Gas accident disaster simulation method and system based on BIM and virtual interaction

Abstract

The invention relates to the technical field of gas accident disaster simulation, and discloses a gas accident disaster simulation method and system based on BIM and virtual interaction. The method comprises the following steps: acquiring real-time acquisition signals of electric equipment, a gas sensor and a temperature sensor in a to-be-detected area, when an abnormal signal is identified, executing disaster situation simulation operation based on an evaluation model, and outputting a simulation evaluation result; according to the method, a simulation evaluation result is obtained, the simulation evaluation result is verified to output an evaluation result with credibility, an evaluation model is constructed by extracting common information of a known gas accident, and evaluation nodes are determined based on the equipment level to load the corresponding evaluation model. According to the method, the evaluation model is constructed, the differentiated disaster simulation operation is executed, and the result is verified, so that accurate perception and early warning of the disaster precursor are realized, and the accuracy and reliability of the evaluation result are ensured.

Description

Technical Field

[0001] This invention belongs to the field of gas accident disaster simulation technology, specifically relating to a gas accident disaster simulation method and system based on BIM and virtual interaction. Background Technology

[0002] With the in-depth development of smart mines, intelligent construction and other fields, it has become an industry consensus to use digital technology to improve the safety production management level of underground engineering. Based on Building Information Modeling (BIM), a high-fidelity digital twin environment is constructed, and combined with virtual reality technology, disaster and accident simulation exercises are carried out. Through dynamic simulation of the disaster process, it is conducive to personnel emergency training, rescue plan formulation and risk assessment.

[0003] Existing BIM-based disaster simulation technologies lack evaluation and verification mechanisms for simulation results. Current technologies often directly output the gas diffusion paths, concentration distributions, and other disaster evolution processes generated by simulation calculations as the final conclusions, without verifying, analyzing, or assessing the credibility of the simulation results. This results in poor accuracy and reliability of the output results. Existing disaster simulation technologies only focus on the physical diffusion process of the gas itself in gas accident simulations, ignoring the dynamic interaction with various electrical devices and key equipment in the environment. They fail to judge and analyze the risk level of potential ignition sources and cannot effectively predict the critical turning point from gas leakage to explosion. Existing disaster simulation technologies lack interactivity and dynamism, and cannot support drill participants in taking countermeasures such as turning on ventilation and cutting off power in real time during the simulation to influence the evolution of the disaster, thus limiting their functionality in practical drills.

[0004] To address the aforementioned problems, this invention provides a method and system for simulating gas accident disasters based on BIM and virtual interaction. Summary of the Invention

[0005] The purpose of this invention is to provide a gas accident disaster simulation method based on BIM and virtual interaction, so as to solve the technical problem that the existing technology does not have an analysis process for the gas disaster simulation results, which leads to the evaluation results being directly output as the final result without being reviewed.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows: A gas accident simulation method based on BIM and virtual interaction includes the following steps: Acquire real-time signals from electrical equipment, gas sensors, and temperature sensors within the area to be tested; When an abnormal signal is identified from the real-time acquired signal, execute: Constructing an assessment model includes: building a dataset based on known gas accidents; and extracting common information from the dataset to construct the assessment model. Based on the assessment model, a disaster simulation operation is performed to output the simulation assessment results; the simulation assessment results are verified to output a reliable assessment result. The common information of the data set includes information consistency and common dimension value. The information consistency is used to describe the pairwise similarity of parameter information among known gas accidents, and the common dimension value is the identification content whose information consistency is greater than or equal to a set threshold. The process of obtaining the information consistency includes: vectorizing the information in the dataset; and extracting the similarity by calculating the cosine of the angle between each pair of vectors, as the information consistency.

[0007] Preferably, identifying the abnormal signal includes: determining the critical point signal in the real-time acquired signal and labeling the critical point signal as the abnormal signal; labeling the sampling point signal in the real-time acquired signal as a normal signal, and obtaining the lower limit and upper limit of the fluctuation of the detection parameter based on the statistical analysis of the normal signal.

