Engineering safety management method and system based on BIM (Building Information Modeling) and medium

Abstract

The invention provides a BIM-based engineering safety management method and system and a medium, and relates to the technical field of engineering management, and the method comprises the steps: collecting a first construction event, transmitting the first construction event back to a BIM engineering model, determining a first feature set, carrying out risk control probability verification, and triggering a first safety management instruction; activating the intermediate plug-in, executing hidden risk mining and dynamic rule generation, and performing local lightweight reconstruction on the BIM engineering model; and executing the deduction and rule constraint decision under the multivariate combination, generating a first safety rule and a first risk control plan, and executing engineering safety management based on the first construction event. Through the method and the device, the technical problems of safety management lagging and low accuracy caused by difficulty in coping with dynamic changes and hidden risks of a construction environment in a traditional method in the prior art are solved, short-time simulation analysis of the BIM engineering model is performed by taking the construction event as a management mode, and the response speed and the safety management efficiency of a construction site are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering management, and particularly relates to a BIM-based engineering safety management method and system and a medium. BACKGROUND

[0002] At present, the traditional engineering safety management method usually relies on a pre-constructed model and preset rules. However, the construction site situation changes rapidly, and the static model is difficult to reflect these changes in real time, resulting in lagging risk identification. In the traditional method, safety risks are mostly identified based on explicit factors such as design conflicts or structural problems, but the implicit risks in the construction process are often ignored. The implicit risks not only come from factors such as temporary changes in working conditions, equipment displacement, personnel activities and the like in the construction process, but also come from complex multi-factor coupling.

[0003] Although the BIM technology improves the visualization and coordination capability of safety management in engineering management, the current BIM application mode still mainly focuses on static model review and matching based on fixed rules. This way cannot effectively cope with the complex environment of dynamic changes in the construction site. In particular, in the management of uncertain factors such as temporary working conditions, equipment displacement and personnel activities, the traditional BIM model cannot capture and reflect these changes in real time, resulting in significant lag and blind spots in the identification of safety hazards.

[0004] In summary, in the prior art, there is a technical problem that the traditional engineering safety management method relies on periodic inspection and simple rule matching, which is difficult to cope with the dynamic changes of the construction environment and the implicit risks, resulting in lagging safety management and low precision. SUMMARY

[0005] The purpose of the present application is to provide a BIM-based engineering safety management method, system and medium, which solves the technical problem in the prior art that the traditional engineering safety management method relies on periodic inspection and simple rule matching, which is difficult to cope with the dynamic changes of the construction environment and the implicit risks, resulting in lagging safety management and low precision.

[0006] In view of the above problems, the present application provides a BIM-based engineering safety management method, system and medium.

[0007] In a first aspect, the application provides a BIM-based engineering safety management method, which is implemented by a BIM-based engineering safety management system, wherein the BIM-based engineering safety management method comprises: collecting a first construction event, returning to a BIM engineering model established by a safety management platform, determining a first feature set and performing a risk control probability test based on a rigid condition, and triggering a first safety management instruction; according to the first safety management instruction, activating an intermediate plug-in, performing implicit risk mining and dynamic rule generation, performing local lightweight reconstruction on the BIM engineering model, and determining a first event BIM model; according to the first event BIM model, taking the first feature set and the first implicit event risk as the decision orientation, taking the first event rule as the decision rule, performing deduction and rule constraint decision under multi-element combination, generating the first safety rule and the first risk control plan, and performing engineering safety management based on the first construction event.

[0008] Optionally, connect to an engineering database to retrieve engineering construction records; according to the engineering construction records, mine a construction condition sequence, wherein the construction condition sequence contains engineering features-risk probability satisfying a preset generalization degree; set a condition probability threshold as a rigid condition for the construction condition sequence, and deploy a priori port in the BIM engineering model.

[0009] Optionally, according to the first construction event, perform short-time simulation on the BIM engineering model to determine a first feature set, wherein the first feature set contains event features and dynamic features; trigger the a priori port to perform rigid condition matching and out-of-limit determination on the first feature set, and if there is any feature with probability out of limit, generate the first safety management instruction; if there is no probability out of limit, generate a standard safety management instruction.

[0010] Optionally, according to the first construction event, construct a first event relationship graph, wherein the event relationship graph takes engineering components as nodes and spatial relationships between components as edges, and the nodes and edges are identified with attribute elements; according to the first event relationship graph, a first generator of the intermediate plug-in performs implicit risk generation to determine a first implicit event risk, wherein the first implicit event risk is determined based on multi-perspective implicit risk fusion; according to a second generator of the intermediate plug-in, generate a first event rule, wherein the first event rule is determined based on multi-perspective rule fusion.

[0011] Optionally, establish a multi-perspective experience library, wherein different construction roles are defined as perspectives; the first generator performs multi-perspective thread reasoning by interacting with the multi-perspective experience library to determine multi-perspective implicit risks; and the multi-perspective implicit risks are fused as the first implicit event risk.

[0012] Optionally, the BIM engineering model is event model segmented, and a lightweight event BIM model is determined according to lightweight reconstruction based on the first event relationship diagram; the lightweight event BIM model is mode deployed as a first event BIM model according to a first feature set and a first implicit event risk as decision guidance, and a first event rule as a decision rule.

[0013] Optionally, the decision guidance and the decision rule are combined through mutual matching to determine a plurality of combination pairs; and the first event BIM model is deduced and rule-constrained decision is made for the plurality of combination pairs, and the decision set is added, wherein the decision set includes single combination pair deduction and joint deduction of at least two combination pairs; and the first security rule and the first risk control plan are generated according to the decision set and added to the first event partition of the platform security management partition.

