Fire-fighting real smoke and real fire training simulation system based on rule driving

Abstract

The invention relates to the technical field of fire safety and simulation training, and discloses a rule-driven fire-fighting real smoke and real fire training simulation system, which comprises a data acquisition and modeling module, a rule driving module, a rendering output module and a training evaluation module. The data acquisition and modeling module is used for acquiring initial fire data and constructing a reference model of smoke diffusion and fire spreading; the rule driving module simulates the non-synchronous evolution process of the fire through a particle simulation algorithm, a time coordination rule and a feedback loop algorithm; the rendering output module is used for integrating the environment rendering data and the multi-factor propulsion mode and generating a training effect evaluation index; and the evaluation optimization module judges whether the simulation precision of the reference model reaches a preset standard or not according to the training effect evaluation index until a simulation result meeting a precision requirement is obtained. According to the invention, the reality and accuracy of fire-fighting simulation training are improved.

Description

Technical Field

[0001] This application relates to the field of fire safety and simulation training technology, and in particular to a rule-driven fire simulation system based on real smoke and fire, which can be widely used in fire simulation, virtual reality and emergency drills. Background Technology

[0002] With the acceleration of urbanization and the increasing complexity of building structures, the risk of fire is increasing. The occurrence of fire is often accompanied by a complex evolution such as rapid smoke diffusion, multi-stage fire spread, and structural damage. The casualties and property losses caused by fire accidents remain high, making it a major disaster in the field of urban public safety. Fire training is a key link in improving the emergency response and practical handling capabilities of firefighters.

[0003] However, existing fire training methods mainly include two categories: real-world training and virtual simulation, both of which have significant shortcomings. Real-world training typically creates a near-fire environment by building training facilities and using open flames and smoke generators. While this method offers a degree of intuitiveness, it suffers from high safety risks, poor scenario adaptability, and high costs. Computer-based virtual simulation technology has been applied in fire training in recent years, capable of recreating fire scenarios through fire modeling and animation rendering. However, most existing virtual simulation systems suffer from simplified evolution mechanisms, insufficient realism, and a lack of evaluation systems. Existing training simulation technologies are deficient in realism, safety, dynamism, and evaluability, making it difficult to meet the demands of modern fire training for complex fire scenarios.

[0004] Therefore, this invention proposes a rule-driven fire-fighting real smoke and fire training simulation system, which can dynamically model and simulate the evolution of fire through a rule-driven approach while ensuring safety. It combines the interactive processes of smoke diffusion, fire spread and structural damage to generate an immersive real smoke and fire training scenario and provides quantifiable training effect evaluation indicators, significantly improving the scientificity and effectiveness of fire training. Summary of the Invention

[0005] To address the aforementioned technical issues, this application provides a rule-driven fire-fighting smoke and fire simulation system, offering a safe, realistic, and quantifiable immersive training environment with real smoke and fire for fire training.

[0006] Firstly, this application provides a rule-driven fire-fighting smoke and fire simulation system, comprising:

[0007] The data acquisition and modeling module is used to acquire initial fire data and build a benchmark model that characterizes the initial state of smoke diffusion and fire spread through multi-factor interactions.

[0008] The rule-driven module is used to receive the baseline model transmitted by the data acquisition and modeling module, and obtain a coordinated multi-factor advancement mode through multiple built-in rule-driven units, and generate a stable fire and smoke interaction model after iterative optimization.

[0009] The rendering output module is used to integrate the environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators.

[0010] The evaluation and optimization module is used to determine whether the simulation accuracy of the benchmark model reaches the preset accuracy standard based on the training effect evaluation index, until the asynchronous evolution simulation result that meets the preset accuracy requirement is obtained.

[0011] Secondly, this application provides a rule-driven fire simulation method for live smoke and fire training, the method comprising:

[0012] Step S1: Obtain initial fire data and construct a benchmark model representing the multi-factor interaction of the initial state of smoke diffusion and fire spread;

[0013] Step S2: Receive the baseline model transmitted by the data acquisition and modeling module, and obtain the coordinated multi-factor advancement mode through multiple built-in rule-driven units. After iterative optimization, generate a stable fire and smoke interaction model.

[0014] Step S3: Integrate the environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators;

[0015] Step S4: Based on the training effect evaluation index, determine whether the simulation accuracy of the benchmark model reaches the preset accuracy standard, until the asynchronous evolution simulation result that meets the preset accuracy requirement is obtained.

