Plant evacuation control method in emergency state based on multi-dimensional information fusion algorithm

Abstract

The present application belongs to the technical field of emergency path planning, and aims to solve the problems of traditional plant evacuation mode, such as slow response, unreasonable evacuation path planning and chaotic personnel evacuation. The present application provides a plant evacuation control method under emergency based on multi-dimensional information fusion algorithm, comprising the following steps: constructing an optimal evacuation path planning model based on plant evacuation under emergency; the optimal evacuation path planning model is obtained based on the combination of multi-source heterogeneous data and dynamic weight mechanism; the dynamic weight mechanism is constructed based on the importance determination rule of multi-source heterogeneous data under emergency; obtaining multi-source heterogeneous data of real-time plant evacuation under emergency; inputting the real-time multi-source heterogeneous data into the optimal evacuation path planning model, and dynamically outputting the optimal evacuation path corresponding to each personnel in the plant. The present application can fully consider various potential risks, and ensure that the designated evacuation area can effectively avoid danger and meet the actual needs of personnel evacuation.

Description

Technical Field

[0001] This invention belongs to the field of emergency event path planning technology, specifically relating to a factory evacuation control method under emergency conditions based on a multi-dimensional information fusion algorithm. Background Technology

[0002] During factory production, emergencies such as fires, explosions, and chemical leaks occur frequently. Traditional factory evacuation methods often suffer from slow response, unreasonable evacuation route planning, and chaotic personnel evacuation.

[0003] For example, relying on manual notification for evacuation can lead to untimely and inaccurate information transmission; if evacuation routes are not adjusted according to real-time conditions, it can easily cause congestion or even trap people in dangerous areas, making it difficult to ensure the safety of people's lives and reduce accident losses. Summary of the Invention

[0004] In order to solve at least one of the above-mentioned technical problems in the prior art, the present invention provides a factory evacuation control method in an emergency state based on a multi-dimensional information fusion algorithm.

[0005] This invention employs the following technical solution: an emergency evacuation control method for factory areas based on a multi-dimensional information fusion algorithm, comprising the following steps: An optimal evacuation route planning model for factory evacuation under emergency conditions is constructed, trained, and validated. The optimal evacuation route planning model is obtained based on a combination of multi-source heterogeneous data and a dynamic weighting mechanism. The multi-source heterogeneous data includes risk data, distance data, terrain and building structure data, personnel location and density data, and exit capacity and evacuation direction data. The dynamic weighting mechanism is constructed based on the importance determination rules of multi-source heterogeneous data under emergency conditions. Multi-source heterogeneous data on real-time evacuation of the factory area in emergency situations is obtained. The risk data, terrain and building structure data, exit capacity and evacuation direction data are obtained by comprehensively acquiring data based on factory area monitoring, sensors and electronic drawings of factory buildings. The personnel location and density data are obtained by comprehensively acquiring data based on UWB positioning base stations and factory area monitoring. All personnel in the factory area are wearing mobile positioning devices. Real-time multi-source heterogeneous data is input into the optimal evacuation route planning model, and the optimal evacuation route corresponding to personnel in each factory area is dynamically output. The optimal evacuation route is distributed to the mobile positioning devices of the corresponding personnel within the factory area, and the personnel within the factory area carry out emergency evacuation according to their respective optimal evacuation routes; and the optimal evacuation route is continuously and dynamically updated during emergency evacuation.

[0006] Preferably, the optimal evacuation route planning model is expressed as follows: ; The path that minimizes the sum of the weighted costs of all edges; in, In the formula, Weights for distance-based data; The actual distance to the edge; Weights for risk-related data; Risk level; Weights for terrain and building structure data; Due to the complexity of the terrain; Weights are assigned to export capacity and evacuation direction data. The degree of congestion in the passageway; Weights are assigned to personnel location and density data; For personnel density; It is the sum of the weighted costs of one side.