[0014] Optionally, temporary communication interaction is established between the first event partition and the front-end acquisition array; and according to the construction process of the first construction event, monitoring based on the front-end acquisition array is performed through space targeting, safety management is performed according to the first security rule in the first event partition, and emergency management is performed according to the first risk plan when there is a risk trend exceeding the first security rule.

[0015] In a second aspect, the application further provides a BIM-based engineering safety management system for executing the BIM-based engineering safety management method of the first aspect, wherein the BIM-based engineering safety management system comprises: a risk control probability determination module for collecting a first construction event, returning to a BIM engineering model established by a safety management platform, determining a first feature set, and performing risk control probability determination based on a rigid condition to trigger a first safety management instruction; a local lightweight reconstruction module for activating an intermediate plug-in according to the first safety management instruction, performing implicit risk mining and dynamic rule generation, and performing local lightweight reconstruction on the BIM engineering model to determine a first event BIM model; and an engineering safety management module for performing deduction and rule-constrained decision under multiple combinations according to the first event BIM model, generating a first security rule and a first risk control plan, and performing engineering safety management based on the first construction event according to a first feature set and a first implicit event risk as decision guidance, and a first event rule as a decision rule.

[0016] In a third aspect, a computer readable storage medium has a computer program stored thereon, and the computer program, when executed, implements the steps of the BIM-based engineering safety management method of any one of the first aspect.

[0017] The one or more technical solutions provided in the application have at least the following beneficial effects: by collecting a first construction event, returning to a BIM engineering model established by a safety management platform, determining a first feature set and performing a risk control probability test based on rigid conditions, triggering a first safety management instruction; according to the first safety management instruction, activating an intermediate plug-in, performing implicit risk mining and dynamic rule generation, performing local lightweight reconstruction on the BIM engineering model, determining a first event BIM model; according to the first event BIM model, taking the first feature set and the first implicit event risk as the decision orientation, taking the first event rule as the decision rule, performing deduction and rule constraint decision under multi-element combination, generating the first safety rule and the first risk control plan, and performing engineering safety management based on the first construction event. That is, by taking the construction event as the management mode, short-time simulation analysis is performed on the BIM engineering model, when the risk control exceeds the limit, the optimal safety rule and the risk control plan are dynamically generated, so that the rapid identification and accurate response to potential risks are realized, and the response speed and safety management efficiency of the construction site are greatly improved.

[0018] The above description is only a summary of the technical solutions of the application. In order to more clearly understand the technical means of the application, the specific embodiments of the application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and those skilled in the art can obtain other drawings without creative labor on the basis of the provided drawings.

[0020] Figure 1 The flowchart of the BIM-based engineering safety management method of the application.

[0021] Figure 2 The structural schematic diagram of the BIM-based engineering safety management system of the application.

[0022] Explanation of reference numerals: risk control probability test module 11, local lightweight reconstruction module 12, engineering safety management module 13. DETAILED DESCRIPTION

[0023] The application provides a BIM-based engineering safety management method, system and medium, which solves the technical problem that the existing technology is difficult to cope with the dynamic changes of the construction environment and the implicit risks due to the dependence of the traditional engineering safety management method on periodic inspection and simple rule matching, resulting in lagging safety management and low precision. By taking the construction event as the management mode, short-time simulation analysis of the BIM engineering model is performed, and when the risk control exceeds the limit, the optimal safety rule and risk control plan are dynamically generated, thereby realizing rapid identification and accurate response to potential risks, greatly improving the response speed and safety management efficiency of the construction site.

[0024] The technical solutions in the application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. It should be understood that the application is not limited by the example embodiments described herein. Based on the embodiments of the application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the application. In addition, it should be noted that, for the convenience of description, only parts related to the application are shown in the drawings, not all.

[0025] Embodiment one, please refer to the attached Figure 1 The application provides a BIM-based engineering safety management method, wherein the BIM-based engineering safety management method is executed by a BIM-based engineering safety management system, and the BIM-based engineering safety management method specifically includes the following steps: A first construction event is collected, returned to a BIM engineering model established by a safety management platform, a first feature set is determined, and a risk control probability test based on rigid conditions is performed, triggering a first safety management instruction.

[0026] Further, the application further includes the following steps: connecting an engineering database, and calling engineering construction records; according to the engineering construction records, construction condition sequences are mined, wherein the construction condition sequences contain engineering features-risk probabilities meeting a preset generalization degree; for the construction condition sequences, a condition probability threshold is set as a rigid condition, and a priori port is deployed in the BIM engineering model.

[0027] Specifically, the engineering database is a structured database that centrally stores and manages project engineering related data, which often contains data of the whole process of the engineering project from planning, design to construction and maintenance, including construction logs, test reports, accident records, material inspection data, environmental monitoring data, equipment operation logs, etc. The engineering database is connected through a data interface, and all relevant historical engineering construction records are called. The engineering construction records refer to detailed documents recorded by construction personnel or project management personnel during the construction process, related to construction progress, technology, equipment usage, safety management, etc.

[0028] The engineering construction records are mined by applying data mining algorithms to extract the construction condition sequences. Each construction stage is composed of a set of condition data, including construction environment, material configuration, personnel arrangement, etc. The engineering construction records are filtered and matched according to the preset generalization degree, so as to obtain the construction condition sequences. The construction condition sequence refers to the condition-result combination rule generated by data mining, which represents the probability of occurrence of certain risks under specific conditions, not only revealing the potential risks under different conditions, but also giving a quantitative risk probability. Specifically, the engineering construction records are processed by the data cleaning and preprocessing module, such as processing missing values and unifying data formats. The association rule mining algorithm is used to scan all records to find those frequently appearing engineering feature combinations. According to the preset generalization degree requirement, for each frequently appearing feature combination, the corresponding risk probability is calculated to obtain the engineering feature-risk probability that meets the preset generalization degree, which constitutes the construction condition sequence. The definition of the preset generalization degree emphasizes the generalization degree of the rule, which enhances the generality of the rule by abstracting and promoting the specific values in the data, avoids the over-specific rule being too limited, and thus improves its application value in various construction environments. For example, assuming that a certain construction condition sequence is: the work type is steel box girder hoisting, the wind speed is high wind (7.9-13.8 m / s), and the height from the ground is high (more than 20 m), this combination has appeared 1000 times in the historical data, of which 700 times have occurred in the dangerous situations such as swinging of the hoisted object. Through calculation, the risk probability of this condition combination is 70%.