[0016] A third aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the methods described above.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] This invention provides a rule-driven fire simulation training system based on real smoke and fire. By integrating modules for data acquisition and modeling, rule-driven operation, rendering output, and evaluation optimization, it effectively solves the problems of insufficient realism, dynamism, and weak evaluation capabilities in existing fire training systems. The data acquisition and modeling module acquires the location of the ignition point and environmental variables in complex scenarios in real time through a sensor network, and constructs a baseline model of smoke diffusion and fire spread based on combustion characteristics, providing accurate initial data for subsequent simulations. The rule-driven module simulates the asynchronous evolution of a fire using particle simulation algorithms, time coordination rules, structural damage analysis, and feedback loop algorithms, ensuring that the expansion of smoke and fire remains coordinated over time while accurately simulating structural damage. The rendering output module integrates environmental rendering data and progression modes to generate training effect evaluation indicators, ensuring simulation accuracy and realism. The evaluation optimization module continuously optimizes the baseline model based on the deviation between the simulation results and the real scene, improving the accuracy of the fire simulation until a preset accuracy standard is reached. This invention enables high-precision fire simulation in a safe virtual environment through the collaborative work of multiple modules, enhancing the realism and effectiveness of the training process, improving firefighters' emergency response capabilities in complex fire scenarios, and providing data support and decision-making basis for the formulation of fire emergency plans. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the structure of the rule-driven fire-fighting real smoke and fire training simulation system in the embodiments of this application.

[0021] Figure 2 This is a flowchart of the rule-driven fire-fighting real smoke and fire training simulation method in the embodiments of this application; Detailed Implementation

[0022] This application provides a rule-driven fire simulation system for live smoke and fire training. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0023] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 The structural diagram of the rule-driven fire-fighting smoke and fire simulation system in this application includes:

[0024] The data acquisition and modeling module is used to acquire initial fire data and construct a benchmark model that characterizes the initial state of smoke diffusion and fire spread through multi-factor interaction. The initial fire data includes the location of the ignition point and environmental variables in the complex scene collected by the sensor network. The environmental variables include temperature, humidity, wind speed, and combustible material distribution. The data acquisition and modeling module constructs a benchmark model by analyzing environmental variables and combustion characteristics through multi-factor interaction to determine the initial propagation state of smoke and fire. The combustion characteristics include parameters that characterize the combustion behavior of combustible materials obtained from the material property database. The parameters include heat release rate, combustion rate, and ignition temperature.

[0025] Specifically, the data acquisition and modeling module uses data collected in real time by a sensor array, combined with the three-dimensional coordinates of the ignition point in the scene, and simultaneously acquires various environmental parameters in the scene to form a set of initial fire data for fire simulation. For example, in a fire training scenario inside a building, the location sensor records the coordinates of the ignition point, the temperature sensor collects the temperature, the humidity sensor collects the humidity, the wind speed sensor collects the wind speed, and the combustible material distribution sensor measures the spatial uniformity density. These parameters are analyzed through multi-factor interaction to form the initial concentration data of smoke diffusion, which is used to build the baseline model.

[0026] The data acquisition and modeling module further interacts with a material property database to obtain combustion characteristics related to the ignition point material. These characteristics include parameters such as heat release rate, combustion rate, and ignition temperature. For example, in a high-rise building scenario, the database stores the combustion rate, heat release rate, and ignition temperature of wood materials. The system extracts the corresponding values ​​based on the type of ignition point material and uses them as input to the model.

[0027] The data acquisition and modeling module inputs the ignition point location, environmental variables, and combustion characteristics into the baseline model's construction equations, forming an initial state representation of smoke diffusion and fire spread. The baseline model calculates the initial propagation behavior of smoke and fire through multi-factor interaction analysis, considering factors such as the effect of temperature on humidity changes, the influence of wind speed on smoke diffusion direction, and the control of combustible material distribution on the fire path. In the warehouse scenario, due to the uneven distribution of combustible materials, the baseline model iteratively analyzes and concludes that the initial fire propagation speed increases with rising temperature, generating a predicted smoke coverage radius of 10m within 5 seconds. In the outdoor forest scenario, wind speed is the dominant variable, and a wind direction factor is added to the baseline model to simulate the eastward propagation behavior of smoke at a speed of 2m / s.