[0008] Preferably, the disaster simulation operation includes: determining an assessment node based on the equipment level; and loading the corresponding assessment model under the assessment node to perform the disaster simulation operation.

[0009] Preferably, the equipment level includes Level 1 equipment, Level 2 equipment and Level 3 equipment, wherein Level 1 equipment is the equipment with the highest risk level, Level 2 equipment is the equipment that causes current fluctuations due to gas concentration fluctuations, and Level 3 equipment is the equipment with the lowest risk level.

[0010] Preferably, the verification of the simulation assessment results includes: obtaining benchmark parameters representing the actual risk status of the area to be tested; generating an assessment offset value based on the benchmark parameters and the simulation assessment results; and determining the reliability of the simulation assessment results based on the assessment offset value.

[0011] A gas accident simulation system based on BIM and virtual interaction includes the following modules: The information acquisition module is used to acquire real-time signals from electrical equipment, gas sensors, and temperature sensors in the area under test. An anomaly detection module is used to identify abnormal signals from the real-time acquired signals; The simulation execution module is used to perform disaster simulation operations based on the evaluation model in response to the anomaly identification module's identification of the anomaly signal, so as to generate simulation evaluation results; The verification output module is used to verify the simulation evaluation results in order to output a reliable evaluation result. The system construction module is used to optimize subsequent disaster simulation operations based on the credible assessment results.

[0012] Preferably, the system further includes a model building module for constructing the evaluation model by extracting common information from a dataset based on known gas accidents.

[0013] Preferably, the model building module extracts common information from the dataset constructed based on known gas accidents, including information consistency and common dimension values; The process of obtaining the information consistency includes: vectorizing the information in the dataset; and extracting the similarity by calculating the cosine of the angle between each pair of vectors, as the information consistency.

[0014] Preferably, the simulation execution module is specifically used to: determine an evaluation node based on the equipment level associated with the electrical equipment in the area to be tested; and load the corresponding evaluation model under the evaluation node.

[0015] Preferably, the equipment level includes Level 1 equipment, Level 2 equipment and Level 3 equipment, with Level 1 equipment being the equipment with the highest risk level, Level 2 equipment being the equipment that causes current fluctuations due to gas concentration fluctuations, and Level 3 equipment being the equipment with the lowest risk level.

[0016] Beneficial effects This invention acquires real-time signals from electrical equipment, gas sensors, and temperature sensors within the test area, identifies anomalies in the signals, determines critical point signals and normal signals, and statistically analyzes the lower and upper limits of fluctuation for each detection parameter. This enables accurate perception and early warning of precursors to gas accidents, providing a timely and accurate data foundation for disaster simulation and emergency response, and avoiding the problem of warning failure caused by data lag or inaccuracy.

[0017] This invention constructs an evaluation model based on known gas accidents. Known gas accidents are divided into collection sets, and common information is extracted from these sets, including information consistency and common dimension values. Information consistency describes the pairwise similarity of parameter information among known gas accidents, while common dimension values ​​are identifiers where information consistency is greater than or equal to a set threshold. Information from each collection set is vectorized, and the cosine of the angle between any two vectors is calculated as the similarity to obtain information consistency. Common dimension values ​​are extracted based on the set threshold. An evaluation model is constructed based on the inherent patterns and common characteristics of historical gas accidents, improving the accuracy and reliability of simulation evaluation results and overcoming simulation biases caused by the limitations of single-accident data.

[0018] In the assessment node determination and simulation execution phase, this invention loads the corresponding assessment model based on the equipment level to determine the assessment information. Combining the electrical equipment and gas present in the test area, it performs disaster simulation operation and outputs simulation assessment results. The equipment level includes the highest risk level Level 1 equipment, Level 2 equipment with gas concentration fluctuations causing current fluctuations, and the lowest risk level Level 3 equipment. The risk level is gradually reduced from Level 1 equipment as the benchmark. According to the risk level and characteristics of different equipment, the simulation strategy and assessment model are dynamically adjusted to make the disaster simulation closer to the actual working conditions and improve the accuracy and relevance of the simulation.

