A fire smoke monitoring and early warning method, system and storage medium in an industrial building

Abstract

This invention discloses a method, system, and storage medium for fire smoke monitoring and early warning in industrial buildings. It obtains building fire boundary conditions from a building information model and fire source parameters from a sensor system. Then, based on the building fire boundary conditions and fire source parameters, it uses RANS equations to obtain the fully diffused smoke state within the industrial building, acquires a three-dimensional velocity field, and calculates the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet within the building. If the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet is less than the design smoke exhaust volume, subsequent protective actions are not triggered; otherwise, the fused flow rate of each smoke exhaust outlet is calculated, and a preset fire protection command is triggered when the fused flow rate is greater than the design smoke exhaust volume. This method, through the two-dimensional cross-sectional simulated flow rate under various operating conditions and the judgment logic at each level of the fused flow rate, can accurately achieve fire smoke monitoring and alarm, effectively reducing the emergency actions of fire-fighting equipment and avoiding excessively high false alarm rates that could affect normal production activities.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method, system, and storage medium for fire smoke monitoring and early warning in industrial buildings. Background Technology

[0002] With the increasing automation of industry, more and more industrial buildings are beginning to plan the functions of their upper spaces in greater detail to improve land use efficiency. However, this also creates fire safety hazards in industrial buildings, including public buildings, where proprietary energy and power equipment is stored. For the manufacturing industry, due to changes in external demand, the stacking and storage of materials and products often change significantly. Traditional methods of relying on manual spot checks to ensure that industrial buildings meet fire safety and other design specifications are not suitable for preventing random fire hazards during logistics and production processes.

[0003] As industrial buildings become increasingly automated and unmanned, space utilization increases (leading to increased obstruction), equipment logic becomes more complex, and labor costs rise. For a period, the reduction in the number of workers or the more efficient allocation of worker energy to production activities can potentially weaken the fire detection capabilities of industrial buildings. On the other hand, although factories are becoming more information-driven and have deployed sensors with fire detection capabilities, industrial buildings currently lack the ability to infer smoke evolution and accurately monitor and warn of fire smoke because they lack real-time smoke simulation systems and connections to relevant IoT systems. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a method for monitoring and early warning of fire smoke in industrial buildings, comprising the following steps:

[0005] S1, obtain the building fire boundary conditions from the building information model and obtain the fire source parameters from the sensor system;

[0006] S2, based on the building fire boundary conditions and fire source parameters, obtain the smoke state that is fully diffused in the industrial building through the RANS equation, obtain the three-dimensional velocity field, and calculate the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet in the building.

[0007] S3. If the simulated flow rate of the two-dimensional cross-section of each smoke exhaust outlet is less than the designed smoke exhaust volume, the subsequent protection action will not be triggered; otherwise, the combined flow rate of each smoke exhaust outlet will be calculated.

[0008] S4. Determine if the combined flow rate is less than the design smoke exhaust rate. If it is less than the design smoke exhaust rate, the protection action will not be triggered; otherwise, the preset fire protection command will be triggered.

[0009] Preferably, step S1 includes: the building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions; and the fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

[0010] Preferably, step S2 specifically includes:

[0011] Set initial parameter x i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3, ..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set, according to the 3σ criterion, divides the side length into 3 parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field v j The simulated flow rate q of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. j , in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j = 1, 2, 3…3 n .

[0012] Preferably, step S3 specifically includes:

[0013] If the simulated flow rate q of the two-dimensional cross-section of all exhaust outlets j If all values ​​are less than the designed smoke exhaust volume, subsequent protective actions will not be triggered.

[0014] If there exists at least one smoke outlet with a two-dimensional cross-sectional simulated flow rate q j If the exhaust volume is greater than the design exhaust volume, then calculate the combined flow rate q at each exhaust outlet.

[0015] Where the coefficient a j =p j w j p j w is the reliability factor. j p represents the importance weight. j From the confusion matrix C j The confusion matrix C is calculated. j For a 3 n Binary classification problems within the identification framework, i.e., C j There are 3 n One, measuring 2*2: in This represents the probability that the j-th working condition is judged as not requiring fire protection. This represents the probability that the j-th working condition is judged to require fire prevention instead of not requiring it. This represents the probability that the j-th working condition is judged to be fireproof instead of requiring fireproofing. This represents the probability that the j-th working condition is judged as requiring fire prevention.

