Logistics park fire risk dynamic assessment method based on virtual-real linkage

Abstract

The invention provides a logistics park fire-fighting risk dynamic assessment method based on virtual-real linkage, and the key points of the technical scheme comprise the steps: deploying physical layer equipment, and obtaining the building data of a logistics park; based on the building data of the logistics park, constructing a digital twinborn model and marking fire danger hidden danger areas and fire-fighting facility points; constructing a dynamic risk assessment model based on the detection factors of the physical layer equipment; based on the digital twinborn model and the dynamic risk assessment model, verifying the performance of the model through virtual-real linkage, and adjusting and optimizing the model; acquiring real-time detection data based on the dynamic risk assessment model, calculating the risk value of the fire hazard in each fire hazard area at the current sampling moment, and respectively judging the risk level; in response to the risk level reaching the set alarm level, fire alarm is triggered in a graded manner, and an emergency mechanism is triggered in a regional manner; according to the invention, the fire risk of the logistics park can be effectively and dynamically evaluated, and the early warning capability and the emergency processing efficiency of major fire risks are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fire risk assessment, and in particular to a dynamic assessment method for fire risk in a logistics park based on virtual-real linkage. Background Art

[0002] With the rapid development of the logistics industry, modern logistics parks typically occupy large areas, with centralized functions such as warehousing, sorting, and transportation. The diverse range of goods, including flammable and explosive items, has significantly increased fire hazards. Simultaneously, with the widespread adoption of automated equipment, such as automated warehouses, automated guided vehicles (AGVs), and intelligent sorting systems, the risk of electrical fires has increased. Furthermore, modern logistics parks typically utilize high-bay, densely packed storage systems, which create high mobility of personnel and vehicles and make management difficult. Fires can easily be caused by operational errors or violations. Once a fire occurs, the high-bay, densely packed storage system can spread the fire quickly, making it difficult to extinguish. Furthermore, it can make it difficult for fire management personnel to promptly detect the fire source and its severity, making emergency evacuation of personnel challenging.

[0003] Therefore, it is very important to do a good job in fire safety prevention and control management in modern logistics parks. Effectively assessing the fire risks in logistics parks and promptly discovering and eliminating safety hazards that cause fires are of great significance to fire safety prevention and control in modern logistics parks.

[0004] However, traditional static fire risk assessment methods have great limitations. Traditional methods are mostly based on regular inspections by firefighters or historical data, and cannot reflect dynamic risks in real time, such as risks that may be caused by factors such as cargo turnover and changes in environmental temperature and humidity. At the same time, the identification and assessment of fire risks in traditional methods are highly dependent on the experience of firefighters, are highly subjective, and lack data support. Moreover, the alarm response of traditional methods is prone to lag, making it difficult to detect sudden risks such as electrical short circuits and equipment failures in a timely manner, resulting in insufficient early warning capabilities.

[0005] Therefore, in view of the warehousing characteristics, storage modes and difficulties in fire safety prevention and control in modern logistics parks, a dynamic fire risk assessment method is developed that can reflect dynamic risks in real time and detect sudden risks in a timely manner, which plays a substantial and important role in the fire safety prevention and control of modern logistics parks. Summary of the Invention

[0006] In order to solve the difficulties in fire safety prevention and control in modern logistics parks, realize the role of real-time reflection of dynamic risks, timely detection of sudden risks, and effectively assess the fire risks of logistics parks, the present invention provides a dynamic assessment method for fire risks in logistics parks based on virtual-real linkage. The technical solution is as follows:

[0007] A dynamic fire risk assessment method for a logistics park based on virtual-physical linkage includes the following steps: deploying physical-layer equipment to obtain building data for the logistics park; constructing a digital twin model based on the building data of the logistics park and marking fire hazard areas and fire protection facilities; constructing a dynamic risk assessment model based on the detection factors of the physical-layer equipment; verifying the performance of the model through virtual-physical linkage and optimizing it; obtaining real-time detection data based on the dynamic risk assessment model, calculating the risk value of fire in each fire hazard area at the current sampling moment, and determining the risk level; and triggering fire alarms and emergency response mechanisms corresponding to the risk level in each area according to the risk level.

[0008] Among them, when calculating the risk value, priority is given to obtaining the cargo information stacked in each fire hazard area, and the temperature sensitivity coefficient matching each fire hazard area is selected based on the cargo type information and cargo quantity information. The calculation formula is:

[0009]

[0010] Where R n represents the risk value of fire in the nth fire hazard area at the current sampling time, e represents a natural constant, k n represents the temperature sensitivity coefficient of the nth fire hazard area, T n,max It represents the maximum detected temperature of the nth fire hazard area at the current sampling time, and T0 represents the weather environment temperature value at the current sampling time;

[0011] When the risk level of a fire hazard area reaches the highest level or the next highest level, a secondary risk level determination is performed based on the smoke concentration data and flame thermal radiation data of the corresponding fire hazard area.