[0019] After the simulation is completed, the present invention uses the simulation evaluation results as the data to be verified, obtains the benchmark parameters that can represent the actual risk status of the area to be tested, generates the evaluation offset value based on the benchmark parameters, judges the credibility of the simulation evaluation results based on the evaluation offset value, and outputs the simulation evaluation results with credibility. The simulation evaluation results are objectively and quantitatively verified and evaluated, ensuring the authenticity and reliability of the simulation evaluation results. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method provided by the present invention; Figure 2 This is a system module diagram provided by the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Example 1 Please see Figure 1 A gas accident disaster simulation method based on BIM and virtual interaction includes the following steps: It acquires real-time signals from electrical equipment, gas sensors, and temperature sensors deployed in the area to be tested.

[0023] Furthermore, abnormal signals are identified from the real-time acquired signals: among the acquired real-time acquired signals, the critical point signals whose values ​​reach or exceed the preset threshold are identified and calibrated as abnormal signals. Abnormal signals usually indicate that the equipment deviates from the normal operating state or the preset safety range. Abnormal signals are usually manifested as sudden changes in values, continuous exceedance of the threshold, or abnormal fluctuation patterns. The sampling point signals are calibrated as normal signals. The sampling point signals are discrete data point signals acquired in the continuous real-time acquired signal stream according to a specific time interval or event triggering mechanism. Normal signals indicate that the equipment is in a stable operating state and that the various parameter indicators are within the preset normal fluctuation range. The normal signals are statistically analyzed and processed to obtain the lower limit of fluctuation and the upper limit of fluctuation of each detection parameter. Specifically, after acquiring a normal signal, the fluctuation values ​​of each signal are continuously statistically analyzed. The fluctuation values ​​are calculated by determining the difference between the previous and subsequent signals and treating this difference as the fluctuation value. All obtained fluctuation data are extracted, aggregated into a dataset, and then analyzed. By calculating statistical distribution characteristics such as the mean, standard deviation, or specific percentiles, the lower and upper limits of the fluctuation of the detection parameters are determined. If the fluctuation range of the current sampling point signal is between the determined lower and upper limits of the fluctuation of the detection parameters, it is determined to be in an abnormal state. If it is not within this range, it is considered that the corresponding device may be affected by some external factors and is experiencing fluctuations.

[0024] Furthermore, an evaluation model is constructed based on historical data of known gas accidents with detailed records. Data subsets are collected and divided according to specific criteria such as accident type, location, and time period. Based on these subsets, a comprehensive dataset is constructed. This comprehensive dataset is a large dataset containing a wider range of historical gas accident data. Common information is extracted from the dataset, including information consistency and common dimension values. Common information reflects the general patterns and common characteristics among different gas accidents. Information consistency describes the pairwise similarity of parameter information among different known gas accidents, while common dimension values ​​refer to the identifiers where information consistency is greater than or equal to a set threshold. These values ​​represent the core characteristics of gas accidents. Information consistency aims to quantify the similarity among different known gas accidents in key technical parameters such as gas concentration change rate, temperature rise rate, and equipment current fluctuation, thereby identifying the common characteristics of accident patterns. The set threshold is a pre-defined critical value during data processing or decision-making. This threshold is used to filter out highly similar accident patterns, ensuring that the subsequently constructed evaluation model accurately reflects the general patterns of gas accidents. Preferably, in the process of obtaining information consistency, the information in each collection set is vectorized. Vectorization converts the parameter information such as the gas concentration change sequence over time, temperature change sequence, and equipment operating status sequence of each known gas accident into numerical vectors in a high-dimensional space for mathematical operations. Then, the vectors are compared pairwise, and the similarity is extracted by calculating the cosine value of the angle between the two vectors. The similarity is used as the information consistency. By calculating the cosine value of the angle between these vectors, the directional similarity of different accident modes in the multi-dimensional parameter space is quantified. The closer the cosine value is to 1, the more similar the parameter change trends of the two accident modes are, thus more accurately assessing the information consistency between them. Preferably, in the process of obtaining common dimension values, an information consistency threshold is set, and the information with an information consistency of not less than the information consistency threshold in the collected information is extracted as common dimension values. Common dimension values ​​are a set of parameters that have been screened and refined and can represent the core characteristics of gas accidents. By setting the information consistency threshold, gas accident patterns with high similarity can be screened to ensure that only those accident patterns that are highly similar in the trend of key parameter changes are classified into one category, and more representative and universal common dimension values ​​are extracted. Common dimension values ​​will serve as the basis for building the evaluation model and improve the generalization ability and prediction accuracy of subsequent disaster simulation operations.