[0028] Through the above implementation methods, the data acquisition and modeling module can dynamically construct a benchmark model in complex training scenarios, ensuring the accuracy and reliability of the initial propagation state of smoke and fire. This effectively solves the technical problems of incomplete initial condition settings and large deviations between simulation results and real fires in traditional training methods, thereby improving the applicability and simulation accuracy of the rule-driven fire-fighting real smoke and fire training simulation system in different scenarios.

[0029] The rule-driven module receives the baseline model transmitted by the data acquisition and modeling module, and obtains a coordinated multi-factor advancement mode through multiple built-in rule-driven units. After iterative optimization, a stable fire and smoke interaction model is generated. Specifically, it includes: a smoke simulation unit, which, based on the baseline model, uses a particle simulation algorithm to simulate the path of smoke spreading outward from the ignition point, calculates the range of smoke diffusion in space, and determines the smoke-covered spatial area. The particle simulation algorithm simulates the diffusion behavior of smoke in complex environments through particle motion trajectories and collision rules. The spatial area is used for subsequent adjustment of the fire spread sequence; and a fire control unit, based on the smoke coverage... The system covers a spatial region and adjusts the timing of fire spread through time coordination rules to ensure that the dynamic evolution of fire spread and smoke diffusion is consistent, thus obtaining the evolution sequence of the fire in the time dimension. The structural damage analysis unit is used to analyze the stress distribution in the fire-affected area based on the fire evolution sequence, using a grid division method to extract damage trigger points and obtain the spatiotemporal coordinates of damage events. The feedback optimization unit is used to extract parameters from the spatiotemporal coordinates of damage events, analyze the temporal deviation of parameters using a feedback loop algorithm, and adjust the fire spread speed according to preset optimization rules to determine the coordinated multi-factor propagation mode and generate a stable baseline model of fire and smoke interaction.

[0030] Specifically, the rule-driven module receives the baseline model transmitted by the data acquisition and modeling module. Through multiple built-in rule-driven units, it dynamically simulates the asynchronous evolution of a fire, addressing the disconnect, temporal inconsistencies, and model distortion issues present in existing fire training simulations regarding the evolution of smoke, fire intensity, and structural damage. This module includes a smoke simulation unit, a fire control unit, a structural damage analysis unit, and a feedback optimization unit. These units collaborate to achieve dynamic evolution modeling of the entire fire process, collectively forming a multi-factor coupled simulation core based on physical rules. The smoke simulation unit, based on the ignition point location and environmental variables in the baseline model, uses a particle simulation algorithm to simulate the smoke diffusion process. By calculating particle trajectories and collision rules with scene obstacles, it determines the real-time coverage of smoke in three-dimensional space. Each particle represents a smoke microparticle, released from the ignition point. Propulsion is based on environmental variables such as wind force, and the particle position update formula is based on the Euler integral method, accumulating velocity multiplied by time. During the simulation, when particles encounter obstacles, collision rules are applied for reflection or absorption, ensuring the realism of smoke diffusion behavior. In complex scenarios, such as multi-story buildings or outdoor fires, particle simulation algorithms can handle the diffusion behavior of smoke according to the specific environmental conditions. For example, in indoor fire scenarios, the diffusion path of smoke is affected by thermal convection, expands upward, and collides with obstacles such as furniture and walls, changing direction. This enables accurate path simulation, accurately characterizes the diffusion behavior of smoke in complex environments, and overcomes the technical problems of simplification and range distortion in traditional methods.

[0031] The fire control unit, based on the spatial coverage information output by the smoke simulation unit, dynamically adjusts the fire spread sequence through time coordination rules. Specifically, it monitors the spatiotemporal distribution data of smoke, determines whether it has reached a predefined logical area, and then activates the fire enhancement stage based on the smoke arrival time delay or advance, generating a fire evolution sequence that is out of sync with the smoke diffusion process. This effectively achieves asynchronous and coordinated evolution of fire and smoke, solving the realism problem caused by the mismatch between the fire spread and smoke diffusion sequence in existing simulations. The structural damage analysis unit, based on the fire evolution sequence, uses a mesh generation method to spatially discretize the fire-affected area and calculates the thermodynamic stress distribution of structural components based on finite element analysis. By comparing with material strength thresholds, it determines the damage trigger point and generates spatiotemporal coordinates that identify the location and time of damage, thereby accurately quantifying the spatiotemporal evolution of structural damage under fire conditions and overcoming simulation biases caused by the neglect or simplification of the structural damage process in the original simulation system. The feedback optimization unit extracts the time difference parameter from the spatiotemporal coordinates of the damage event and uses a feedback loop algorithm to analyze the time deviation between the time difference parameter and the time when the smoke spreads to that coordinate. If the deviation exceeds a preset threshold, the fire spread speed parameter is automatically adjusted. Through iterative optimization, the interaction behavior between the fire and the smoke tends to stabilize, generating a benchmark model that can reproduce the coupling effect of multiple factors. This improves the consistency and reliability of the overall system simulation and solves the problems of error accumulation and model divergence caused by the lack of a feedback mechanism in traditional simulations.