[0012] Preferably, physical layer equipment is deployed to obtain building data of the logistics park, including the installation of corresponding IoT sensor detection devices in fire hazard areas and fire protection facilities, and connecting to the server through the network; at the same time, the BIM model of the logistics park buildings is obtained, the real-life data of the logistics park is collected, the precise three-dimensional coordinates of the fire protection facilities and IoT sensor detection devices are obtained and calibrated in the BIM model.

[0013] Preferably, a digital twin model is constructed, including point cloud denoising and image processing of the acquired real-scene data of the logistics park, generating a real-scene model with geographic coordinates, unifying the coordinate systems of the BIM model and the real-scene model, spatially aligning the BIM model and the real-scene model, and generating a high-precision three-dimensional park model; deleting unnecessary components in the three-dimensional park model, reducing the volume of the three-dimensional park model, and making the three-dimensional park model lightweight; marking fire hazard areas, fire-fighting facilities, Internet of Things sensor detection devices, and fire escape routes in the three-dimensional park model.

[0014] Preferably, constructing a digital twin model also includes associating attributes that need to be dynamically updated in the three-dimensional park model with corresponding data, performing dynamic attribute mapping, binding the detection data of the Internet of Things sensor detection device to the corresponding detection areas in the three-dimensional park model, and binding the fire protection facility status data collected by the Internet of Things sensor detection device to the corresponding fire protection facility points in the three-dimensional park model; based on the temperature data collected by the Internet of Things sensor detection device, dynamically rendering the model surface corresponding to each fire hazard area, driving the color gradient of the corresponding fire hazard area in the three-dimensional park model, and generating a visual heat map; setting abnormal trigger conditions for the fire protection facility status, and marking abnormal fire protection facility points in the three-dimensional park model with warning icons based on the fire protection facility status data collected by the Internet of Things sensor detection device.

[0015] Preferably, a dynamic risk assessment model is constructed, including numbering the fire hazard areas in the logistics park from 1 to N, where N represents the total number of fire hazard areas in the logistics park; obtaining the maximum detected temperature of each fire hazard area at the current sampling moment, wherein the calculation formula for the maximum detected temperature of the nth fire hazard area at the current sampling moment is:

[0016] T n,max =max(T n,1 ,...,T n,M ),

[0017] In the formula, max() represents the maximum value acquisition function, T n,1 represents the temperature value detected by the first sensor in the nth fire hazard area at the current sampling time, T n,M Represents the temperature value detected by the Mth sensor in the nth fire hazard area at the current sampling time, where n ranges from 1 to N. Calculates the risk value of fire in each fire hazard area at the current sampling time. Based on the range of risk values, a five-level risk classification standard is established, with level 1 being the lowest and level 5 being the highest. The risk level of each fire hazard area is divided according to the risk value at the current sampling time.

[0018] At the same time, the IoT sensor detection devices installed at the fire protection facilities are used to detect the status of the fire protection facilities at each fire protection facility at the current sampling time. When the status of the fire protection facilities at a certain fire protection facility exceeds the set threshold range, it is determined that the fire protection facilities at the corresponding fire protection facility are abnormal, and the risk level of the corresponding fire protection facility is determined to be level three, and the corresponding emergency mechanism is triggered;

[0019] At the same time, according to the risk levels of various fire hazard areas and fire protection facilities at the current sampling moment, the corresponding warning signs are mapped into the digital twin model.

[0020] Preferably, the construction of the risk level classification standard includes that when the risk value is greater than 0 and less than or equal to 0.2, the risk level is level one, indicating no significant risk; when the risk value is greater than 0.2 and less than or equal to 0.4, the risk level is level two, indicating potential hidden dangers; when the risk value is greater than 0.4 and less than or equal to 0.6, the risk level is level three, indicating obvious abnormalities; when the risk value is greater than 0.6 and less than or equal to 0.8, the risk level is level four, indicating major hidden dangers; when the risk value is greater than 0.8 and less than or equal to 1, the risk level is level five, indicating that a fire has occurred.

[0021] Preferably, based on the digital twin model and the dynamic risk assessment model, the virtual-reality linkage verifies the model performance and adjusts and optimizes it, including geometric accuracy verification, randomly selecting a number of real feature points, measuring the coordinates of the real feature points and comparing them with the model coordinates, requiring the coordinate error to be no more than 5 cm; sensor detection device accuracy verification, randomly selecting sensor detection devices in different areas, comparing their detection data with manual measurement data, requiring the error of the detection data to be no more than 10%; simulation operation verification, randomly simulating a fire scenario in a fire hazard area to verify the real-time and reliability of the digital twin model and the dynamic risk assessment model; at the same time, periodically update the cargo category and fire-fighting facility change information to optimize the accuracy of the model.