[0025] The functional relationship of the evaluation model is:

[0026] In the formula, This represents a vectorization operation, which means converting the first vector into the second vector. Parameter sequence of the accident Convert to a high-dimensional numerical vector The operation, This represents a set of historical data on known gas accidents. This means that it contains detailed parameter records of multiple gas accidents that have occurred, which is the basis for model learning; Indicates the first The parameter sequence of the incident. It means the collection of various technical parameters such as gas concentration and temperature that change over time in a single gas accident; Indicates the first The first accident A parameter sequence, which means a sequence of data describing the change of a specific technical parameter over time in a single accident; This represents the vectorized accident parameter vector, which means that the multivariate parameter sequence of a single gas accident is converted into a high-dimensional numerical vector for similarity calculation. This represents the vectorized accident parameter vector, which means that the multivariate parameter sequence of a single gas accident is converted into a high-dimensional numerical vector for similarity calculation. Indicates the consistency of information, which means that the first... accident and The cosine similarity between the parameter vectors of each accident is used to quantify their degree of similarity. This represents the angle between vectors, and its meaning is that the vectors... and The angle between them reflects their similarity in direction; This indicates a fused data structure. ,right Cluster analysis was performed on the feature vectors in the dataset, using the K-means clustering algorithm. The data was divided into groups by minimizing the intra-cluster distance. Each cluster is formed by extracting and clustering common dimension values ​​representing the core features of gas accidents from the selected high-similarity accident pairs. This means that it includes all the extracted common dimension values, which serve as the basic data template for subsequent disaster simulation. Indicates the first Each common dimension value represents a vector that signifies a specific gas accident mode or its key characteristics. This represents the number of known gas accidents, which is the total number of gas accidents in the historical dataset. This indicates the number of parameter sequences in a single accident, which means the number of types of technical parameters recorded in a single gas accident. This represents the number of common dimension values, which means the number of core feature vectors of gas accident patterns contained in the fused data structure.