[0007] Preferably, when initially setting the weights for multi-source heterogeneous data, the weight of risk data accounts for 40%-60%, the weight of distance data accounts for 20%-30%, the weight of terrain and building structure data accounts for 10%-20%, the weight of exit capacity and evacuation direction data accounts for 5%-10%, and the weight of personnel location and density data accounts for 10%-20%.

[0008] Preferably, the risk data includes, but is not limited to: open flame identification results, smoke concentration or spread rate, explosion risk area, and toxic gas diffusion rate; distance data includes, but is not limited to: distance to the exit or path distance, number of turns, and effective distance affected by passage width; terrain and building structure data includes, but is not limited to: data slope, whether stairs are available, whether doors are open, whether passages are blocked, and building fire resistance rating; personnel location and density data includes, but is not limited to: current personnel location, personnel density, and personnel movement speed; and exit capacity and evacuation direction data includes, but is not limited to: exit width, maximum exit throughput, and whether it conflicts with the smoke diffusion direction.

[0009] Preferably, the system also includes setting up one-click evacuation physical buttons in the factory monitoring center and key management offices. These one-click evacuation physical buttons are used to trigger the optimal evacuation route planning model to plan evacuation routes in the factory area under emergency conditions; simultaneously trigger all emergency lighting systems in the factory area, including but not limited to passageway lights and exit indicator lights; simultaneously trigger the voice broadcasting system to broadcast evacuation messages in real time through speakers distributed in various corners of the factory area; and simultaneously trigger the access control system to unlock all relevant access control systems and ensure unobstructed personnel evacuation routes.

[0010] Preferably, the mobile positioning devices worn by personnel within the factory area interact with UWB positioning base stations deployed around the factory area via UWB positioning tags. The mobile positioning devices are smart bracelets equipped with vibration modules. The smart bracelet screen can display the current location information in real time, and simultaneously indicate the direction of the nearest evacuation route in the form of arrows and text. Through the built-in voice module of the smart bracelet, it can receive and play key information transmitted by sound columns to help personnel quickly locate their own location. If personnel within the factory area discover any abnormal situations in the surrounding area, including but not limited to blocked passages or injuries to other personnel, they can report to the central control system through the emergency SOS button on the smart bracelet. After receiving the feedback information, the central control system will promptly dispatch rescue forces for targeted handling.

[0011] Compared with the prior art, the beneficial effects of the present invention are: This invention breaks through the traditional single-factor delineation method, innovatively integrating multi-dimensional factors such as risk data, distance data, terrain and building structure data, personnel location and density data, exit capacity, and evacuation direction data to construct a scientific and accurate method for delineating evacuation areas. This strategy can fully consider various potential risks, ensuring that the delineated evacuation areas can effectively avoid dangers while meeting the actual needs of personnel evacuation, providing a reliable foundation for emergency evacuation.

[0012] The deep integration of the one-button evacuation physical control with the optimal evacuation route planning model, emergency lighting system, voice broadcasting system, and access control system enables instantaneous activation from the issuance of an instruction to a coordinated response from multiple systems in an emergency. This design simplifies the originally complex and time-consuming manual decision-making and separate operation of multiple systems into a one-button automated process, which can greatly shorten the emergency response time, buy precious golden time for personnel evacuation, and significantly improve the factory's response speed and efficiency in dealing with emergencies.

[0013] Leveraging the high precision of UWB (Ultra-Wideband) positioning technology, this system enables real-time and accurate tracking of personnel and mobile devices within the factory area. Combined with dynamic path planning algorithms, it adjusts evacuation routes promptly based on actual conditions (personnel distribution, changes in hazardous situations, etc.). This system overcomes the limitations of traditional fixed-route evacuations, providing personalized and optimized evacuation plans for each individual, effectively preventing congestion and accidental entry into dangerous areas, and ensuring efficient and safe evacuation.

[0014] The design combining smart bracelets with speaker columns not only enables multimodal alerts for evacuation information (vibration, voice, and screen display), but also establishes a two-way information exchange channel between personnel and the central control system. Personnel can promptly report any abnormal situations on-site via the bracelets, allowing the central control system to quickly allocate resources, forming a closed-loop emergency command system. This enhances the flexibility and controllability of the evacuation process and improves the overall emergency management level. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationships, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should fall within the scope of the technical content disclosed in the present invention. It should be noted that in this specification, relational terms such as "first" and "second" are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.