[0029] For the sequence of construction conditions, the condition probability threshold is a risk control standard used to determine which construction condition combinations require special attention. When the risk probability of a combination exceeds the threshold, the condition combination is considered a high-risk condition and requires appropriate safety measures. The condition probability threshold is input as a rigid condition into the prior port of the BIM engineering model. The BIM engineering model matches the rigid condition to determine which construction stages, work types, or environmental conditions may lead to high risks. For example, a large construction project's construction record contains data from 15 projects over the past 10 years, totaling 450,000 records. All records of hoisting operations, totaling 87,000 records, were extracted, and a generalization rule was set: wind speed 0-5.4 m / s is light wind, wind speed 5.4-7.9 m / s is moderate wind, and wind speed 7.9-13.8 m / s is strong wind; height 0-10 m is low, 10-20 m is medium, and above 20 m is high. An important risk rule was discovered: the work type is steel box girder hoisting, the wind speed is strong wind (7.9-13.8 m / s), and the height is high (more than 20 m). This combination appeared 1250 times in the historical data, of which 900 times caused the hoisted object to sway or the work to stop, resulting in a risk probability of 0.72. According to historical data and safety standards, the condition probability threshold is set to 70%.

[0030] The prior port refers to a pre-set data access point in the BIM engineering model for importing relevant data for analysis and processing. The BIM engineering model is a digital tool that allows integration of all data for a construction project through three-dimensional modeling techniques, enabling visualization of various aspects of the building, including structure, facilities, construction progress, and other information. Rigid conditions refer to conditions that must be met during construction and cannot be compromised for safety. By connecting to the engineering database and extracting construction records, combined with the BIM engineering model for dynamic display and risk assessment, real-time monitoring and identification of potential risks during construction can greatly improve the safety of the construction site.

[0031] Further, the application further includes the following steps: based on the first construction event, performing short-time simulation on the BIM engineering model to determine a first feature set, wherein the first feature set includes event features and dynamic features; triggering the prior port to perform rigid condition matching and out-of-limit determination on the first feature set, and if there is any feature with an out-of-limit probability, generating the first safety management instruction; if there is no out-of-limit probability, generating a standard safety management instruction.

[0032] Specifically, the first construction event occurring at the construction site is collected and uploaded to the safety management platform through the data collection module, and then the data is transmitted to the BIM engineering model. The first construction event is a specific work activity that is occurring or about to occur at the construction site, and is a trigger, not a time period. Based on the first construction event, a short-time simulation is performed in the BIM engineering model to simulate the risk situation during the construction process, and the physical effects that may occur under the first construction event are calculated, such as component swing under wind load, foundation pressure change, etc., thereby outputting a first feature set containing the event itself and the real-time environmental state. The first feature set includes event features and dynamic features. The event features are static attributes that describe the event itself, such as the work type being large component hoisting, the component weight being 45 tons, etc. The dynamic features are parameters or conditions that change constantly during the construction process, such as wind speed changes, temperature fluctuations, equipment load changes, etc.

[0033] The first feature set is quickly matched with the built-in rigid conditions by triggering the prior port of the BIM engineering model, and it is determined whether there is a risk control overrun. If the risk probability of a certain feature exceeds the set safety threshold, an overrun determination is made and the first safety management instruction is triggered. For example, if the weight > 40t and the height > 50m and the wind speed > 8m / s, the risk probability is 89%, the current feature value (45t, 60m, 10.5m / s) is substituted into the rule for determination, 45t > 40t, 65m > 50m, 10.5m / s > 8m / s, it is found that all conditions are met, and the calculated probability far exceeds the threshold, i.e. there is a probability overrun, and the first safety management instruction is immediately generated, the content is: high-risk warning! Immediately suspend hoisting operations! Reason: Overweight component hoisting at high altitude in strong wind environment risk overrun, please wait for the wind to improve or start the standby risk control plan. The rigid condition matching refers to determining whether it meets the safety standard under certain conditions. If the risk probability of a certain condition exceeds the preset threshold, an overrun determination is triggered, i.e. it is determined whether there is a risk control overrun. If a risk control overrun or potential risk is detected, the first safety management instruction is generated, prompting the site to take appropriate safety measures, such as suspending work, adding support or adjusting the work plan, etc. The risk control overrun refers to the probability of a certain risk condition exceeding the preset safety threshold, and exceeding the threshold means that there is a high safety risk, which must be addressed.

[0034] If no risk control overrun is detected, a standard safety management instruction is generated, i.e. the work continues according to the standard safety process and no additional intervention is required. Through short-time simulation and real-time data analysis, high-risk conditions that may exist during the construction process are identified in a timely manner, and appropriate safety management instructions are generated to ensure the safety of the construction process.

[0035] According to the first safety management instruction, an intermediate plug-in is activated, implicit risk mining and dynamic rule generation are performed, and a local lightweight reconstruction of the BIM engineering model is performed to determine a first event BIM model.

[0036] Further, the application further includes the following steps: constructing a first event relationship graph according to the first construction event, wherein the event relationship graph takes engineering components as nodes and spatial relationships between components as edges, and the nodes and edges are identified with attribute elements; the first generator of the intermediate plug-in performs implicit risk generation according to the first event relationship graph to determine a first implicit event risk, wherein the first implicit event risk is determined based on multi-perspective implicit risk fusion; and the second generator of the intermediate plug-in generates a first event rule, wherein the first event rule is determined based on multi-perspective rule fusion.