[0016] Preferably, step S1 specifically includes:

[0017] Create an image sample set of combustible and flammable explosive materials that need to be stored in industrial buildings, and train a YOLO network model;

[0018] The image captured by the camera is scaled up to the size required by YOLO and divided into an equally divided grid, where each grid is used to detect targets whose center point falls within that grid.

[0019] Each grid cell predicts a conditional probability value for each category of stacked objects, and generates B square bounding boxes that enclose the objects based on the grid. Each bounding box predicts five regression values ​​x. Y ,y Y ,w Y ,h Y ,c Y , where x Y ,y Y ,w Y ,h Y These represent the center coordinates, width, and height of the bounding box, respectively. Y The probability and location accuracy of the bounding box containing an object are represented; a coarse prediction box is obtained by filtering using the NMS method.

[0020] Extract the rough prediction bounding box image and convert it into a grayscale image. Use the Canny operator to extract its edges and obtain a binary image. Use the findContours function to obtain the target contour line outline. Optimize the contour line and draw the fitting contour. Measure the contour perimeter or filling area. Combine multiple camera calibration information to estimate the scale of the fire source accumulation at the scene.

[0021] This invention also discloses a fire smoke monitoring and early warning system for industrial buildings, comprising: a data acquisition module for acquiring building fire boundary conditions from a building information model and acquiring fire source parameters from a sensor system; a flow acquisition module for obtaining the fully diffused smoke state within the industrial building based on the building fire boundary conditions and fire source parameters using RANS equations, acquiring a three-dimensional velocity field, and calculating the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet within the building; a fused flow calculation module for calculating the fused flow rate of each smoke exhaust outlet when at least one of the two-dimensional cross-sectional simulated flow rates of each smoke exhaust outlet is greater than the design smoke exhaust rate; and a flow judgment module for judging whether the fused flow rate is less than the design smoke exhaust rate. If it is less than the design smoke exhaust rate, no protective action is triggered; otherwise, a preset fire protection command is triggered.

[0022] Preferably, the building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions, and the fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

[0023] Preferably, the traffic acquisition module is configured to set the initial parameter x. i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3, ..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set, according to the 3σ criterion, divides the side length into 3 parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field v j The simulated flow rate q of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. j , in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j = 1, 2, 3…3 n .

[0024] The present invention also discloses a fire smoke monitoring and early warning device for industrial buildings, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0025] The present invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the methods described above.

[0026] The fire smoke monitoring and early warning method disclosed in this embodiment for industrial buildings obtains building fire boundary conditions from a building information model and fire source parameters from a sensor system. Then, based on the building fire boundary conditions and fire source parameters, it obtains the fully diffused smoke state within the industrial building using RANS equations, acquires the three-dimensional velocity field, and calculates the two-dimensional cross-sectional simulated flow rate q of each smoke exhaust outlet within the building. j If the simulated flow rate of the two-dimensional cross-section at each smoke exhaust outlet is less than the design smoke exhaust volume, subsequent protective actions are not triggered. Otherwise, the combined flow rate q of each smoke exhaust outlet is calculated. If the combined flow rate q is less than the design smoke exhaust volume, protective actions are not triggered; otherwise, a preset fire protection command is triggered. This method utilizes sensor alarms and the simulated flow rate q of the two-dimensional cross-section under various operating conditions. j The system integrates judgment logic at three levels: fire, smoke, and q. This allows for accurate monitoring and alarm of fire smoke, effectively reducing the emergency actions of fire-fighting equipment and preventing excessively high false alarm rates from affecting normal production activities.

[0027] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0029] Figure 1 This is a schematic flowchart of a fire smoke monitoring and early warning method for industrial buildings disclosed in an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram illustrating the process of obtaining the scale of on-site fire source accumulation according to an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0033] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0034] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms “first,” “second,” and similar terms used in the specification and claims of this patent application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “a” and similar terms do not indicate a limitation of quantity, but rather indicate the presence of at least one.