[0022] Preferably, real-time detection data is obtained, and the risk value of fire in each fire hazard area at the current sampling moment is calculated, including obtaining the detection information data of the physical layer equipment in each fire hazard area at the current sampling moment, and aligning the timestamps of the temperature data, smoke concentration data, and flame thermal radiation data collected by the Internet of Things sensor detection device; according to the dynamic risk assessment model, the risk value of fire in each fire hazard area at the current sampling moment is calculated, and at the same time, the status of the fire-fighting facilities at each fire-fighting facility point at the current sampling moment is detected and an abnormality judgment is made.

[0023] Preferably, the risk level is determined according to the risk value and the risk level classification standard, including determining the risk level of each fire hazard area at the current sampling moment according to the risk value of fire in each fire hazard area at the current sampling moment; and determining the risk level of the abnormal fire protection facility point as level three according to the facility abnormality judgment result of each fire protection facility point at the current sampling moment;

[0024] Among them, the risk value of the nth fire hazard area at the current sampling time is R n , when 0.6 <R n ≤0.8, the risk level of the nth fire hazard area is preliminarily determined to be level 4. <R nWhen the value of smoke concentration data or flame thermal radiation data is less than or equal to the corresponding set threshold value, the risk level of the nth fire hazard area is preliminarily determined to be level five. At this time, the corresponding smoke concentration data and flame thermal radiation data are compared with the corresponding set threshold values respectively, and a secondary risk level determination is performed on the nth fire hazard area. When either the smoke concentration data or the flame thermal radiation data is greater than or equal to the corresponding set threshold value, the risk level of the nth fire hazard area at the current sampling moment remains unchanged. When both the smoke concentration data and the flame thermal radiation data are less than the corresponding set threshold values at the same time, the risk level of the nth fire hazard area is reduced by one level.

[0025] Preferably, fire alarms and emergency mechanisms corresponding to risk levels are triggered in different areas according to risk levels, including: when the risk level is level one, each fire hazard area and each fire-fighting facility point is normal and no alarm is triggered; when the risk level is level two, information is sent to remind fire management personnel in the corresponding area to conduct inspections; when the risk level is level three, information is sent to notify fire management personnel in the corresponding area to conduct timely investigations on abnormal situations; when the risk level is level four, the fire alarm is triggered to remind staff in the corresponding area to evacuate and leave, and the fire management personnel in the corresponding area are notified to immediately conduct verification and processing; when the risk level is level five, the fire alarm and automatic fire-fighting facilities are triggered, and the fire management personnel in the corresponding area and adjacent areas are notified to immediately conduct emergency evacuation, and the fire situation is simulated and predicted through the digital twin model to plan safe escape routes and fire-fighting routes.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. The present invention provides a dynamic assessment method for fire risks in logistics parks based on virtual-real linkage, which can solve the difficulties in fire safety prevention and control in modern logistics parks, such as large area, concentrated functions, a wide variety of goods, and intensive storage. It can reflect fire risks in real time and discover sudden risks in a timely manner; through the corresponding Internet of Things sensor detection devices, the abnormal conditions of each area are monitored in real time. At the same time, the fire risks of each area in the logistics park are effectively and dynamically assessed through the digital twin model and the dynamic risk assessment model, and the risk levels are divided respectively. According to the risk assessment results, the fire-fighting equipment and personnel of the logistics park can be reasonably configured, the scheduling and management of fire-fighting resources can be optimized, and an all-weather, multi-dimensional, traceable fire safety risk assessment plan can be realized to predict and control fire risks, and significantly improve the early warning capability and emergency response efficiency of major fire risks.

[0028] 2. The present invention provides a dynamic assessment method for fire risk in logistics parks based on virtual-reality linkage. By establishing a three-dimensional visualization platform through a digital twin model, it can realize functions such as dynamic risk assessment, risk heat map generation, intelligent matching of emergency plans, and dynamic allocation of emergency resources. The digital twin model has strong scalability and can be further combined with augmented reality technology to realize virtual-reality linkage control of fire protection facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0030] Figure 1 It is a flow chart of the dynamic assessment method of fire risk; DETAILED DESCRIPTION

[0031] The technical features of the present invention are further described in detail below with reference to the accompanying drawings so that those skilled in the art can understand them.

[0032] The application scenario of the present invention is a modern logistics park. A modern logistics park is a complex system with a large area, concentrated functions such as warehousing, sorting, and transportation, a wide variety of goods, and popular automation equipment. Modern logistics parks have many warehouses, equipment, and personnel, and many fire risk factors. Traditional static assessments may not be timely enough, so dynamic assessments are needed. Dynamic assessment means that risk status can be monitored and updated in real time or near real time. When storing goods in a logistics park, the goods need to be divided into three types: flammable, combustible, and non-flammable. Goods of the same type are stored and placed in the same place to facilitate fire management and prevention.