[0037] Specifically, after receiving the first safety management instruction, the intermediate plug-in is automatically activated. The intermediate plug-in is a module in the safety management platform, which is used to perform specific functions such as risk identification, rule generation, etc. According to the first construction event, a first event relationship graph is constructed. Each component, such as a beam, column, plate, pipeline, etc., is regarded as a node in the graph; the spatial relationship between components, such as support, connection, adjacency, crossing, etc., is defined as the edge of the graph; further, attributes such as component type, material, mechanical property, distance, angle, etc. are assigned to the nodes and edges, which are converted into a logical and intuitive form, completing the construction of the first event relationship graph.

[0038] The adversarial network is a deep learning architecture, which is usually used to generate data based on the adversarial training between the generator and the discriminator. The generator is responsible for generating data, and the discriminator is responsible for judging whether the generated data is real, that is, for the first generator, the generator is used to generate implicit risks, and the discriminator is used to judge whether the generated implicit risks are in line with the actual situation. The goal of the two is to continuously improve the quality of the generated data through mutual competition. The generator receives input information such as the first event relationship graph, construction environment data, equipment state, etc., generates implicit risks, and the input information is processed through certain embedding, then sent to the network of the generator, and finally outputs the implicit risk result. The discriminator is responsible for evaluating whether the generated implicit risk is reasonable and whether it conforms to the actual construction site risk, and judges whether the generated implicit risk has credibility and usability through training historical data and known real implicit risk cases. The first event relationship graph includes data of each node and its spatial relationship in the construction process, and each node and edge has attribute information. In order to train the discriminator, a set of implicit risk data of historical construction projects needs to be prepared, including the implicit risks that have occurred and their actual occurrence probability, such as equipment failure, hoisting out of control, unstable construction environment, etc. In the training process, real-time environment data, equipment data, personnel data, etc. of the construction site also need to be collected.

[0039] In the adversarial network, the training objectives of the generator and the discriminator are opposite, and the corresponding loss functions are defined. The goal of the generator is to deceive the discriminator, i.e., the generated data looks as real as possible. Fix the generator, train the discriminator, extract a batch of multiple real samples from the training set, generate multiple generated risk vectors from random noise by the generator, merge the multiple real risk vectors and the multiple generated risk vectors, and label them as 1 for real and 0 for generated, and input them to the discriminator. Through the known binary cross-entropy loss function, the loss of the discriminator is calculated, and only the parameters of the discriminator are updated through backpropagation, and the goal is to minimize the loss, i.e., to make the discriminator accurately distinguish between true and false.

[0040] Fix the discriminator, train the generator, and again generate multiple generated risk vectors by the generator, input the generated risk vectors to the discriminator with fixed parameters, get the discrimination result A, and calculate the loss function of the generator: LOSS = -log(D(A)), the meaning of this loss function is: the generator hopes that the risk limit A generated by it is judged as real by the discriminator, i.e., the probability of D(A) approaching 1 is maximized. Through backpropagation, only the parameters of the generator are updated, and the goal is to minimize the loss. The training data amount is 50000 historical events and their risk pairs; the batch size is 32; the feature vector dimension: the dimensions of the graph vector and the risk vector are both set to 512; the optimizer is the Adam optimizer, the learning rate is 0.0002; the number of training rounds is 1000.

[0041] The discriminator first looks at 32 real risks, such as brittle fracture of the connection weld at low temperature, and 32 false risks generated by the generator, such as fatigue damage caused by wind vibration. After learning, the discriminator can accurately distinguish that the first description is real (label 1) and the second is false (label 0). Then, the generator tries to improve and generate a more realistic risk, such as: under the condition of 5℃ ambient temperature and dynamic load, the impact toughness of the weld is insufficient, and there is a risk of brittle fracture. In the next round, the discriminator sees this new generated description and may be confused, thinking that it looks very real, and thus gives a higher score, such as 0.6; through thousands of rounds of such games, the generator can eventually generate high-quality, high-fidelity hidden risk descriptions that are almost indistinguishable from real risk records in statistics. During the training process, the generator continuously optimizes its ability so that its output hidden risk is more and more consistent with the distribution of real risks. After the training is completed, the first generator can be used to generate new hidden risks.

[0042] The first generator of the intermediate plug-in generates the first implicit event risk according to the first event relationship graph. Through multi-perspective implicit risk fusion, the generator will combine multiple different perspectives to determine the first implicit event risk. Multi-perspective implicit risk fusion is an analysis method that considers and evaluates implicit risks through different perspectives. Through the second generator of the intermediate plug-in, the first event rule is generated according to the implicit risk analysis result and the event relationship graph. The training process of the second generator is similar to the process of the aforementioned first generator, with the difference being that the first generator is a module responsible for risk identification, and the second generator is a module responsible for rule generation. The input data of the second generator comes from the first event relationship graph and the implicit risk generation module, and based on multi-dimensional data analysis, one or more safety rules are output. Using multi-perspective rule fusion technology means considering information from multiple angles when generating safety rules, such as environmental perspective, equipment perspective, and personnel perspective.

[0043] With the first implicit event risk as the core input, multi-perspective rule fusion decision is started, multiple alternative rule schemes are generated, and one optimal solution is selected from them to finally generate the first event rule tailored for this event. For example, the current lifting safety rule: immediately form a 4-person cable wind rope team; connect 4 Dilmama cable wind ropes with a diameter of 20 mm at the four corners of the steel box girder, controlled manually by personnel; the initial tension of the cable wind rope needs to be maintained within the range of 5-8 kN to ensure guidance and not excessive constraint; after meeting the above conditions, the work is allowed to continue under the condition that the wind speed is less than 12 m / s. Multi-perspective rule fusion refers to comprehensively considering various safety rules from different angles to generate the optimal safety management rule that adapts to complex construction sites.