[0032] Through the synergistic effect of the aforementioned rule-driven units, the system can achieve dynamic coupling of multiple factors such as smoke diffusion, fire spread, and structural damage, ensuring that the simulation results of the fire evolution process are closer to the real scene in both time and space dimensions. It can achieve high-fidelity dynamic asynchronous evolution simulation of fire in a virtual training environment, effectively solving the problems of static fire evolution process, insufficient interaction, and large prediction accuracy deviation in existing technologies. This improves the simulation realism and training applicability of the rule-driven fire training simulation system, providing a reliable environmental foundation and decision support for fire training.

[0033] The smoke simulation unit is configured to use a particle simulation algorithm to simulate the diffusion behavior of smoke in a complex environment through the motion trajectory and collision rules of particles. The spatial area is used for subsequent adjustment of the fire spread sequence.

[0034] The time coordination rules in the fire control unit are configured as follows: monitor the spatiotemporal distribution data of smoke diffusion output by the smoke simulation unit; based on the spatiotemporal distribution data, determine whether the smoke covers the preset area; if so, dynamically adjust the phased enhancement of the fire spread model according to the time when the smoke arrives at the preset area; through dynamic adjustment, generate a staggered fire evolution sequence that is not synchronized with the smoke diffusion in the time dimension.

[0035] Specifically, the smoke simulation unit simulates the path of smoke spreading outward from the ignition point using a particle simulation algorithm. It calculates the smoke diffusion range based on environmental variables such as temperature and wind speed, and characterizes the smoke diffusion behavior in complex scenarios through particle trajectories and collision rules. This spatiotemporal data of smoke diffusion is monitored and analyzed in real time through a time coordination mechanism to determine whether the smoke has covered a preset area. If the smoke has covered the preset area, the system activates a phased enhancement of the fire spread model through the time coordination mechanism. At this time, the time coordination mechanism adjusts the activation sequence of fire spread based on the time it takes for the smoke to reach the area. For example, if the smoke arrives at a certain area later than a predetermined threshold, the system delays the activation sequence of fire spread; conversely, if the smoke arrives earlier, fire spread is activated earlier. Through this dynamic adjustment, the enhancement process of fire spread can be out of sync with the smoke diffusion process, simulating the asynchronous evolution of smoke and fire in a fire.

[0036] In specific application scenarios, such as high-rise building fires, where smoke spreads from lower to upper floors, the system dynamically adjusts the timing of fire spread based on the time it takes for the smoke to reach the top floor. For example, if smoke covers a passageway within 5 minutes, the fire will begin to intensify in the 6th minute, ensuring that the fire spread process matches the time difference in smoke diffusion. In warehouse fire scenarios, the smoke diffusion path is affected by the stacking of goods, resulting in irregular smoke diffusion. The system adjusts the timing of fire spread based on the actual path of smoke diffusion; for instance, as smoke spreads to the shelving area, the fire spread speed is correspondingly delayed, thus simulating a more realistic dynamic evolution of a fire.

[0037] By adjusting the timing of fire spread, the errors caused by the synchronous expansion of fire and smoke in traditional fire simulation methods are avoided. This ensures the asynchronous nature of fire spread and smoke diffusion during fire simulation, improves the spatiotemporal coordination and prediction accuracy of the simulation, and solves the technical problem of inconsistency between fire and smoke diffusion processes in fire simulation. It has good application effects and practical value.

[0038] The structural damage analysis unit is configured to: use a mesh generation method that calculates the force distribution through finite element analysis, determine points that exceed the material strength threshold from the force distribution characteristics as damage trigger points, and generate spatiotemporal coordinates that characterize the time and spatial location of structural damage.