[0033] The present invention adopts a technical solution that combines virtual models with actual physical systems to effectively evaluate the fire risks in logistics parks. The complete dynamic fire risk assessment system consists of three parts: data layer, model layer, and application layer. The data layer needs to collect various data, such as building structure data, equipment status, environmental data, operational data, and historical data; the sources of these data include sensors, monitoring systems, management platforms, etc. The model layer includes digital twin models and risk assessment models. The application layer is a specific application solution, such as real-time monitoring, early warning, simulation exercises, and decision support. The present invention focuses on describing the establishment and operation methods of the model layer to achieve dynamic assessment of the fire risks in logistics parks, so as to be used for real-time monitoring of fire safety hazards in logistics parks and timely early warning of fire risks.

[0034] A dynamic assessment method for fire risk in logistics parks based on virtual-real linkage, its operation logic is as follows Figure 1 The specific operation steps are as follows:

[0035] Step S1: Deploy physical layer equipment to obtain building data of the logistics park;

[0036] Specifically, deploying physical layer equipment includes: Installing appropriate IoT sensing devices in fire hazard areas and firefighting facilities, and connecting them to servers via the network. For example, IoT sensing devices such as temperature sensors, smoke detectors, flame sensors, infrared thermal imaging cameras, and fire water pressure monitors should be installed in key areas such as warehouses, freight yards, and power distribution rooms. Since goods stored in logistics parks need to be classified as flammable, combustible, and non-flammable, and similar goods are stored and placed in the same area, IoT sensing devices deployed in the same area should be used to detect similar goods. Due to the large space in areas such as warehouses and freight yards, multiple temperature sensors, smoke detectors, flame sensors, or infrared thermal imaging cameras should be installed in different locations to more accurately detect the actual conditions and reduce the impact of spatial distance on the accuracy of detection data. Thermal imaging cameras can be selected from the FLIR-A700 series, with an accuracy of ±2°C. Smoke detectors can be laser smoke detectors that can distinguish between dust and actual smoke. Firefighting facilities can use smart fire hydrants and smart sprinkler systems with pressure sensors and solenoid valve controls to facilitate emergency response.

[0037] In addition, obtaining the building data of the logistics park includes: obtaining the BIM model of the logistics park buildings and collecting the real-life data of the logistics park; at the same time, obtaining the precise three-dimensional coordinates of the fire-fighting facilities and IoT sensor detection devices and calibrating them in the BIM model. Key parameters such as the fire pipe diameter and sprinkler head position can be extracted from the BIM model data; collecting the real-life data of the logistics park can be used to conduct a full-scale scan of the logistics park buildings, fire hazard areas and fire-fighting facilities through laser scanning, and at the same time, use drone aerial photography to perform oblique photographic modeling of high-altitude facilities, such as roof smoke exhaust vents and lightning rods; specific operations: use a ground-based three-dimensional laser scanner, such as the FARO-Focus-S-350, set a 5mm point pitch, and conduct a full-scale scan of the park buildings, shelves, and fire-fighting facilities. The scanning density of key areas such as flammable warehouses and distribution rooms is increased to 2mm; use a drone equipped with a camera, such as the DJI M300-RTK equipped with a Zenmuse P1 lens, at a flight altitude of 50m to generate 5cm resolution orthophotos for oblique photographic modeling of high-altitude facilities.

[0038] Step S2: Based on the building data of the logistics park, a digital twin model is constructed and fire hazard areas and fire protection facilities are marked;

[0039] Specifically, building a digital twin model includes: performing point cloud denoising and image processing on the acquired real-life data of the logistics park, generating a real-life model with geographic coordinates, unifying the coordinate systems of the BIM model and the real-life model, spatially aligning the BIM model and the real-life model, and generating a high-precision three-dimensional park model; specific operations: performing point cloud denoising on the collected real-life data using Cloud-Compare software, generating a real-life model with geographic coordinates using Pix4Dmapper software, and then spatially aligning the BIM model and the real-life model in AutoCAD-Civil-3D to ensure that the error in the width of the fire passage is less than 0.1m; then, based on the high-precision three-dimensional park model, deleting non-essential components in the three-dimensional park model, such as decorative lines, reducing the volume of the three-dimensional park model, and making the three-dimensional park model lightweight; marking fire hazard areas, fire protection facilities, Internet of Things sensor detection devices, and fire escape routes in the three-dimensional park model.