[0044] Through the event relationship graph and multi-perspective implicit risk fusion, explicit and implicit risks on the construction site are comprehensively identified, improving the accuracy and timeliness of risk identification. Through multi-perspective fusion analysis of the event relationship graph and the first generator, deep risks caused by multi-factor coupling that the human brain cannot instantly clarify are discovered. Through the second generator and multi-perspective rule fusion, specific, feasible, and optimal solutions, i.e., the first event rule, are actively provided, thereby ensuring the safety of the construction process.

[0045] Further, the present application further includes the following steps: establishing a multi-perspective experience library, wherein different construction roles are defined for the perspectives; the first generator determines the multi-perspective implicit risk by interacting with the multi-perspective experience library and performing multi-perspective thread reasoning; and the multi-perspective implicit risk is fused as the first implicit event risk.

[0046] Specifically, the multi-perspective experience library refers to a database containing the experience and knowledge of different construction roles, such as operators, supervisors, etc. In different perspectives of construction roles, the cognition, judgment, and handling of safety risks are different, and multi-perspective joint is used to ensure completeness. Each perspective corresponds to different aspects and risk categories that a construction role will focus on when performing its duties. Construction roles refer to groups of people responsible for different tasks and duties in engineering projects, including operators, supervisors, safety officers, etc. For example, the operator perspective focuses on direct risks in construction operations, such as equipment failure, operation errors, environmental factors, etc.; the supervisor perspective focuses on the standardization and quality control of the construction process, identifying design defects, construction deviations, etc.; the safety officer perspective focuses on identifying potential safety hazards in the construction site, assessing environmental safety, equipment safety, and personnel safety.

[0047] The first generator interacts with the multi-perspective experience library to obtain the experience and knowledge of different construction roles. The first generator can obtain risk perspectives and risk data for different construction roles, thereby performing more accurate risk identification and judgment. Based on data and experience from different perspectives, multi-perspective thread reasoning is performed, and the reasoning process of each perspective is independent, generating corresponding risk judgments. For example, the operator perspective generates a risk about equipment out of control, hoisting operations may be out of control when the wind speed is greater than 12 m / s, with a risk probability of 80%; The supervisor perspective generates a risk about construction quality, and loose equipment connections may cause equipment to fall, with a risk probability of 60%; the safety officer perspective generates a risk of personnel injury, and the risk probability of personnel entering by mistake due to lack of safety signs is 40%. By combining the judgments of different roles on implicit risks, a comprehensive first implicit event risk is formed. By combining the judgments of different roles on implicit risks, a comprehensive risk assessment result is generated. The first implicit event risk refers to the potential risk that may cause a safety accident in the construction process based on multi-perspective implicit risk analysis. It integrates the experience and reasoning results of all relevant roles to provide a more comprehensive safety risk assessment. For example, the operator perspective indicates that the wind speed is 14 m / s, the equipment load is 90%, and the hoisting out-of-control risk is 80%; the supervisor perspective indicates that the equipment support beam connection is loose, and the equipment falling risk is 60%; the safety officer perspective indicates that there are not enough safety warning signs on the site, and the risk of entering by mistake is 40%. Through multi-perspective thread reasoning, the operator reasons that when the wind speed is greater than 12 m / s and the equipment load is too large, the risk of hoisting out of control is 80%; the supervisor reasons that the support beam connection is loose, and the risk of equipment falling is 60%; the safety officer reasons that there is a lack of safety signs, and the risk of personnel entering by mistake is 40%. Comprehensive risk: when the wind speed in hoisting operations exceeds 14 m / s, the equipment load exceeds 80%, and the support beam connection is loose, the comprehensive implicit risk assessment is 85%.

[0048] Further, the application further comprises the steps of: performing event model segmentation on the BIM engineering model, performing lightweight reconstruction according to the first event relationship diagram to determine a lightweight event BIM model; performing mode deployment on the lightweight event BIM model as a first event BIM model, taking the first feature set and the first implicit event risk as decision guidance, and taking the first event rule as a decision rule.

[0049] Specifically, according to the characteristics of the construction plan and the construction event, the BIM engineering model is segmented into event models to form multiple sub-models. According to the first event relationship diagram, the relevant data area is identified and extracted from the entire BIM model. According to the spatial and logical relationship between events and components in the first event relationship diagram, unnecessary details are removed, and only the core information directly related to the current event is retained. Lightweight reconstruction is completed to determine a lightweight event BIM model. The lightweight event BIM model is a model after lightweight reconstruction, which retains key information related to a specific construction event and eliminates irrelevant parts.

[0050] The first feature set and the first implicit event risk are taken as decision guidance, and the first event rule is taken as a decision rule. Decision guidance refers to core data or information guiding the decision-making process, and decision rules are rules formulated according to experience or models for making decisions according to specific conditions or characteristics. According to the first feature set and the first implicit event risk, the safety of the current construction event is evaluated. When the feature set and the implicit event risk of the construction site are identified, the first event rule is applied to determine whether to take appropriate safety measures. According to the decision guidance and the decision rule, the lightweight event BIM model is deployed in mode, and the known decision rule, feature set and implicit risk are applied to the lightweight event BIM model to complete real-time risk control decision and safety management. The first event BIM model is generated by mode deployment, which integrates the feature data of the construction event, the implicit risk analysis result and the decision rule to form a complete BIM model containing all risk assessment results and management instructions related to the current construction event. The first event BIM model is a BIM model constructed for a specific construction event.