[0035] Currently, the design industry generally believes that smoke exhaust vents occupy a lot of space, and it is assumed that when a fire breaks out in a large space, the smoke quickly reaches the top and then flows freely, so it is assumed that the smoke can be completely exhausted. Therefore, as long as design specifications are met or simple calculations are performed according to the large space procedure, the number of smoke exhaust vents is usually kept to a minimum. However, in actual operation, smoke exhaust vents often suffer from problems such as obstruction, blockage, and malfunction of active smoke extraction equipment. If such situations are compounded by the failure of fire protection facilities or their inability to operate according to rated requirements, the fire hazard is clearly present. Therefore, this patent aims to fully integrate IoT sensors, Building Information Modeling (BIM), Building Management System (WMS), and Random Access Array (RANS) numerical simulation for fire prediction and intervention. In this embodiment, due to the current technical situation and scenario requirements, RANS is simplified to a single-phase flow of smoke, without considering carbon smoke and other factors. This embodiment directly identifies fire smoke hazards and can also use the final smoke monitoring and early warning results to finely evaluate whether the number and layout of smoke exhaust vents in some types of industrial buildings, set according to the large space standard, are reasonable. Specifically, see attached... Figure 1 As shown in the attached figure, this embodiment discloses a method for monitoring and early warning of fire smoke in industrial buildings. Figure 1 As shown, the method specifically includes the following steps.

[0036] Step S1: Obtain the building fire boundary conditions from the building information model and obtain the fire source parameters from the sensor system.

[0037] Monitoring and assessing fire hazards begins with the preparation of relevant information. This starts with the IoT platform in industrial buildings, including cameras, temperature and electrical sensors installed on energy-powered and high-power equipment, and environmental sensors installed on building surfaces and pipelines. Camera video image analysis can identify flammable materials and their accumulation status; time-series analysis of equipment sensors can trigger alarms for high temperatures and high currents; and environmental sensors can monitor for leaks of gases, liquids, and dust. Additionally, the WMS (Building Management System) contains dimensional and structural information about the equipment within the industrial building, and currently, most of these are modeled in 3D.

[0038] In this embodiment, the building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions. The fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

[0039] What are the speed inlet parameters and their geometric dimensions? How, can it be pre-defined and directly extracted from the BIM structural information model? Passive air supply velocity. The airflow velocity of the active air supply can be directly collected from the deployed environmental sensors. It can be obtained directly or simply from the operating data of the equipment sensors. If the relevant sensors are lacking, information can be designed to replace them while ensuring the normal condition of the equipment and doors and windows.

[0040] What are the speed exit points and their geometric dimensions? How can the velocity field v be pre-defined and directly extracted from the BIM structural information model and calculated using RANS? j Multiply to obtain the estimated smoke volume Estimated exhaust conditions: exhaust velocity Acquisition methods and Similarly; make-up air speed Ideally, data can be collected, but estimation methods can also be used. Since general smoke control and exhaust standards require that the make-up air volume should not be less than 50% of the exhaust volume, it can be estimated based on the lower limit of the relationship with the exhaust volume.

[0041] Initial environmental conditions include the initial ambient temperature t0, which can be directly collected by a temperature sensor; the initial temperature of some heat-generating parts can be collected by a contact temperature sensor. The initial concentration of gases such as CO2 in the air can be obtained by an environmental gas sensor, or it can be assumed to be normal air.

[0042] In addition, the combustion process is simplified in this embodiment. It is assumed that the combustible material is fully combusted throughout the process. The intensity of the fire source P is mainly determined by parameters such as the combustion rate and calorific value. These are all related to the type of combustible material. The type of combustible material has been identified by the classification function of the camera image processing. The correspondence between the type and parameters such as calorific value can be established in the platform database.

[0043] The size of the fire source accumulation can be estimated using multiple cameras through the following specific steps. See attached diagram for details. Figure 2 As shown, the scale of flammable materials, i.e., on-site ignition sources, can be obtained through the following information.

[0044] Step S101: Create an image sample set of combustible and flammable explosive materials that need to be stored in industrial buildings, and train the YOLO network model.

[0045] Step S102: Scale the image captured by the camera to the size required by YOLO and divide it into equally divided grids, where each grid is used to detect targets whose center point falls within that grid.

[0046] Step S103: For each grid cell, predict a conditional probability value for each category of stacked objects, and generate B square bounding boxes that enclose the objects based on the grid cells. Each bounding box predicts five regression values ​​x. Y ,y Y ,w Y ,h Y ,c Y , where x Y ,y Y ,w Y ,h Y These represent the center coordinates, width, and height of the bounding box, respectively. Y The probability and location accuracy of the bounding box containing an object are represented; a coarse prediction box is obtained by filtering using the NMS method.

[0047] Step S104: Extract the rough prediction box image and convert it into a grayscale image. Use the Canny operator to extract its edges and obtain a binary image. Use the findContours function to obtain the target contour. Optimize the contour and draw the fitting contour. Measure the contour perimeter or filling area. Combine the calibration information from multiple cameras to estimate the scale of the fire source accumulation at the scene.