[0040] In addition, building a digital twin model also includes: associating attributes that need to be dynamically updated in the 3D park model with corresponding data to perform dynamic attribute mapping; binding the detection data of the IoT sensor detection device to the corresponding detection area in the 3D park model, and binding the fire protection facility status data collected by the IoT sensor detection device to the corresponding fire protection facility point in the 3D park model; it can also establish parametric models for movable equipment in the logistics park, such as forklifts, unmanned guided vehicles (AGVs), firefighting robots, etc., and import them into the 3D park model, and synchronize the positioning data of the movable equipment with the movable equipment model in the 3D park model in real time;

[0041] Afterwards, based on the temperature data collected by the IoT sensor detection device, the model surface corresponding to each fire hazard area is dynamically rendered, driving the color gradient of the corresponding fire hazard area in the three-dimensional park model to generate a visual heat map; setting the abnormal trigger conditions of the fire protection facility status, based on the fire protection facility status data collected by the IoT sensor detection device, marking the abnormal fire protection facility points in the three-dimensional park model with warning icons; thereby, the digital twin model has the function of a dynamic visualization platform.

[0042] Step S3: constructing a dynamic risk assessment model based on the detection factors of the physical layer equipment;

[0043] Specifically, building a dynamic risk assessment model includes: numbering the fire hazard areas in the logistics park from 1 to N, where N represents the total number of fire hazard areas in the logistics park; obtaining the maximum detected temperature of each fire hazard area at the current sampling time, where the calculation formula for the maximum detected temperature of the nth fire hazard area at the current sampling time is:

[0044] T n,max =max(Tn,1 ,...,T n,M ),

[0045] In the formula, max() represents the maximum value acquisition function, T n,1 represents the temperature value detected by the first sensor in the nth fire hazard area at the current sampling time, T n,M It represents the temperature value detected by the Mth sensor in the nth fire hazard area at the current sampling time, where n ranges from 1 to N.

[0046] Calculate the risk value of fire in each fire hazard area at the current sampling moment; the calculation formula of the risk value is:

[0047]

[0048] Where R n represents the risk value of fire in the nth fire hazard area at the current sampling time, e represents a natural constant, k n represents the temperature sensitivity coefficient of the nth fire hazard area, T n,max Indicates the maximum detected temperature of the nth fire hazard area at the current sampling time. T0 represents the weather environment temperature value at the current sampling time. The weather environment temperature value at the current sampling time can be obtained by connecting to the network or by collecting it using a temperature sensor set outside the corresponding fire hazard area.

[0049] The temperature sensitivity coefficient is related to the flammability and quantity of goods. Therefore, when calculating the risk value, it is necessary to obtain the cargo type information and cargo quantity information of each fire hazard area from the cargo management platform of the logistics park, so as to set the temperature sensitivity coefficient of each fire hazard area according to the cargo type information and cargo quantity information; since when storing goods in the logistics park, it is necessary to divide the goods into three types: flammable, combustible, and non-flammable. The same type of goods are concentrated in the same place for storage and placement. The data collected from the same area is for the same type of goods. The storage density of the goods can be further divided into five status levels: full, large, half-full, small, and empty. Combined with the cargo type and the storage density status of the goods, the corresponding temperature sensitivity coefficient is set for each combination of cargo type and storage density status. The value range of the temperature sensitivity coefficient is 0.01~0.1. When calculating the risk value, the corresponding temperature sensitivity coefficient can be directly called for calculation based on the information obtained from the cargo management platform of the logistics park.

[0050] In addition, the establishment of risk level classification standards includes: constructing a five-level risk level classification standard based on the value range of risk values, with level one being the lowest and level five being the highest, and the risk value range is 0 to 1; when the risk value is greater than 0 and less than or equal to 0.2, the risk level is level one, indicating no significant risk; when the risk value is greater than 0.2 and less than or equal to 0.4, the risk level is level two, indicating potential hidden dangers; when the risk value is greater than 0.4 and less than or equal to 0.6, the risk level is level three, indicating obvious abnormalities; when the risk value is greater than 0.6 and less than or equal to 0.8, the risk level is level four, indicating major hidden dangers; when the risk value is greater than 0.8 and less than or equal to 1, the risk level is level five, indicating that a fire has occurred;

[0051] After obtaining the risk value at the current sampling moment, the risk level of each fire hazard area is divided according to the risk value. When the risk level of a fire hazard area reaches level four or level five, a secondary risk level determination is performed based on the smoke concentration data and flame thermal radiation data collected by the IoT sensor detection device of the corresponding fire hazard area at the current sampling moment. When either the smoke concentration data or the flame thermal radiation data is greater than or equal to the corresponding set threshold, it is confirmed that the risk level at the current sampling moment has reached level four or level five, and the corresponding fire alarm and emergency mechanism are triggered. When both the smoke concentration data and the flame thermal radiation data are less than the set threshold at the same time, the risk level of the corresponding fire hazard area is reduced by one level, triggering the corresponding fire alarm and emergency mechanism.