[0051] By combining all relevant information into the lightweight event BIM model, the response speed of the construction site is improved, and potential risks are ensured to be responded to in real time. The lightweight reconstructed model only contains core data related to a specific event, reducing the complexity of the model and thus reducing the consumption of computing resources. By integrating the first feature set, the first implicit event risk and the first event rule into the model, risks are assessed and predicted in real time to ensure that safety management during construction is strengthened.

[0052] According to the first event BIM model, a first feature set and a first implicit event risk are used as decision guidance, and a first event rule is used as a decision rule to perform deduction and rule-constrained decision under multi-combination to generate a first safety rule and a first risk control plan, and to perform engineering safety management based on the first construction event.

[0053] Further, the application further includes the following steps: by mutual matching, combining the decision guidance and the decision rule to determine a plurality of combination pairs; for the plurality of combination pairs, performing deduction and rule-constrained decision in the first event BIM model to add to the decision set, wherein, single combination pair deduction and joint deduction of at least two combination pairs are included; according to the decision set, generating a first safety rule and a first risk control plan, and adding to the first event partition of the platform safety management partition.

[0054] Specifically, by mutual matching, the decision guidance and the decision rule are combined to generate a plurality of combination pairs. Each combination pair represents a specific feature condition and a corresponding safety rule. For example, when the wind speed exceeds 12 m / s and the equipment load exceeds 80%, a combination pair is generated, indicating that work needs to be suspended and equipment inspection needs to be performed under such conditions. Mutual matching refers to matching the decision guidance (composed of the first feature set and the first implicit event risk) with the decision rule, i.e., comparing the feature data of the current construction site with the preset decision rule. According to the feature data of the current construction environment and construction event, a plurality of combination pairs are generated. Each combination pair represents a condition and a corresponding safety rule.

[0055] According to the generated plurality of combination pairs, deduction is performed in the first event BIM model, including deduction of a single combination pair and joint deduction of a plurality of combination pairs. Deduction of a single combination pair is based on a specific feature and rule combination, and joint deduction of a plurality of combination pairs considers a plurality of feature and rule combinations to predict potential safety risks. Through deduction, various risks that may occur during construction are comprehensively judged, and corresponding safety decisions are generated. Deduction refers to predicting potential construction event risks and their consequences under given specific features and rules. Rule-constrained decision making is based on preset rules to make judgments and controls to ensure that certain safety standards are followed during deduction.

[0056] The decision set is the result set after deduction and rule-constrained decision-making, including single combination pair deduction and multiple combination pair joint deduction. According to the deduction result and rule-constrained decision-making, a first safety rule and a first risk control plan are generated. The first safety rule defines the optimal safety measures to be taken for a specific construction event, while the first risk control plan provides a response scheme for potential safety hazards. The safety rule and the risk control plan will be added to the first event partition of the platform safety management partition for use on the construction site. For example, a building project is carrying out hoisting operations, with a wind speed of 13 m / s, a device load of 85%, and a temperature of 30℃. The decision rules include suspending hoisting operations when the wind speed is greater than 12 m / s, checking the equipment when the device load is greater than 80%, and starting the high-temperature emergency plan when the temperature is greater than 28℃. Each combination pair is deduced separately to decide to suspend hoisting operations, check the equipment, and start the high-temperature emergency plan. The first safety rule is to suspend hoisting operations when the wind speed exceeds 12 m / s and notify the operator to avoid the dangerous area; the first risk control plan is to stop hoisting operations immediately if the device load exceeds 80%, perform equipment inspection, and ensure that the equipment is not damaged.

[0057] The platform safety management partition refers to the safety management area in the construction project, which is divided according to different construction stages, events, or areas. Within each partition, corresponding safety management strategies are generated based on construction events, risk assessment, decision rules, etc. Through deduction and rule-constrained decision-making, various risks that occur during construction are evaluated, and the optimal safety rules and risk control plans are generated to ensure that safety management during construction not only meets the actual situation but also responds to real-time changes on site. By deducing multiple combination pairs, each possible risk situation is accurately determined, and corresponding safety measures are generated for each situation.

[0058] Further, the application further includes the following steps: establishing temporary communication interaction between the first event partition and the front-end acquisition array; with the construction progress of the first construction event, performing monitoring based on the front-end acquisition array through spatial targeting, and performing safety management according to the first safety rule in the first event partition, when there is a risk trend exceeding the first safety rule, performing emergency management according to the first risk plan.

[0059] Specifically, the first event partition establishes a platform security management partition and temporarily communicates with the front-end acquisition array. The front-end acquisition array refers to a group of sensors, monitoring devices, and data acquisition modules distributed at the construction site, used to monitor the state data of the environment, equipment, personnel, etc. in real time during the construction process, including wind speed, temperature, equipment load, work status, etc. Temporary communication interaction is the real-time data exchange process between the platform and the front-end acquisition array during the construction process. This communication is temporary and is usually enabled during a specific construction event or task to transmit data in real time and make appropriate safety decisions as needed. The front-end device will collect data in real time and interact with the back-end system to provide necessary real-time data support during the construction process.

[0060] As the construction process of the first construction event progresses, real-time tracking of the site is performed through spatial targeting technology. Spatial targeting refers to tracking and positioning the real-time state of the construction site through certain spatial positioning technology. Through front-end acquisition array monitoring, after receiving the data from the front-end acquisition array, safety management is performed according to the pre-set first safety rules in the first event partition. If a risk trend is detected, i.e. certain characteristic data exceeds the safety range, an emergency management is performed according to the pre-set first risk plan. The risk plan may include suspending work, adjusting work plan, starting inspection procedures, etc. For example, data of wind speed 13 m / s and information of equipment load 85% are received; according to the first safety rules, it is judged that the wind speed exceeds 12 m / s, triggering the instruction to suspend hoisting work, and the equipment load exceeds 80%, triggering the equipment inspection instruction; it is evaluated that these conditions cause the hoisting work to have the risk of losing control, and according to the first risk plan, the emergency measures are automatically started, the hoisting work is suspended, and the equipment inspection procedure is started. The first risk plan is an emergency response measure generated based on the triggering conditions of the first safety rules. When certain safety risks are detected during the construction process, such as high equipment load, abnormal environmental conditions, etc., the risk plan is executed according to the pre-set risk plan to perform emergency management operations, such as suspending work, starting safety inspection, etc.