[0039] Specifically, the structural damage analysis unit spatially divides the fire-affected area using a gridding method based on the fire's evolution sequence over time, ensuring that the affected area at each time point along the fire's spread path is precisely divided into uniform grid cells. This gridding allows for more detailed analysis and simulation of the fire's affected area, providing accurate data support for subsequent structural damage analysis.

[0040] The size of each grid cell can be flexibly set according to the complexity of the scene. This division method helps to further extract stress distribution characteristics. Within each grid cell, the corresponding thermal stress and mechanical load values ​​are calculated to obtain the stress distribution matrix. Based on the stress distribution matrix, the structural damage analysis unit can determine the stress situation of each grid cell, thereby identifying points that exceed the material strength threshold as potential damage trigger points. For example, in the fire simulation of concrete structures, when the thermal stress at a point exceeds the pressure threshold, that point will be marked as a damage trigger point. By using this threshold comparison method, potential damage locations can be identified early, improving the accuracy of fire simulation.

[0041] Furthermore, the structural damage analysis unit simulates the dynamic evolution of the structure based on the stress conditions at each trigger point, and gradually updates the damage state of the structure using a time-stepping method. At regular time steps, the stress conditions are reassessed to determine if any new damage has emerged. For example, if the stress gradient of the mesh around the trigger point increases during the gradual deformation and crack propagation process, the system will classify it as progressive damage and further advance the simulation of structural damage. This process can track the dynamic changes in structural damage during a fire in real time, ensuring that the simulated fire evolution in the training system is highly realistic.

[0042] After completing the stress analysis and damage assessment, the structural damage analysis unit transforms the damage events into spatiotemporal coordinates. These coordinates include the timestamps and spatial locations of the events, accurately characterizing the time and spatial location of structural damage. These spatiotemporal coordinates can be used for further fire damage assessment or emergency response optimization, ensuring that the specific location and time of occurrence of each structural damage point can be clearly seen during training.

[0043] By employing finite element analysis and mesh generation methods, this invention provides a refined model of the stress and damage to building structures in fire simulations. This addresses the inaccurate or overly simplistic handling of structural damage in traditional fire simulations, significantly improving the accuracy and reliability of damage prediction. By dynamically tracking the stress state and evolution of structures during a fire, this invention provides a scientific basis for real-time simulation of structural damage, effectively enhancing the realism and emergency response capabilities of fire training systems.

[0044] The feedback optimization unit is configured to: analyze the time difference between the spatiotemporal coordinates of the damage event and the time difference between the smoke diffusion to the spatiotemporal coordinates through a feedback loop algorithm, determine whether the time difference exceeds a preset threshold, and if it does, adjust the speed of fire spread, determine the coordinated multi-factor advancement mode, and generate a stable fire and smoke interaction model by iteratively optimizing the adjustment range through the feedback loop algorithm.

[0045] Specifically, the feedback loop algorithm extracts parameter control variables from the spatiotemporal coordinates of the damage event. These variables include fire intensity and smoke concentration thresholds. The extraction of parameters relies on the force distribution characteristics obtained from the mesh generation method in the baseline model. By calculating the time difference between the occurrence of the damage event and the arrival of smoke at that coordinate, the feedback loop algorithm can dynamically determine whether the time difference exceeds a preset threshold. If the time difference exceeds the preset threshold, the fire spread rate is adjusted. By adjusting the fire spread rate, the timing of fire and smoke diffusion is ensured to be consistent, avoiding distortion of simulation results due to excessively fast or slow fire spread.

[0046] The adjustment process involves reducing the fire spread rate based on the time difference and adjusting the activation sequence of the fire model through a phased enhancement parameter introduced by a time coordination mechanism, ensuring that the adjusted rate is consistent with the smoke coverage. During iterative optimization, the algorithm repeatedly evaluates the adjustment magnitude until the model stabilizes. The generated fire and smoke interaction model can be used for subsequent iterative updates of the baseline model, ensuring model stability and high accuracy. In high-rise building fire scenarios, if smoke diffusion is obstructed, the feedback loop algorithm can adjust the fire spread rate to synchronize fire and smoke diffusion, preventing premature fire activation or excessively slow fire expansion, thereby improving simulation accuracy.

[0047] Through the iterative optimization process of the feedback loop algorithm, the problem of asynchronous fire spread and smoke diffusion in traditional fire simulation is solved, which significantly improves the accuracy and stability of the simulation. Especially in complex and dynamically changing fire scenarios, it can provide more reliable training data support, optimize emergency response strategies, and improve the training effect and emergency response capabilities of firefighters.