[0052] At the same time, the IoT sensor detection devices installed at the fire protection facilities are used to detect the status of the fire protection facilities at each fire protection facility at the current sampling time. When the status of the fire protection facilities at a certain fire protection facility exceeds the set threshold range, it is determined that the fire protection facilities at the corresponding fire protection facility are abnormal, and the risk level of the corresponding fire protection facility is determined to be level three, and the corresponding emergency mechanism is triggered;

[0053] At the same time, according to the risk levels of various fire hazard areas and fire protection facilities at the current sampling moment, the corresponding warning signs are mapped into the digital twin model to visualize the risk levels of various areas.

[0054] Step S4: Based on the digital twin model and dynamic risk assessment model, the model performance is verified and optimized through virtual-real linkage;

[0055] Specifically, based on the digital twin model and dynamic risk assessment model, the virtual-reality linkage verification model performance and adjustment optimization include: geometric accuracy verification, randomly selecting a number of real feature points, measuring the coordinates of the real feature points and comparing them with the model coordinates, requiring the coordinate error to be no more than 5cm; sensor detection device accuracy verification, randomly selecting sensor detection devices in different areas, and comparing their detection data with manual measurement data, requiring the error of the detection data to be no more than 10%; simulation operation verification, randomly simulating a fire scenario in a fire hazard area to verify the real-time and reliability of the digital twin model and dynamic risk assessment model; at the same time, periodically updating cargo categories and fire-fighting facility change information to optimize the accuracy of the model.

[0056] In addition, when the verification error of the geometric accuracy verification exceeds the set standard, the coordinate system of the three-dimensional park model needs to be adjusted and optimized; when the verification error of the accuracy verification of the sensor detection device exceeds the set standard, the installation position or installation quantity of the corresponding sensor detection device can be adjusted and optimized to obtain detection data that is closer to the actual measurement value.

[0057] In addition, after the model is verified and optimized, the digital twin model and dynamic risk assessment model can be trained through historical data and scenario settings to generate emergency plans corresponding to various risk scenarios, thereby optimizing the corresponding fire emergency mechanism; it can also integrate the LSTM time series prediction algorithm to analyze the time series characteristics of temperature data and smoke data, and predict risk trends in the next 5-10 minutes.

[0058] Step S5: Based on the dynamic risk assessment model, real-time detection data is obtained, the risk value of fire in each fire hazard area at the current sampling moment is calculated, and the risk level is determined;

[0059] Specifically, obtaining real-time detection data and calculating the risk value of fire in each fire hazard area at the current sampling moment include: obtaining the detection information data of the physical layer equipment in each fire hazard area at the current sampling moment, and the data collected from the same area is for the same type of goods; aligning the timestamps of the temperature data, smoke concentration data, and flame thermal radiation data collected by the Internet of Things sensor detection device; calculating and obtaining the risk value of fire in each fire hazard area at the current sampling moment according to the dynamic risk assessment model, and at the same time, detecting the status of the firefighting facilities at each firefighting facility point at the current sampling moment and making abnormal judgments.

[0060] In addition, the risk level determination based on the risk value and risk level classification standards includes: determining the risk level of each fire hazard area at the current sampling moment based on the risk value of fire in each fire hazard area at the current sampling moment; determining the risk level of the abnormal fire protection facility point as level three based on the facility abnormality judgment result of each fire protection facility point at the current sampling moment;

[0061] Among them, the risk value of the nth fire hazard area at the current sampling time is R n , when 0.6 <R n ≤0.8, the risk level of the nth fire hazard area is preliminarily determined to be level 4. <R n When the value of smoke concentration data or flame thermal radiation data is less than or equal to the corresponding set threshold value, the risk level of the nth fire hazard area is preliminarily determined to be level five. At this time, the corresponding smoke concentration data and flame thermal radiation data are compared with the corresponding set threshold values respectively, and a secondary risk level determination is performed on the nth fire hazard area. When either the smoke concentration data or the flame thermal radiation data is greater than or equal to the corresponding set threshold value, the risk level of the nth fire hazard area at the current sampling moment remains unchanged. When both the smoke concentration data and the flame thermal radiation data are less than the corresponding set threshold values at the same time, the risk level of the nth fire hazard area is reduced by one level.

[0062] Step S6: triggering fire alarms and emergency mechanisms corresponding to the risk levels in different areas according to the risk levels;

[0063] Specifically, the fire alarms and emergency mechanisms that trigger corresponding risk levels in different areas include: when the risk level is level one, all fire hazard areas and fire-fighting facilities are normal and no alarms are triggered; when the risk level is level two, information is sent to remind fire management personnel in the corresponding area to conduct inspections; when the risk level is level three, information is sent to notify fire management personnel in the corresponding area to conduct timely investigations on abnormal situations; when the risk level is level four, the fire alarm is triggered to remind staff in the corresponding area to evacuate and leave, and the fire management personnel in the corresponding area are notified to immediately conduct verification and processing; when the risk level is level five, the fire alarm and automatic fire-fighting facilities are triggered, and the fire management personnel in the corresponding area and adjacent areas are notified to immediately conduct emergency evacuation. The fire situation is simulated and predicted through the digital twin model, and safe escape routes and fire-fighting routes are planned.