[0061] By establishing temporary communication interaction, real-time reception of environmental and equipment data from the construction site is achieved, and through spatial targeting technology and front-end monitoring array, the state changes of the construction site can be dynamically tracked, so that real-time adjustment of safety management measures can be performed. When a risk trend is detected, an emergency plan is automatically triggered and relevant management measures are executed, reducing the risk of human intervention and improving the efficiency and response speed of emergency management.

[0062] In summary, the BIM-based engineering safety management method provided in the application has the following beneficial effects: by collecting a first construction event, returning to the BIM engineering model established by the safety management platform, determining a first feature set and performing a risk control probability test based on rigid conditions, triggering a first safety management instruction; according to the first safety management instruction, activating the intermediate plug-in, performing implicit risk mining and dynamic rule generation, performing local lightweight reconstruction on the BIM engineering model, determining a first event BIM model; according to the first event BIM model, taking the first feature set and the first implicit event risk as the decision orientation, taking the first event rule as the decision rule, performing deduction and rule constraint decision under multi-element combination, generating the first safety rule and the first risk control plan, and performing engineering safety management based on the first construction event. That is, by taking the construction event as the management mode, short-time simulation analysis is performed on the BIM engineering model, and when the risk control exceeds the limit, the optimal safety rule and the risk control plan are dynamically generated, so that the rapid identification and accurate response to potential risks are realized, and the response speed and safety management efficiency of the construction site are greatly improved.

[0063] In the second embodiment, based on the same inventive concept as the BIM-based engineering safety management method in the foregoing first embodiment, the application further provides a BIM-based engineering safety management system. Please refer to the accompanying drawings Figure 2 The BIM-based engineering safety management system comprises: A risk control probability test module 11 is configured to collect a first construction event, return to a BIM engineering model established by a safety management platform, determine a first feature set, and perform a risk control probability test based on rigid conditions, thereby triggering a first safety management instruction. A local lightweight reconstruction module 12 is configured to activate an intermediate plug-in according to the first safety management instruction, perform implicit risk mining and dynamic rule generation, perform local lightweight reconstruction on the BIM engineering model, and determine a first event BIM model. An engineering safety management module 13 is configured to take the first feature set and the first implicit event risk as the decision orientation, take the first event rule as the decision rule, perform deduction and rule constraint decision under multi-element combination, generate a first safety rule and a first risk control plan, and perform engineering safety management based on the first construction event.

[0064] Further, the risk control probability test module 11 in the BIM-based engineering safety management system is further configured to connect an engineering database, call engineering construction records, mine a construction condition sequence according to the engineering construction records, wherein the construction condition sequence contains engineering features-risk probability satisfying a preset generalization degree, set a condition probability threshold as a rigid condition, and deploy a prior port in the BIM engineering model.

[0065] Further, the risk control probability verification module 11 in the BIM-based engineering safety management system is further configured to: according to the first construction event, perform short-time simulation on the BIM engineering model to determine a first feature set, wherein the first feature set includes event features and dynamic features; trigger the priori port to perform rigid condition matching and out-of-limit determination on the first feature set, and generate the first safety management instruction if there is any feature with an out-of-limit probability; and generate a standard safety management instruction if there is no out-of-limit probability.

[0066] Further, the local lightweight reconstruction module 12 in the BIM-based engineering safety management system is further configured to: according to the first construction event, construct a first event relationship graph, wherein the event relationship graph takes engineering components as nodes and spatial relationships between components as edges, and the nodes and edges are identified with attribute elements; according to the first event relationship graph, the first generator of the intermediate plug-in performs implicit risk generation to determine a first implicit event risk, wherein the first implicit event risk is determined based on multi-perspective implicit risk fusion; and according to the second generator of the intermediate plug-in, generate a first event rule, wherein the first event rule is determined based on multi-perspective rule fusion.

[0067] Further, the local lightweight reconstruction module 12 in the BIM-based engineering safety management system is further configured to: establish a multi-perspective experience library, wherein different construction roles are defined as perspectives; the first generator performs multi-perspective thread reasoning by interacting with the multi-perspective experience library to determine multi-perspective implicit risks; and fuse the multi-perspective implicit risks as the first implicit event risk.

[0068] Further, the local lightweight reconstruction module 12 in the BIM-based engineering safety management system is further configured to: perform event model segmentation on the BIM engineering model, perform lightweight reconstruction according to the first event relationship graph to determine a lightweight event BIM model; and perform mode deployment on the lightweight event BIM model as a first event BIM model with the first feature set and the first implicit event risk as decision guidance and the first event rule as a decision rule.

[0069] Further, the engineering safety management module 13 in the BIM-based engineering safety management system is further configured to: combine decision guidance and decision rules by mutual matching to determine a plurality of combination pairs; perform deduction and rule constraint decision in the first event BIM model for the plurality of combination pairs to add to a decision set, wherein the decision set includes single combination pair deduction and joint deduction of at least two combination pairs; generate a first safety rule and a first risk control plan according to the decision set and add them to a first event partition of a platform safety management partition.

[0070] Further, the engineering safety management module 13 in the BIM-based engineering safety management system is further configured to: establish temporary communication interaction between the first event partition and the front-end acquisition array; with the construction progress of the first construction event, perform front-end acquisition array-based monitoring through spatial targeting, and perform safety management according to the first safety rule in the first event partition; and when there is a risk trend exceeding the first safety rule, perform emergency management according to the first risk plan.