[0048] The rendering output module is used to integrate environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators. The configuration is as follows: based on the coordinated multi-factor advancement mode, the environmental rendering data of the real scene is obtained, and the multi-factor advancement mode of the fire output by the rule-driven module is integrated with the environmental rendering data and the fire multi-factor advancement mode output by the rule-driven module using 3D rendering to generate training effect evaluation indicators. The environmental rendering data includes lighting, material and scene geometry information.

[0049] Specifically, the environmental rendering data includes lighting, material, and scene geometry information, ensuring it matches the spatiotemporal coordinates of the progression mode to guarantee high accuracy and realism in the fire simulation. In the implementation process, the rendering output module uses ray tracing to calculate light intensity and attenuation coefficients, simulating the reaction of smoke and fire under different environments. For example, in a high-density building fire simulation scenario, the module calculates the initial smoke diffusion path and simulates light attenuation based on the scene's light intensity distribution, surface reflectivity, and geometric structure. Based on this, the lighting information is combined with the smoke diffusion path to calculate the visibility of smoke within a specific area and the smoke coverage area.

[0050] By integrating this data, the rendering output module generates training performance evaluation metrics, including changes in light attenuation coefficients, smoke visibility, and fire spread area. These metrics quantify the differences between the simulation and real-world scenarios, guiding iterative updates to the baseline model. In indoor fire simulations, the rendering output module pays particular attention to the effects of lighting changes and object materials. For example, during a simulated fire, the interaction between indoor lighting and smoke affects smoke visibility. Through detailed modeling of ambient lighting and materials, the module provides an assessment of adaptability to low-light environments, improving simulation reliability. In outdoor forest fire scenarios, the module also considers the effects of changing sunlight angles and geometric obstacles like trees, calculating the propagation of smoke and fire under different lighting conditions, and optimizing shadow and lighting simulations to make the simulation results closer to reality. In complex fire scenarios such as those involving high-rise buildings, the rendering output module adjusts lighting, materials, and geometry based on the complexity of the actual scene to optimize the simulation of the dynamic evolution of fire and smoke, ensuring the accuracy of the simulation training. This method effectively solves the problems caused by insufficient lighting, material and geometric information in traditional simulations, improves the realism and adaptability of the simulation, and provides a high-precision training environment for fire training, which helps to improve training effectiveness and emergency response capabilities.

[0051] The evaluation and optimization module is used to determine whether the simulation accuracy of the baseline model meets the preset accuracy standard based on the training effect evaluation index, until an asynchronous evolution simulation result that meets the preset accuracy requirement is obtained. The configuration is as follows: based on the training effect evaluation index, by comparing the deviation between the simulation result and the real scene, it determines whether the simulation accuracy of the baseline model is less than the preset accuracy threshold. If so, the parameters related to smoke diffusion and fire spread in the baseline model are adjusted, and the particle simulation algorithm, time coordination rule, and feedback loop algorithm in the rule-driven module are re-executed to generate updated asynchronous evolution simulation results. This process is repeated until the simulation accuracy is greater than or equal to the preset accuracy threshold. The asynchronous evolution simulation result represents the dynamic interaction process of smoke diffusion, fire spread, and structural damage.

[0052] Specifically, the evaluation and optimization module determines whether the simulation accuracy meets preset accuracy standards by comparing the deviation values ​​with those of the actual scenario. For example, if the deviation values ​​of the fire spread rate or smoke diffusion path in the simulation results exceed the allowable range, the evaluation and optimization module will automatically adjust the relevant parameters in the baseline model. Specific operations include increasing the particle density in the particle simulation to improve the precision of smoke diffusion, or adjusting the combustion rate of the fire spread to make the fire and smoke expansion process more closely resemble reality. In high-rise building fire simulation scenarios, if there are deviations in simulation accuracy, such as the simulated fire spreading too quickly or the smoke diffusion being uneven, the evaluation and optimization module will adjust the relevant parameters based on the differences in fire spread and smoke diffusion in the simulation results and re-execute the particle simulation and feedback loop algorithm.