[0064] The embodiments described in the present invention are merely descriptions of preferred implementations of the present invention and are not limited to the precise structures described above and shown in the accompanying drawings. Various modifications and changes can be made without departing from the scope of protection thereof. Without departing from the design concept of the present invention, various variations and improvements made to the technical solutions of the present invention by engineers and technicians in this field should fall within the scope of protection of the present invention.

Claims

1. A dynamic assessment method for fire risk in a logistics park based on virtual-real linkage, characterized by: Deploy physical layer equipment to obtain building data in the logistics park; Based on the building data of the logistics park, a digital twin model is constructed and fire hazard areas and fire protection facilities are marked; based on the detection factors of physical layer equipment, a dynamic risk assessment model is constructed; Based on the digital twin model and dynamic risk assessment model, the virtual and real linkage verifies the model performance and adjusts and optimizes it; Based on the dynamic risk assessment model, real-time detection data is obtained to calculate the risk value of fire in each fire hazard area at the current sampling time and determine the risk level; Trigger fire alarms and emergency response mechanisms based on risk levels in different areas; Among them, when calculating the risk value, priority is given to obtaining the cargo information stacked in each fire hazard area, and the temperature sensitivity coefficient matching each fire hazard area is selected based on the cargo type information and cargo quantity information. The calculation formula is: Where R n represents the risk value of fire in the nth fire hazard area at the current sampling time, e represents a natural constant, k n represents the temperature sensitivity coefficient of the nth fire hazard area, T n,max It represents the maximum detected temperature of the nth fire hazard area at the current sampling time, and T0 represents the weather environment temperature value at the current sampling time; When the risk level of a fire hazard area reaches the highest level or the next highest level, a secondary risk level determination is performed based on the smoke concentration data and flame thermal radiation data of the corresponding fire hazard area.

2. The fire risk dynamic assessment method according to claim 1, characterized in that: Deploy physical layer equipment to obtain building data of the logistics park, including installing corresponding IoT sensor detection devices in fire hazard areas and fire protection facilities, and connecting to the server through the network; at the same time, obtain the BIM model of the logistics park buildings, collect real-life data of the logistics park, obtain the precise three-dimensional coordinates of fire protection facilities and IoT sensor detection devices, and calibrate them in the BIM model.

3. The fire risk dynamic assessment method according to claim 2, characterized in that: Constructing a digital twin model involves performing point cloud denoising and image processing on the acquired real-life data of the logistics park, generating a real-life model with geographic coordinates, unifying the coordinate systems of the BIM model and the real-life model, spatially aligning the BIM model and the real-life model, and generating a high-precision three-dimensional park model; deleting unnecessary components in the three-dimensional park model, reducing the volume of the three-dimensional park model, and making the three-dimensional park model lightweight; marking fire hazard areas, fire-fighting facilities, Internet of Things sensor detection devices, and fire escape routes in the three-dimensional park model.

4. The method for dynamic fire risk assessment according to claim 3, characterized in that: Building a digital twin model also includes associating attributes that need to be dynamically updated in the three-dimensional campus model with corresponding data, performing dynamic attribute mapping, binding the detection data of the Internet of Things sensor detection device to the corresponding detection area in the three-dimensional campus model, and binding the fire protection facility status data collected by the Internet of Things sensor detection device to the corresponding fire protection facility point in the three-dimensional campus model; Based on the temperature data collected by the IoT sensor detection device, the model surface corresponding to each fire hazard area is dynamically rendered, driving the color gradient of the corresponding fire hazard area in the 3D campus model to generate a visual heat map; Set abnormal trigger conditions for the status of fire-fighting facilities, and mark abnormal fire-fighting facility points in the three-dimensional campus model with warning icons based on the fire-fighting facility status data collected by the Internet of Things sensor detection device.

5. The fire risk dynamic assessment method according to claim 2, characterized in that: A dynamic risk assessment model is constructed, which includes numbering the fire hazard areas in the logistics park from 1 to N, where N represents the total number of fire hazard areas in the logistics park; obtaining the maximum detection temperature of each fire hazard area at the current sampling time, where the calculation formula for the maximum detection temperature of the nth fire hazard area at the current sampling time is: T n,max =max(T n,1 ,...,T n,M ), In the formula, max() represents the maximum value acquisition function, T n,1 represents the temperature value detected by the first sensor in the nth fire hazard area at the current sampling time, T n,M Represents the temperature value detected by the Mth sensor in the nth fire hazard area at the current sampling time, where n ranges from 1 to N. Calculates the risk value of fire in each fire hazard area at the current sampling time. Based on the range of risk values, a five-level risk classification standard is established, with level 1 being the lowest and level 5 being the highest. The risk level of each fire hazard area is divided according to the risk value at the current sampling time. At the same time, the IoT sensor detection devices installed at the fire protection facilities are used to detect the status of the fire protection facilities at each fire protection facility at the current sampling time. When the status of the fire protection facilities at a certain fire protection facility exceeds the set threshold range, it is determined that the fire protection facilities at the corresponding fire protection facility are abnormal, and the risk level of the corresponding fire protection facility is determined to be level three, and the corresponding emergency mechanism is triggered; At the same time, according to the risk levels of various fire hazard areas and fire protection facilities at the current sampling moment, the corresponding warning signs are mapped into the digital twin model.