[0048] Step S2: Based on the building fire boundary conditions and fire source parameters, obtain the smoke state that is fully diffused in the industrial building through the RANS equation, acquire the three-dimensional velocity field, and calculate the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet in the building.

[0049] After preparing the data through the aforementioned step S1, the location of the fire hazard can be alerted by the IoT multimodal sensor, which continuously produces a certain amount of smoke. The smoke state that is fully diffused in the industrial building can be obtained through the RANS method, and a three-dimensional velocity field can be obtained.

[0050] In this embodiment, the initial parameter x is set. i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3, ..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set of variables is divided into three parts according to the 3σ criterion. If not enough candidate diffusion configurations are obtained, this can be extended to 5σ, dividing the side length into five parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field v j The simulated flow rate q of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. j , in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j = 1, 2, 3…3 n .

[0051] Where v j The velocity of each mesh, i.e., the infinitesimal element, is generated by RANS with the entire space as the mesh. It can be described by a three-dimensional coordinate system of xyz and follows the conservation law. Therefore, the following system of equations can be established.

[0052]

[0053] ρ is the density of the smoke fluid; t is time; u x u y u z Let denot u be the velocity components of the velocity vector in the x, y, and z directions. div is the vector divergence operator; μ is the dynamic viscosity; grad is the vector gradient operator; P is the pressure; S... x S y S z It is a generalized source term; by calculating u for each grid cell, the scalar distribution v of the velocity can be obtained. j .

[0054] Step S3: If the simulated flow rate of the two-dimensional cross-section of each smoke exhaust outlet is less than the designed smoke exhaust volume, the subsequent protection action will not be triggered; otherwise, the combined flow rate of each smoke exhaust outlet will be calculated.

[0055] In this embodiment, if all q j If all values ​​are less than the designed smoke exhaust volume, subsequent fire prevention actions will not be triggered. If q exists... j If the smoke exhaust volume exceeds the design exhaust volume, the calculation of the fusion flow rate q will be triggered.

[0056] The initial parameters input into RANS inevitably contain a certain amount of error, which will inevitably cause the simulation results to fluctuate and become unstable within a certain range. Therefore, the flow rate q will objectively have multiple simulation results. The problem then becomes how to merge these multiple flow rates q into one. Naturally, this requires designing or introducing an information fusion method. The information fusion method is used to stabilize the results of fluid computation simulation.

[0057] In this embodiment, step S3 may specifically include the following:

[0058] Step S31, if the simulated flow rate q of the two-dimensional cross-section of all smoke exhaust outlets j If all values ​​are less than the designed smoke exhaust volume, subsequent protective actions will not be triggered.

[0059] Step S32, if there is at least one smoke exhaust outlet with a two-dimensional cross-sectional simulated flow rate q j If the exhaust volume is greater than the design exhaust volume, then calculate the combined flow rate q at each exhaust outlet.

[0060] Where the coefficient a j =p j w j p j w is the reliability factor. j p represents the importance weight. j From the confusion matrix C j The confusion matrix C is calculated. j For a 3 n Binary classification problems within the identification framework, i.e., C j There are 3 n One, measuring 2*2: in This represents the probability that the j-th working condition is judged as not requiring fire protection. This represents the probability that the j-th working condition is judged to require fire prevention instead of not requiring it. This represents the probability that the j-th working condition is judged to be fireproof instead of requiring fireproofing. This represents the probability that the j-th working condition is judged as requiring fire prevention.

[0061] The importance weight w j This value is obtained from the output of a width learning system (BLS) and adaptively updated based on feedback. Specifically, assuming this value is recorded in the output node Y, Y can be calculated using the following system of equations:

[0062]

[0063] W1 is a randomly generated set of weights, and β1 is a threshold of 1. Input data X is transformed into a new set of mapped feature nodes Z under the action of the linear function φ(·). W2 is a randomly generated set of weights connecting the mapped nodes and the augmentation nodes, and β2 is a threshold of 2. Under the continuous transformation of the linear function ξ(·), the original input data completes a non-linear transformation from X to H. The feature mapping nodes Z and augmentation nodes H are combined into the hidden layer node A of the breadth learning model. W represents the connection weights. In this embodiment, since the distribution configuration obtained from RANS calculation is indeterminate, the number of nodes in BLS needs to remain flexible and scalable. The number of added nodes is denoted as m, the number of mapped feature nodes is set to n, the number of augmentation nodes is m, and the intermediate layer is A. m If A represents... m =[Z n |H m Adding a set of augmenting nodes to the original width learning network consisting of p nodes, the intermediate layer can be represented as A after adding the set of augmenting nodes. m+1 :