[0053] The evaluation and optimization module continuously optimizes the baseline model through real-time feedback iteration. When the spatiotemporal differences between simulated smoke diffusion and fire spread exceed a threshold, the feedback loop algorithm adjusts the initiation sequence of fire spread according to a time coordination mechanism to prevent fire spread from occurring too early or too late, thus improving the spatiotemporal consistency of the simulation. In complex fire scenarios, such as warehouse fires, the evaluation and optimization module not only optimizes parameters based on specific environmental variables but also improves simulation accuracy by adjusting the feedback mechanism in the model, ensuring that the simulation results differ minimally from the actual scenario.

[0054] Through this continuous optimization process, the evaluation and optimization module can precisely control the dynamic interaction between fire, smoke, and structural damage in the simulation, ensuring that the simulation results meet high accuracy requirements. After each optimization iteration, the baseline model is updated based on the actual training results to obtain more accurate simulation results of asynchronous evolution. This method effectively solves the problems of inconsistent timing and asynchronous fire spread and smoke diffusion in traditional simulations, significantly improving the realism and reliability of the simulation.

[0055] This invention also provides a rule-driven fire simulation method for live smoke and fire training, implemented through the aforementioned system, such as... Figure 2 As shown, the method includes:

[0056] Step S1: Obtain initial fire data and construct a benchmark model representing the multi-factor interaction of the initial state of smoke diffusion and fire spread.

[0057] Step S2: Receive the baseline model transmitted by the data acquisition and modeling module, and obtain the coordinated multi-factor advancement mode through multiple built-in rule-driven units. After iterative optimization, generate a stable fire and smoke interaction model.

[0058] Step S3: Integrate the environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators.

[0059] Step S4: Based on the training effect evaluation index, determine whether the simulation accuracy of the benchmark model has reached the preset accuracy standard, until the asynchronous evolution simulation results that meet the preset accuracy requirements are obtained.

[0060] This application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a rule-driven fire simulation method for real smoke and fire training.

[0061] In summary, this application effectively addresses the shortcomings of existing fire training systems, such as insufficient realism, dynamism, and weak evaluation capabilities, by integrating modules for data acquisition and modeling, rule-driven simulation, rendering output, and evaluation optimization. The data acquisition and modeling module uses a sensor network to acquire the location of the ignition point and environmental variables in complex scenarios in real time, and combines this with combustion characteristics to construct a baseline model of smoke diffusion and fire spread, providing accurate initial data for subsequent simulations. The rule-driven module uses particle simulation algorithms, time-coordination rules, structural damage analysis, and feedback loop algorithms to simulate the asynchronous evolution of a fire, ensuring that the expansion of smoke and fire remains coordinated over time, while accurately simulating structural damage. The rendering output module integrates environmental rendering data and progression patterns to generate training effect evaluation indicators, ensuring simulation accuracy and realism. The evaluation optimization module continuously optimizes the baseline model based on the deviation between the simulation results and the real scene, improving the accuracy of fire simulation until a preset accuracy standard is reached. This invention enables high-precision fire simulation in a safe virtual environment through the collaborative work of multiple modules, enhancing the realism and effectiveness of the training process, improving firefighters' emergency response capabilities in complex fire scenarios, and providing data support and decision-making basis for the formulation of fire emergency plans.

[0062] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0063] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0064] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A rule-driven fire-fighting smoke and fire simulation system, characterized in that, The system includes: The data acquisition and modeling module is used to acquire initial fire data and build a benchmark model that characterizes the initial state of smoke diffusion and fire spread through multi-factor interactions. The rule-driven module is used to receive the baseline model transmitted by the data acquisition and modeling module, and obtain a coordinated multi-factor advancement mode through multiple built-in rule-driven units, and generate a stable fire and smoke interaction model after iterative optimization. The rendering output module is used to integrate the environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators. The evaluation and optimization module is used to determine whether the simulation accuracy of the benchmark model reaches the preset accuracy standard based on the training effect evaluation index, until the asynchronous evolution simulation result that meets the preset accuracy requirement is obtained.

2. The system according to claim 1, characterized in that, The data acquisition and modeling module is configured as follows: The initial fire data includes the location of the ignition point and environmental variables in the complex scene collected by a sensor network, including temperature, humidity, wind speed and distribution of combustibles; The data acquisition and modeling module analyzes the environmental variables and combustion characteristics through multi-factor interaction to construct the benchmark model, which is used to determine the initial propagation state of smoke and fire. The combustion characteristics include parameters characterizing the combustion behavior of combustibles obtained from the material property database, including heat release rate, combustion rate, and ignition temperature.

