a material storage area and a personnel gathering area; Based on the real-time construction stage, the a priori focus area set is called in stages in combination with the area-stage association relationship to obtain a stage-a priori focus area set; A plurality of the stage regions of interest are combined with the stage-prior region of interest set to obtain the real-time region of interest set.

5. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 4 is characterized in that: Collecting real-time wind data, combining the real-time wind data with the numerical wind tunnel model, traversing the real-time focus area set to perform dynamic risk assessment, and obtaining a real-time risk coefficient set, further comprising: Based on the real-time construction stage, extracting the regional building information model accordingly; Analyzing the real-time wind data, extracting real-time wind speed and real-time wind direction, and updating the regional building information model according to the real-time wind speed and the real-time wind direction; Performing a transient wind field simulation based on the updated regional building information model to obtain real-time transient wind field simulation results, wherein the real-time transient wind field simulation results at least include wind pressure distribution and turbulence distribution; Position matching is performed on the real-time focus area set and the real-time transient wind field simulation result, and risk assessment indicators of multiple real-time focus areas are extracted, wherein the risk assessment indicators include at least turbulence intensity, turbulence kinetic energy, pressure fluctuation rate and dynamic pressure; Dynamic risk assessment of the plurality of real-time focus areas is performed according to the risk assessment index, and a plurality of real-time risk coefficients are obtained and output as the real-time risk coefficient set.

6. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 5, characterized in that: Real-time wind data is collected, and the real-time wind data is combined with the numerical wind tunnel model to traverse the real-time focus area set for dynamic risk assessment to obtain a real-time risk coefficient set. Thereafter, the method further includes: Obtaining the most unfavorable environmental convection information and the transient wind field simulation results corresponding to the real-time construction phase, and performing a priori risk assessment accordingly; Acquire a plurality of the real-time risk coefficient sets and a plurality of the real-time wind force data corresponding to the real-time construction stage; Iteratively calculating a plurality of difference wind data between the real-time wind data and the most unfavorable environmental convection information to obtain a difference wind data set; Iteratively calculating risk coefficient difference data between a plurality of the real-time risk coefficient sets and a priori risk assessment results to obtain a plurality of risk difference data sets; Taking the differential wind dataset as input and multiple risk differential datasets as supervision, a migration assessment model is constructed and trained, and dynamic risk assessment is performed according to the migration assessment model.

7. The method for intelligent identification and early warning of construction risks based on the BIM model according to claim 5, characterized in that: In the visualization interface of the building information model, visually marking a plurality of real-time areas of concern according to the real-time risk coefficient set, and performing risk warning in combination with the visual marking results and the real-time risk coefficient set, including: Traversing the plurality of real-time focus areas, combining the real-time risk coefficient with a preset color mapping rule, obtaining color marking information, and rendering the corresponding information to the visualization interface to obtain the visualization marking result; Continuously monitoring the real-time risk coefficient set, and generating and sending a risk warning signal when it is determined that any of the real-time risk coefficients exceeds a preset risk threshold; The risk warning signal is used to trigger at least one of the following operations: Highlighting and flashing the corresponding real-time focus area in the visual interface; Pushing a warning message including the visual marking result and the real-time risk coefficient to a predetermined terminal device; Activate the sound and light alarm devices and electronic fences deployed in the target scene.

8. The intelligent identification and early warning system for construction risks based on the BIM model is characterized by: Used to perform the method according to any one of claims 1 to 7, comprising: Data acquisition module, used to obtain the building information model and historical meteorological data of the target scene; A region identification module is used to construct a numerical wind tunnel model based on the building information model and historical meteorological data, perform transient wind field simulation, and perform region identification based on the transient wind field simulation results to obtain a priori focus region set; An area of ​​interest definition module, configured to define a real-time area of ​​interest set of the target scene based on the construction progress information of the target scene and the a priori area of ​​interest set; a risk assessment module, configured to collect real-time wind data, and combine the real-time wind data with the numerical wind tunnel model to traverse the real-time focus area set for dynamic risk assessment to obtain a real-time risk coefficient set; The risk warning module is used to visually mark multiple real-time focus areas according to the real-time risk coefficient set in the visualization interface of the building information model, and to perform risk warning in combination with the visualization marking results and the real-time risk coefficient set.

Images