[0064] A m+1 =[A m |ξ(Z n W p +β p )]=[A m |H p ];

[0065] Accordingly, the intermediate layer connection weights become: In the formula, D and B are merely simplified mathematical symbols with no practical meaning. The calculation method is as follows:

[0066] D = (A m ) + ξ(Z n W p +β p )

[0067]

[0068]

[0069] C has the same properties as D and B, and the calculation method is as follows: C=ξ(Z) n Wp +β p )-A m D; Use A in problems involving node expansion. m+1 With W m+1 Updating A and W will give you Y, which means outputting w. j .

[0070] As mentioned earlier, the distinction between operating conditions is determined by the CFD input features x and σ, and has universality. The result is assigned by the evaluation function, and the prior probability P0 is obtained after statistical analysis of this assignment. In this embodiment, P0 is a historical experience of whether "fire protection is required" or "fire protection is not required". Where P j =P0C j BetP j It is P j After the Pignatic transformation, the Pignatic probability can be obtained.

[0071] Step S4: Determine whether the combined flow rate is less than the design smoke exhaust rate. If it is less than the design smoke exhaust rate, the protection action will not be triggered; otherwise, the preset fire protection command will be triggered.

[0072] For situations where the combined flow rate q is greater than the designed smoke exhaust volume, taking an automated warehouse in an industrial building as an example, robots can be controlled to perform fire prevention actions to eliminate fire hazards. The types of robots and the fire prevention operations they perform vary depending on the industrial building. This embodiment does not discuss specific fire prevention actions and related robot control issues. It can be assumed that they ultimately act according to a set of instructions A based on a certain strategy. The result of A is observed by the corresponding sensors under the IoT platform, forming an observation result. If this observation result is unsuccessful, i.e., the sensor observations have not yet returned to normal, the weight w is directly adjusted without going through an evaluation function. j The general principle is to assign a larger value to q. j For a more advanced consideration, the process continues until the sensor timing returns to normal operating range. If, after the robot's fire prevention actions, the sensors indicate that the fire hazard is rapidly worsening or even that a fire has already started, then the fire-fighting instruction set action B is initiated until the fire is extinguished.

[0073] If the environmental observation is successful, meaning the sensor observations return to normal, a positive incentive is given in the evaluation function, and adjustments are made based on the actual results. This case study uses, for illustration, the Q-learning method, which, in particular, involves constructing a Q-matrix where rows represent states and columns represent actions. If an action is successful, the corresponding matrix element changes from 0 to 100; if unsuccessful, 0 changes to -1; and if the fire is accelerated, it is recorded as -100. The feedback mechanism of the entire evaluation function has three levels: first, it directly feeds back to the agent, reinforcing its fire prevention and extinguishing actions and optimizing instruction sets A and B or their generation strategies; second, it optimizes the prior probability P0 needed for confusion matrix calculation, theoretically making p... j The statistical learning results are more accurate; thirdly, the samples that are ultimately marked as failures in fire prevention and fire extinguishing judgments and behaviors can be used to trace whether the smoke exhaust and fire protection designs have truly played their role and optimize related management; the conclusions on layout optimization formed will help to write more refined smoke exhaust and fire protection specifications for industrial buildings.

[0074] The fire smoke monitoring and early warning method disclosed in this embodiment for industrial buildings obtains building fire boundary conditions from a building information model and fire source parameters from a sensor system. Then, based on the building fire boundary conditions and fire source parameters, it obtains the fully diffused smoke state within the industrial building using RANS equations, acquires the three-dimensional velocity field, and calculates the two-dimensional cross-sectional simulated flow rate q of each smoke exhaust outlet within the building. j If the simulated flow rate of the two-dimensional cross-section at each smoke exhaust outlet is less than the design smoke exhaust volume, subsequent protective actions are not triggered. Otherwise, the combined flow rate q of each smoke exhaust outlet is calculated. If the combined flow rate q is less than the design smoke exhaust volume, protective actions are not triggered; otherwise, a preset fire protection command is triggered. This method utilizes sensor alarms and the simulated flow rate q of the two-dimensional cross-section under various operating conditions. j The system integrates judgment logic at three levels: fire, smoke, and q. This allows for accurate monitoring and alarm of fire smoke, effectively reducing the emergency actions of fire-fighting equipment and preventing excessively high false alarm rates from affecting normal production activities.