[0027] Furthermore, based on the assessment model, a disaster simulation operation is performed to output the simulation assessment results: During the disaster simulation process, several assessment nodes are determined based on the assessment information. The assessment nodes are key time points or spatial locations where risk assessment and processing logic loading are required. Under each assessment node, the corresponding assessment processing logic is loaded. The assessment processing logic is a set of preset rules or algorithms for analyzing, judging, and assessing the equipment status, environmental parameters, or accident evolution trend. Combined with the electrical equipment and gas distribution in the area to be tested, the disaster simulation operation is performed. The output data obtained after predicting and quantifying the possible consequences, impact range, and risk level of the gas accident is used as the simulation assessment result. The assessment information, determined based on equipment levels, typically includes the equipment type, importance, environment, and historical failure data. The introduction of equipment levels aims to achieve differentiated risk assessment and optimized resource allocation in disaster simulation operations. Different levels of equipment may play different roles or have different vulnerabilities in gas accidents; therefore, the priority and depth of the assessment need to be determined based on their importance and risk exposure level. Equipment levels include Level 1, Level 2, and Level 3 equipment, with Level 1 equipment representing the highest risk. Level 1 equipment typically refers to critical equipment that, if it fails in a gas accident, will directly lead to significant casualties, equipment damage, or a rapid expansion of the accident area. Examples include main ventilation fans, main power supply equipment, and electrical equipment in densely populated areas. Assessing them with the highest risk level ensures that these equipment receive priority attention in the simulation. The potential risks of the equipment and the development of the most stringent emergency plans are considered. Level 2 equipment refers to equipment that exhibits current fluctuations caused by gas concentration fluctuations. Level 2 equipment is those that are sensitive to changes in gas concentration, and whose operating status, such as current and temperature, changes significantly due to gas concentration fluctuations. Examples include certain explosion-proof electrical equipment and gas extraction pumps. In the early stages of a gas accident, this type of equipment may serve as an indicator of gas anomalies, and its current fluctuations can serve as an important basis for assessing the evolution of the accident. Level 3 equipment represents the equipment with the lowest risk level. Level 3 equipment refers to equipment whose failures have a relatively small impact on the overall disaster situation or whose own risk exposure is low in a gas accident. Examples include auxiliary lighting equipment and non-critical monitoring equipment. The purpose of conducting the lowest risk level assessment is to focus on more critical equipment with limited simulation resources, while also comprehensively covering all equipment in the area to be tested. Preferably, the risk level of Level 1 equipment is used as a benchmark, and the risk is gradually reduced to Level 2 equipment until all equipment is downgraded. The remaining equipment is then marked as Level 3 equipment. This step-by-step downgrade method based on risk level aims to establish a clear and operable equipment risk assessment system. First, the most critical Level 1 equipment is identified. Then, Level 2 equipment is identified based on specific technical characteristics such as current fluctuations caused by gas concentration fluctuations. Finally, all equipment not identified as Level 1 or Level 2 equipment is automatically classified as Level 3 equipment. This grading strategy ensures that all equipment is included in the risk assessment scope, and the level of granularity of the assessment matches the potential risk contribution of the equipment, thereby improving the efficiency and accuracy of disaster simulation operations.