[0071] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The foregoing Figure 1 The BIM-based engineering safety management method and specific examples in Embodiment One are also applicable to the BIM-based engineering safety management system of the present embodiment. Based on the foregoing detailed description of the BIM-based engineering safety management method, those skilled in the art can clearly understand the BIM-based engineering safety management system in the present embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0072] Embodiment Three, based on the same inventive concept as the BIM-based engineering safety management method in the foregoing Embodiment One, the present application also provides a computer readable storage medium, which stores a computer program. The computer program, when executed, implements the steps of the BIM-based engineering safety management method in any one of the foregoing Embodiments One.

[0073] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0074] Obviously, for those skilled in the art, without departing from the principles of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for engineering safety management based on BIM, characterized by, The method comprises the following steps: Collecting a first construction event, returning to a BIM engineering model established by a safety management platform, determining a first feature set and performing a risk control probability test based on rigid conditions, triggering a first safety management instruction; According to the first safety management instruction, activate the intermediate plug-in, execute implicit risk mining and dynamic rule generation, perform local lightweight reconstruction on the BIM engineering model, and determine the first event BIM model; According to the first event BIM model, taking the first feature set and the first implicit event risk as the decision orientation, and taking the first event rule as the decision rule, the deduction and rule constraint decision under the multi-element combination are executed, the first safety rule and the first risk control plan are generated, and the engineering safety management based on the first construction event is executed.

2. The BIM-based engineering safety management method of claim 1, wherein, Before performing the risk control probability test based on the rigid condition, the rigid condition is obtained, including: Connect the engineering database and call the engineering construction record; According to the engineering construction record, the construction condition sequence is mined, wherein the construction condition sequence contains the engineering feature-risk probability meeting the preset generalization degree; For the construction condition sequence, set the condition probability threshold as the rigid condition, and deploy the prior port in the BIM engineering model.

3. The BIM-based engineering safety management method of claim 2, wherein, Performing a risk control probability test based on rigid conditions includes: According to the first construction event, short-time simulation is performed on the BIM engineering model to determine the first feature set, wherein the first feature set includes event features and dynamic features; Trigger the prior port to perform rigid condition matching and out-of-limit judgment on the first feature set. If there is any feature with probability out of limit, generate the first safety management instruction; If there is no probability out of limit, generate a standard safety management instruction.

4. The BIM-based engineering safety management method of claim 1, wherein, Performing implicit risk mining and dynamic rule generation includes: According to the first construction event, a first event relationship graph is constructed, wherein the event relationship graph takes engineering components as nodes and spatial relationships between components as edges, and nodes and edges are identified with attribute elements; According to the first event relationship graph, the first generator of the intermediate plug-in performs implicit risk generation to determine the first implicit event risk, wherein the first implicit event risk is determined based on multi-perspective implicit risk fusion; According to the second generator of the intermediate plug-in, the first event rule is generated, wherein the first event rule is determined based on multi-perspective rule fusion.

5. The BIM-based engineering safety management method of claim 4, wherein, Performing implicit risk generation to determine the first implicit event risk includes: Establish a multi-perspective experience library, wherein different construction roles are defined as perspectives; The first generator determines the multi-perspective implicit risk by interacting with the multi-perspective experience library through multi-perspective thread reasoning; Fuse the multi-perspective implicit risk as the first implicit event risk.

6. The BIM-based engineering safety management method of claim 5, wherein, Performing local lightweight reconstruction on the BIM engineering model to determine the first event BIM model includes: Performing event model segmentation on the BIM engineering model, performing lightweight reconstruction according to the first event relationship graph, and determining a lightweight event BIM model; Taking the first feature set and the first implicit event risk as the decision orientation, and taking the first event rule as the decision rule, the lightweight event BIM model is deployed as the first event BIM model.

7. The BIM-based engineering safety management method of claim 1, wherein, The deduction and rule constraint decision under multi-element combination are performed to generate the first safety rule and the first risk control plan, including: Through mutual matching, the decision orientation and the decision rule are combined to determine a plurality of combination pairs; For the plurality of combination pairs, deduction and rule constraint decision are performed in the first event BIM model to add to the decision set, wherein, single combination pair deduction and joint deduction of at least two combination pairs are included; According to the decision set, the first safety rule and the first risk control plan are generated and added to the first event partition of the platform safety management partition.

8. The BIM-based engineering safety management method of claim 7, wherein, The engineering safety management based on the first construction event is performed, including: Establishing temporary communication interaction of the first event partition and the front-end acquisition array; With the construction process of the first construction event, the monitoring based on the front-end acquisition array is performed through spatial targeting, and the safety management is performed according to the first safety rule in the first event partition. When there is a risk trend exceeding the first safety rule, the emergency management is performed according to the first risk plan.

9. A BIM-based engineering safety management system, characterized by, The steps of the BIM-based engineering safety management method in any one of claims 1 to 8 are implemented, and the BIM-based engineering safety management system includes: A risk control probability verification module is configured to collect a first construction event, return to a BIM engineering model established by a safety management platform, determine a first feature set, and perform risk control probability verification based on a rigid condition to trigger a first safety management instruction; A local lightweight reconstruction module is configured to activate an intermediate plug-in according to the first safety management instruction, perform implicit risk mining and dynamic rule generation, perform local lightweight reconstruction on the BIM engineering model, and determine a first event BIM model; An engineering safety management module is configured to perform deduction and rule constraint decision under multi-element combination according to the first event BIM model, take the first feature set and the first implicit event risk as the decision orientation, and take the first event rule as the decision rule to generate the first safety rule and the first risk control plan, and perform engineering safety management based on the first construction event.

10. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium, and the computer program is implemented when executed to realize the steps of the BIM-based engineering safety management method in any one of claims 1 to 8.

Images