[0075] This invention also discloses a fire smoke monitoring and early warning system for industrial buildings, comprising: a data acquisition module for acquiring building fire boundary conditions from a building information model and acquiring fire source parameters from a sensor system; a flow acquisition module for obtaining the fully diffused smoke state within the industrial building based on the building fire boundary conditions and fire source parameters using RANS equations, acquiring a three-dimensional velocity field, and calculating the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet within the building; a fused flow calculation module for calculating the fused flow rate of each smoke exhaust outlet when at least one of the two-dimensional cross-sectional simulated flow rates of each smoke exhaust outlet is greater than the design smoke exhaust rate; and a flow judgment module for judging whether the fused flow rate is less than the design smoke exhaust rate. If it is less than the design smoke exhaust rate, no protective action is triggered; otherwise, a preset fire protection command is triggered.

[0076] In this embodiment, the building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions; the fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

[0077] In this embodiment, the traffic acquisition module is configured to set the initial parameter x. i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3, ..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set, according to the 3σ criterion, divides the side length into 3 parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field v j The simulated flow rate q of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. j , in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j = 1, 2, 3…3 n .

[0078] The specific functions of the fire smoke monitoring and early warning system in the aforementioned industrial building correspond one-to-one with the fire smoke monitoring and early warning method in the industrial building disclosed in the previous embodiments. Therefore, they will not be described in detail here. For details, please refer to the various embodiments of the fire smoke monitoring and early warning method in the industrial building disclosed in the previous embodiments. It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to mutually.

[0079] In other embodiments, a fire and smoke monitoring device for industrial buildings is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the various steps of the fire and smoke monitoring method for industrial buildings as described in the above embodiments.

[0080] The fire and smoke monitoring device in the industrial building may include, but is not limited to, a processor and a memory. The server may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the schematic diagram is merely an example of a server and does not constitute a limitation on the server device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the server device may also include input / output devices, network access devices, buses, etc.

[0081] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the server device, connecting various parts of the server device via various interfaces and lines.

[0082] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the server device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc. In addition, the memory may include high-speed random access memory and non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0083] If the fire smoke monitoring method in the industrial building 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, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

[0085] In summary, the above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should be covered by the present invention.

Claims

1. A method for monitoring and early warning of fire smoke in industrial buildings, characterized in that, Includes the following steps: S1, obtain the building fire boundary conditions from the building information model and obtain the fire source parameters from the sensor system; S2, based on the building fire boundary conditions and fire source parameters, obtain the smoke state that is fully diffused in the industrial building through the RANS equation, obtain the three-dimensional velocity field, and calculate the two-dimensional cross-sectional simulated flow rate of each smoke exhaust outlet in the building. Step S2 specifically includes: setting the initial parameter x i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set, according to the 3σ criterion, divides the side length into 3 parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field The simulated flow rate of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. , ,in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j=1,2,3…3 n ; S3, if the simulated flow rate of the two-dimensional cross-section of each smoke exhaust outlet is less than the designed smoke exhaust volume, then subsequent protection actions are not triggered; otherwise, the combined flow rate of each smoke exhaust outlet is calculated. Step S3 specifically includes: If the simulated flow rate of the two-dimensional cross-section of all exhaust outlets is... If all values ​​are less than the design smoke exhaust volume, subsequent protective actions will not be triggered; if there is at least one smoke exhaust outlet with a two-dimensional cross-sectional simulated flow rate... If the exhaust volume is greater than the design exhaust volume, then calculate the combined flow rate q at each exhaust outlet. ; where the coefficient , As a reliability factor, As importance weight, From the confusion matrix The confusion matrix was calculated. For a 3 n Binary classification problems within the framework of identification, i.e. There are 3 n One, measuring 2*2: ;in This represents the probability that the j-th working condition is judged as not requiring fire protection. This represents the probability that the j-th working condition is judged to require fire prevention instead of not requiring it. This represents the probability that the j-th working condition is judged to be fireproof instead of requiring fireproofing. This represents the probability that the j-th working condition is judged as requiring fire prevention. S4. Determine if the combined flow rate is less than the design smoke exhaust rate. If it is less than the design smoke exhaust rate, the protection action will not be triggered; otherwise, the preset fire protection command will be triggered.