[0028] Furthermore, the simulation evaluation results are verified to output a credible evaluation result: After the simulation, the simulation evaluation results are used as data to be verified to assess their accuracy and credibility. Benchmark parameters that can represent the actual risk status of the area under test are obtained. Based on the difference between the benchmark parameters and the data to be verified, an evaluation offset value is generated to quantify the degree of deviation between the simulation results and the actual situation. The credibility of the simulation evaluation results is then judged based on the evaluation offset value. Finally, a credible evaluation result that accurately reflects the actual situation is output. The purpose of result verification is to ensure that the output results of the disaster simulation operation are highly consistent with the actual situation. Through the evaluation offset value, the credibility of the simulation can be objectively evaluated. If the evaluation offset value is within an acceptable range, the simulation evaluation result is considered credible; otherwise, the simulation parameters or processing logic need to be adjusted until the credibility standard is met. Preferably, when acquiring simulation data, the equipment status, environmental parameters, alarm and fault time points of the alarm nodes and fault nodes corresponding to the equipment in each test area are imported as simulation data. Both alarm nodes and fault nodes are time nodes. The values ​​of alarm nodes and fault nodes are defined as relay nodes. An alarm node refers to the warning time point triggered when the equipment reaches the preset warning value when parameters such as gas concentration and temperature reach the preset warning value, while a fault node refers to the time point when the equipment actually experiences functional failure or damage. Importing alarm nodes and fault nodes as simulation data accurately defines the key time axis of accident occurrence and development, providing accurate time reference for subsequent relay node division and distributed simulation. Relay nodes are intermediate time points between alarm nodes and fault nodes, further subdivided according to time intervals and technical logic, used to capture state changes in the accident evolution process more precisely. Preferably, an alarm node is taken as the starting point and a fault node as the ending point, forming a continuous time period to be analyzed. Within this time period, based on the time interval between the starting point and the ending point, a set of intermediate time points are obtained by dividing equally or according to specific rules. The intermediate time points can be used for subsequent distributed comparison. The relay node can also be called the secondary alarm node. Multi-point simulation is performed according to the distribution of normal or abnormal equipment status. The time sequence of the alarm node, fault node and relay node is arranged into a continuous time axis. Then, all known historical fault points in the test area are also arranged on the time axis in chronological order. If the time difference between the node and the fault point is less than the preset time tolerance, it indicates that the node is highly correlated with the fault point. The nodes with high correlation to be evaluated are calculated and identified and called evaluation nodes. The evaluation node serves as the risk offset point induced by the fault. The corresponding evaluation processing logic will be loaded under the evaluation node. Evaluation processing logic is loaded under each evaluation node, such as temperature evaluation processing logic, gas evaluation processing logic or joint evaluation processing logic. Specifically, the temperature assessment processing logic is as follows: When the parameter under the assessment node is mainly temperature, only a single temperature assessment processing logic needs to be loaded to analyze the temperature change trend, judge the temperature anomaly, and assess its risk. The gas assessment processing logic is as follows: When the parameter under the assessment node only involves gas concentration, a single gas assessment processing logic is loaded to analyze gas concentration changes, judge gas anomalies, and assess its risk. The joint assessment processing logic is as follows: When the extracted fluctuation parameter contains multiple variables, such as gas concentration, temperature, current, etc., and simultaneously affects the final equipment status, the joint assessment processing logic is adopted to uniformly process the multi-variable parameters and comprehensively judge the risk level. Preferably, when performing disaster simulation operations, based on the real-time distribution location of each device and in combination with relevant risk coefficients, disaster simulation operations are performed in the constructed BIM 3D scene to obtain simulation evaluation results. The simulation evaluation results are output as evaluation offset values. If the evaluation offset value is not greater than the allowable range, the simulation evaluation result is considered reliable and marked as a qualified result. If the evaluation offset value exceeds the allowable range, the offset direction and offset degree will be re-evaluated, and new parameters will be recalculated based on the evaluation results. The new parameters will be re-imported into the corresponding evaluation node, and the simulation will be performed again until the simulation evaluation result meets the requirements. The advantage of rapid evaluation and processing is that it can continuously improve the realism of the simulation evaluation results, gradually reduce the degree of simulation deviation from the actual situation, and ensure that the final disaster simulation operation is closer to the actual environment.

[0029] In this embodiment, the safety status of different equipment in the mining area is monitored and identified throughout the entire process by real-time signal acquisition. Then, an adaptive simulation basic data structure is constructed by multi-dimensional sampling and comparison of known gas accidents. Relay node partitioning technology is used to realize the automatic extraction of risk-inducing nodes and dynamic adjustment of risk levels. Finally, the offset value is evaluated by rapid iteration of multi-dimensional parameters to obtain a credible evaluation result. Data acquisition, fault identification, multi-dimensional analysis and disaster simulation operation are closely integrated to construct a complete closed-loop disaster drill logic path.

[0030] Example 2 Please refer to Figure 2 This embodiment provides a gas accident disaster simulation system based on BIM and virtual interaction. The system monitors the status of the area under test in real time. When abnormal signals are identified, it automatically performs disaster simulation operations and verifies the results, thereby outputting a credible assessment result to support accident early warning and emergency decision-making.

[0031] In its implementation, this system can be deployed on servers, workstations, or dedicated embedded computing devices in a mine monitoring center. It communicates with electrical equipment, gas sensors, and temperature sensors within the monitored area via industrial Ethernet, wireless sensor networks, and other methods. Logically, the system can be divided into the following collaborative modules: The model building module is used to construct the assessment model required for disaster simulation operations. Based on known gas accidents in historical records, a dataset is constructed and analyzed to extract common information. The process of extracting common information includes: vectorizing the parameter information describing each accident in the dataset into vectors in a high-dimensional space; calculating the cosine value of the angle between each pair of vectors to quantify the similarity as the information consistency; identifying the content with an information consistency greater than or equal to a set threshold as common dimension values. Based on the extracted common information, an assessment model that can be called by subsequent processes is finally constructed.