3. The system according to claim 1, characterized in that, The rule-driven module includes: The smoke simulation unit, based on the baseline model, uses a particle simulation algorithm to simulate the path of smoke spreading outward from the ignition point, calculates the range of smoke gradually spreading in the spatial dimension, and determines the space area covered by smoke. The particle simulation algorithm simulates the diffusion behavior of smoke in a complex environment through the motion trajectory and collision rules of particles. The space area is used for subsequent adjustment of the fire spread sequence. The fire control unit, based on the smoke-covered spatial area, adjusts the timing of fire spread through time coordination rules to keep the dynamic evolution of fire spread and smoke diffusion consistent, thereby obtaining the evolution sequence of fire in the time dimension. The structural damage analysis unit is used to analyze the stress distribution in the fire-affected area based on the fire evolution sequence, using a grid division method, to extract damage trigger points and obtain the spatiotemporal coordinates of the damage event. The feedback optimization unit is used to extract parameters from the spatiotemporal coordinates of the damage event, analyze the temporal deviation of the parameters using a feedback loop algorithm, adjust the fire spread speed according to preset optimization rules, determine the coordinated multi-factor propulsion mode, and generate a stable benchmark model of fire and smoke interaction.

4. The system according to claim 3, characterized in that, The time coordination rule in the fire control unit is configured as follows: Monitor the spatiotemporal distribution data of smoke diffusion output by the smoke simulation unit; Based on the spatiotemporal distribution data, it is determined whether the smoke covers the preset area. If it is determined to be yes, the phased enhancement of the fire spread model is dynamically adjusted according to the time when the smoke arrives at the preset area. Through the aforementioned dynamic adjustment, a staggered fire evolution sequence that is out of sync with smoke diffusion in the time dimension is generated.

5. The system according to claim 3, characterized in that, The structural damage analysis unit is configured as follows: A mesh generation method based on finite element analysis to calculate the stress distribution is adopted. Points exceeding the material strength threshold are identified from the stress distribution characteristics as damage trigger points, and spatiotemporal coordinates characterizing the time and spatial location of structural damage are generated.

6. The system according to claim 3, characterized in that, The feedback optimization unit is configured as follows: The feedback loop algorithm analyzes the spatiotemporal coordinates of the damage event and the time difference between the smoke spreading to the spatiotemporal coordinates, and determines whether the time difference exceeds a preset threshold. If it does, the speed of fire spread is adjusted, and a coordinated multi-factor propulsion mode is determined. The feedback loop algorithm generates a stable fire and smoke interaction model by iteratively optimizing the adjustment range.

7. The system according to claim 1, characterized in that, The rendering output module is configured as follows: Based on the coordinated multi-factor advancement mode, real-world scene environment rendering data is obtained. 3D rendering is used to integrate the environment rendering data and the fire multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators. The environment rendering data includes lighting, material, and scene geometry information.

8. The system according to claim 1, characterized in that, The evaluation and optimization module is configured as follows: Based on the training effect evaluation index, by comparing the deviation between the simulation results and the real scene, it is determined whether the simulation accuracy of the benchmark model is less than a preset accuracy threshold. If so, the parameters related to smoke diffusion and fire spread in the baseline model are adjusted, and the particle simulation algorithm, time coordination rule, and feedback loop algorithm in the rule-driven module are re-executed to generate updated asynchronous evolution simulation results. The simulation results are compared cyclically until the simulation accuracy is greater than or equal to the preset accuracy threshold. The asynchronous evolution simulation results characterize the dynamic interaction process of smoke diffusion, fire spread, and structural damage.

9. A rule-driven fire-fighting smoke and fire simulation method, used to implement a rule-driven fire-fighting smoke and fire simulation system as described in any one of claims 1-8, characterized in that, The method includes: Step S1: Obtain initial fire data and construct a benchmark model representing the multi-factor interaction of the initial state of smoke diffusion and fire spread; Step S2: Receive the baseline model transmitted by the data acquisition and modeling module, and obtain the coordinated multi-factor advancement mode through multiple built-in rule-driven units. After iterative optimization, generate a stable fire and smoke interaction model. Step S3: Integrate the environmental rendering data with the multi-factor advancement mode output by the rule-driven module to generate training effect evaluation indicators; Step S4: Based on the training effect evaluation index, determine whether the simulation accuracy of the benchmark model reaches the preset accuracy standard, until the asynchronous evolution simulation result that meets the preset accuracy requirement is obtained.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in claim 9.

Images