2. The fire smoke monitoring and early warning method for industrial buildings according to claim 1, characterized in that, Step S1 includes: the building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions; the fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

3. The fire smoke monitoring and early warning method in industrial buildings according to claim 2, characterized in that, Step S1 specifically includes: Create an image sample set of combustible and flammable explosive materials that need to be stored in industrial buildings, and train a YOLO network model; The image captured by the camera is scaled up to the size required by YOLO and divided into an equally divided grid, where each grid is used to detect targets whose center point falls within that grid. Each grid cell predicts a conditional probability value for each category of stacked objects, and generates B square bounding boxes that enclose the objects based on the grid. Each bounding box predicts five regression values ​​x. Y , y Y , w Y , h Y , c Y , where x Y , y Y , w Y , h Y These represent the center coordinates, width, and height of the bounding box, respectively. Y The probability and location accuracy of the bounding box are represented by the NMS method, which is used to filter and obtain a coarse prediction box. Extract the rough prediction bounding box image and convert it into a grayscale image. Use the Canny operator to extract its edges and obtain a binary image. Use the findContours function to obtain the target contour line outline. Optimize the contour line and draw the fitting contour. Measure the contour perimeter or filling area. Combine multiple camera calibration information to estimate the scale of the fire source accumulation at the scene.

4. A fire smoke monitoring and early warning system for industrial buildings, characterized in that, include: The data acquisition module is used to obtain building fire boundary conditions from the building information model and to obtain fire source parameters from the sensor system. The flow acquisition module is used to obtain the smoke state that is fully diffused in the industrial building based on the building fire boundary conditions and fire source parameters through RANS equations, obtain the three-dimensional velocity field, and calculate the two-dimensional cross-sectional simulated flow of each smoke exhaust outlet in the building. Set initial parameter x i There exists a range of values ​​[x] bi x ui ], where i = 1, 2, 3..., and the probability of taking values ​​within this range follows a Gaussian distribution Φ. i Then these mutually independent Φ i The superposition will yield a multivariate Gaussian distribution, where σ is defined as Φ. i Variance σ i The set, according to the 3σ criterion, divides the side length into 3 parts. For an n-dimensional multivariate Gaussian distribution, the initial number of test cases is 3. n The obtained three-dimensional velocity field The simulated flow rate of the two-dimensional cross-section of each smoke exhaust outlet inside the building was calculated. , ,in The geometric dimensions of the smoke exhaust outlets under different operating conditions are obtained from the building information model, where j=1,2,3…3 n ; The fusion flow calculation module is used to calculate the fusion flow of each smoke exhaust outlet when at least one of the two-dimensional cross-sectional simulated flow rates in each smoke exhaust outlet is greater than the design smoke exhaust volume. If the simulated flow rate of the two-dimensional cross-section of all exhaust outlets is... If all values ​​are less than the design smoke exhaust volume, subsequent protective actions will not be triggered; if there is at least one smoke exhaust outlet with a two-dimensional cross-sectional simulated flow rate... If the exhaust volume is greater than the design exhaust volume, then calculate the combined flow rate q at each exhaust outlet. ; where the coefficient , As a reliability factor, As importance weight, From the confusion matrix The confusion matrix was calculated. For a 3 n Binary classification problems within the framework of identification, i.e. There are 3 n One, measuring 2*2: ;in This represents the probability that the j-th working condition is judged as not requiring fire protection. This represents the probability that the j-th working condition is judged to require fire prevention instead of not requiring it. This represents the probability that the j-th working condition is judged to be fireproof instead of requiring fireproofing. This represents the probability that the j-th working condition is judged as requiring fire prevention. The flow rate determination module is used to determine whether the combined flow rate is less than the design smoke exhaust volume. If it is less than the design smoke exhaust volume, the protection action will not be triggered; otherwise, the preset fire protection command will be triggered.

5. The fire smoke monitoring and early warning system for industrial buildings according to claim 4, characterized in that: The building fire boundary conditions include, but are not limited to, the smoke exhaust port inlet velocity, the smoke exhaust port outlet velocity, and the initial environmental conditions. The fire source parameters include, but are not limited to, the fire source intensity and the fire source accumulation scale.

6. A fire smoke monitoring and early warning device for industrial buildings, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-3.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-3.

Images