[0032] The information acquisition module is used to continuously acquire real-time signals from the area under test. Through the data interface, it periodically or event-triggeredly receives data from various sensing and monitoring devices deployed in a working face under the mine, including electrical signals such as the operating current and voltage of various electrical equipment, the concentration reading of the gas sensor, and the temperature reading of the temperature sensor. These data signals together constitute a real-time description of the environment and equipment status of the area.

[0033] The anomaly identification module is responsible for processing the real-time acquired signals collected by the information acquisition module and identifying abnormal signals. It analyzes the received signal stream and determines the signal at that moment as a critical point signal when the detected signal value exceeds a preset safety threshold or its rate of change changes abruptly, thus marking it as an anomaly signal to trigger subsequent disaster simulation operations. Other sampling point signals not identified as critical points are marked as normal signals. In order to dynamically optimize the sensitivity and accuracy of detection, this module also performs statistical analysis based on continuously acquired normal signals to calculate the lower limit and upper limit of the fluctuation of key detection parameters under normal operating conditions, providing a dynamic benchmark for subsequent anomaly judgment.

[0034] The simulation execution module is activated after the anomaly identification module identifies anomaly signals. Its core task is to perform disaster simulation operations based on the assessment model. According to the source of the anomaly signal, i.e., the associated electrical equipment, the module determines the equipment level. The equipment level is pre-classified according to the risk level of the equipment. Core equipment such as tunneling machines and coal mining machines can be classified as the highest risk level, Level 1 equipment, while auxiliary ventilation or drainage equipment can be classified as Level 2 or Level 3 equipment. Based on the determined equipment level, the module selects one or more corresponding assessment nodes. Under the selected assessment node, the module loads the assessment model pre-generated by the model building module that matches the equipment level and scenario, and performs disaster simulation operations. Finally, it generates a simulation assessment result describing the development trend of the disaster.

[0035] The verification output module verifies the simulation evaluation results generated by the simulation execution module to ensure the reliability of the output results. It obtains a benchmark parameter that represents the actual risk state of the area under test. This benchmark parameter can be derived from an independent, high-precision verification sensor or calculated in real time through a simplified physical model. The simulation evaluation results are compared with the benchmark parameter to generate an evaluation offset value, which quantifies the difference between the simulation results and the actual state. The reliability of the simulation evaluation results is judged based on the magnitude of the evaluation offset value. If the offset value is within an acceptable range, the simulation evaluation result, along with its reliability information, is output as a reliable evaluation result; otherwise, the result can be marked as low reliability or a re-simulation can be triggered.

[0036] The system construction module optimizes subsequent disaster simulation operations based on credible assessment results, forming a closed-loop adaptive system. It receives simulation assessment results and assessment offset values ​​from the verification output module, analyzes feedback data, fine-tunes the parameters of the assessment model itself, or adjusts the selection strategy of assessment nodes and the loading rules of the assessment model in the simulation execution module. If an assessment model is found to continuously generate large assessment offset values ​​under specific working conditions, the system construction module can reduce the weight of that model in similar scenarios or trigger the model construction module to update it. Through this continuous optimization, the system can continuously improve the accuracy and reliability of disaster simulation operations.

[0037] Through the collaborative work of the aforementioned model building module, information acquisition module, anomaly identification module, simulation execution module, verification output module, and system construction module, a disaster assessment model is built and dynamically optimized to identify early warning signs of disasters, execute verified disaster simulation operations, and finally output a reliable assessment result.