6. The fire risk dynamic assessment method according to claim 5, characterized in that: The construction of the risk level classification standard includes: when the risk value is greater than 0 and less than or equal to 0.2, the risk level is level one, indicating no significant risk; when the risk value is greater than 0.2 and less than or equal to 0.4, the risk level is level two, indicating potential hidden dangers; when the risk value is greater than 0.4 and less than or equal to 0.6, the risk level is level three, indicating obvious abnormalities; when the risk value is greater than 0.6 and less than or equal to 0.8, the risk level is level four, indicating major hidden dangers; when the risk value is greater than 0.8 and less than or equal to 1, the risk level is level five, indicating that a fire has occurred.

7. The fire risk dynamic assessment method according to claim 2, characterized in that: Based on the digital twin model and dynamic risk assessment model, the virtual-real linkage verifies the model performance and adjusts and optimizes it. This includes geometric accuracy verification, randomly selecting several real-world feature points, measuring their coordinates, and comparing them with the model coordinates, with the requirement that the coordinate error does not exceed 5cm; sensor detection device accuracy verification, randomly selecting sensor detection devices in different areas, and comparing their detection data with manual measurement data, with the requirement that the error of the detection data does not exceed 10%; Simulation operation verification randomly simulates a fire scenario in a fire hazard area to verify the real-time and reliability of the digital twin model and dynamic risk assessment model; at the same time, periodically update the cargo category and fire protection facility change information to optimize the accuracy of the model.

8. The method for dynamic fire risk assessment according to claim 6, characterized in that: Acquire real-time detection data and calculate the risk value of fire in each fire hazard area at the current sampling time. This includes obtaining detection information data of physical layer devices in each fire hazard area at the current sampling time and aligning the timestamps of temperature data, smoke concentration data, and flame thermal radiation data collected by IoT sensor detection devices. According to the dynamic risk assessment model, the risk value of fire in each fire hazard area at the current sampling moment is calculated. At the same time, the status of fire-fighting facilities at each fire-fighting facility point at the current sampling moment is detected and abnormal judgment is made.

9. The fire risk dynamic assessment method according to claim 8, characterized in that: Determine the risk level according to the risk value and risk level classification standard, including determining the risk level of each fire hazard area at the current sampling moment according to the risk value of fire in each fire hazard area at the current sampling moment; According to the abnormality judgment results of each fire protection facility at the current sampling time, the risk level of the abnormal fire protection facility point is determined to be level three; Among them, the risk value of the nth fire hazard area at the current sampling time is R n , when 0.6 <R n ≤0.8, the risk level of the nth fire hazard area is preliminarily determined to be level 4. <R n When the value of smoke concentration data or flame thermal radiation data is less than or equal to the corresponding set threshold, the risk level of the nth fire hazard area is preliminarily determined to be level 5. At this time, the corresponding smoke concentration data and flame thermal radiation data are compared with the corresponding set thresholds respectively, and a secondary risk level determination is performed on the nth fire hazard area. When either the smoke concentration data or the flame thermal radiation data is greater than or equal to the corresponding set threshold, the risk level of the nth fire hazard area at the current sampling moment remains unchanged. When the smoke concentration data and flame thermal radiation data are both less than the corresponding set thresholds, the risk level of the nth fire hazard area will be reduced by one level.

10. The fire risk dynamic assessment method according to claim 9, characterized in that: Fire alarms and emergency mechanisms corresponding to risk levels are triggered in different areas according to the risk level. When the risk level is level one, all fire hazard areas and fire-fighting facilities are normal and no alarms are triggered. When the risk level is level two, information is sent to remind fire management personnel in the corresponding area to conduct inspections. When the risk level is level three, information is sent to notify fire management personnel in the corresponding area to conduct timely investigations on abnormal situations. When the risk level is level four, the fire alarm is triggered to remind staff in the corresponding area to evacuate and the fire management personnel in the corresponding area are notified to conduct immediate verification and processing. When the risk level is level five, the fire alarm and automatic fire-fighting facilities are triggered, and the fire management personnel in the corresponding area and adjacent areas are notified to conduct emergency evacuation immediately. The fire situation is simulated and predicted through the digital twin model to plan safe escape routes and fire-fighting routes.

Images