[0038] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for simulating gas accident disasters based on BIM and virtual interaction, characterized in that, Includes the following steps: Acquire real-time signals from electrical equipment, gas sensors, and temperature sensors within the area to be tested; When an abnormal signal is identified from the real-time acquired signal, execute: Constructing an assessment model includes: building a dataset based on known gas accidents; and extracting common information from the dataset to construct the assessment model. Based on the assessment model, a disaster simulation operation is performed to output the simulation assessment results; the simulation assessment results are verified to output a reliable assessment result. The common information of the data set includes information consistency and common dimension value. The information consistency is used to describe the pairwise similarity of parameter information among known gas accidents, and the common dimension value is the identification content whose information consistency is greater than or equal to a set threshold. The process of obtaining the information consistency includes: vectorizing the information in the dataset; and extracting the similarity by calculating the cosine of the angle between each pair of vectors, as the information consistency.

2. The gas accident disaster simulation method based on BIM and virtual interaction according to claim 1, characterized in that, The identification of abnormal signals includes: determining the critical point signal in the real-time acquired signal and labeling the critical point signal as the abnormal signal; labeling the sampling point signal in the real-time acquired signal as a normal signal, and obtaining the lower limit and upper limit of the fluctuation of the detection parameter based on the statistical analysis of the normal signal.

3. The gas accident disaster simulation method based on BIM and virtual interaction according to claim 1, characterized in that, The disaster simulation operation includes: determining the assessment node based on the equipment level; and loading the corresponding assessment model under the assessment node to perform the disaster simulation operation.

4. The gas accident disaster simulation method based on BIM and virtual interaction according to claim 3, characterized in that, The equipment levels include Level 1 equipment, Level 2 equipment, and Level 3 equipment. Level 1 equipment is the equipment with the highest risk level, Level 2 equipment is the equipment that causes current fluctuations due to gas concentration fluctuations, and Level 3 equipment is the equipment with the lowest risk level.

5. The gas accident disaster simulation method based on BIM and virtual interaction according to claim 1, characterized in that, The verification of the simulation assessment results includes: obtaining benchmark parameters representing the actual risk status of the area to be tested; generating an assessment offset value based on the benchmark parameters and the simulation assessment results; and determining the reliability of the simulation assessment results based on the assessment offset value.

6. A gas accident disaster simulation system based on BIM and virtual interaction, characterized in that, Includes the following modules: The information acquisition module is used to acquire real-time signals from electrical equipment, gas sensors, and temperature sensors in the area under test. An anomaly detection module is used to identify abnormal signals from the real-time acquired signals; The simulation execution module is used to perform disaster simulation operations based on the evaluation model in response to the anomaly identification module's identification of the anomaly signal, so as to generate simulation evaluation results; The verification output module is used to verify the simulation evaluation results in order to output a reliable evaluation result. The system construction module is used to optimize subsequent disaster simulation operations based on the credible assessment results.

7. A gas accident disaster simulation system based on BIM and virtual interaction according to claim 6, characterized in that, The system also includes a model building module, which is used to build the evaluation model by extracting common information from a dataset based on known gas accidents.

8. A gas accident disaster simulation system based on BIM and virtual interaction according to claim 7, characterized in that, The model building module extracts common information from the dataset constructed based on known gas accidents, including information consistency and common dimension values; The process of obtaining the consistency of the information includes: vectorizing the information in the data set; And similarity is extracted by calculating the cosine of the angle between each pair of vectors, which is used as the consistency of the information.

9. A gas accident disaster simulation system based on BIM and virtual interaction according to claim 6, characterized in that, The simulation execution module is specifically used to: determine the evaluation node based on the equipment level associated with the electrical equipment in the area to be tested; and load the corresponding evaluation model under the evaluation node.

10. A gas accident disaster simulation system based on BIM and virtual interaction according to claim 9, characterized in that, The equipment levels include Level 1 equipment, Level 2 equipment, and Level 3 equipment. Level 1 equipment is the equipment with the highest risk level, Level 2 equipment is the equipment that causes current fluctuations due to gas concentration fluctuations, and Level 3 equipment is the equipment with the lowest risk level.

Images