[0019] This invention provides an embodiment: like Figure 1 As shown, an emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm includes the following steps: S1: Construct and train / verify an optimal evacuation route planning model for factory evacuation under emergency conditions; the optimal evacuation route planning model is obtained based on a combination of multi-source heterogeneous data and a dynamic weighting mechanism; the multi-source heterogeneous data includes risk data, distance data, terrain and building structure data, personnel location and density data, and exit capacity and evacuation direction data; the dynamic weighting mechanism is constructed based on the importance determination rules of multi-source heterogeneous data under emergency conditions; In this embodiment, when delineating boundaries based on comprehensive factors, the factory's building layout drawings are used as a foundation, combined with the actual terrain (such as elevation and slope) and building structure (such as wall thickness and door / window location). Simultaneously, video surveillance and personnel positioning systems are integrated. The former uses an open flame algorithm to identify and pinpoint the fire source in real time, while the latter accurately displays the distribution of personnel in each area. Then, artificial passageways such as corridors, fire doors, and staircases within the factory are overlaid. Areas that can block the spread of hazards (such as firewalls that can delay the spread of fire and smoke), passageways that facilitate personnel evacuation (such as spacious, unobstructed corridors), areas avoiding open flames and the fire's reach, and densely populated areas are all considered in the evacuation zone designation. Finally, by integrating multi-dimensional information, a logically clear and feasible evacuation zone boundary is precisely delineated, ensuring that the boundary avoids risk sources while matching the current personnel distribution, effectively guiding the evacuation of personnel from different areas.

[0020] The optimal evacuation path planning model (core: Multi-attribute decision MADM + graph network shortest path) is expressed as follows: ; The path with the minimum weighted cost of all edges; in this embodiment, an edge is simply understood as "a passable physical passage", such as a passage from the workshop gate to the outdoor safety area; The "weighted sum of costs of all edges" is essentially the optimal evacuation path: from the current node of a person to the safe exit node, the route that has the smallest sum of costs for every passage along the entire path is the optimal evacuation path.

[0021] in, In the formula, Weights for distance-based data; The actual distance to the edge; Weights for risk-related data; Risk level; Weights for terrain and building structure data; Due to the complexity of the terrain; Weights are assigned to export capacity and evacuation direction data. The degree of congestion in the passageway; Weights are assigned to personnel location and density data; For personnel density; It is the sum of the weighted costs of one side.

[0022] When initially setting the weights for multi-source heterogeneous data, the weights for risk-related data are 40%-60%, distance-related data are 20%-30%, terrain and building structure data are 10%-20%, exit capacity and evacuation direction data are 5%-10%, and personnel location and density data are 10%-20%.

[0023] Furthermore, the dynamic weighting mechanism built upon the importance determination rules for multi-source heterogeneous data in emergency situations includes, but is not limited to: 1. If the smoke concentration exceeds the preset smoke concentration threshold, the weight of risk-related data will be increased to 80%. 2. If a channel is blocked, set the cost of that side to infinity; 3. If the number of people queuing at a certain exit exceeds the maximum capacity of that exit, the weighting of the exit capacity and evacuation direction data for that exit will be reduced. 4. If the population density in a certain area is too high, the distance data corresponding to that area will be reduced to prevent more people from entering the congested area.

[0024] S2: Obtain multi-source heterogeneous data on real-time evacuation of the factory area in an emergency. The risk data, terrain and building structure data, exit capacity and evacuation direction data are obtained by comprehensively acquiring data based on factory area monitoring, sensors and electronic drawings of factory buildings. The personnel location and density data are obtained by comprehensively acquiring data based on UWB positioning base stations and factory area monitoring. All personnel in the factory area are wearing mobile positioning devices. The risk-related data includes, but is not limited to: open flame identification results, smoke concentration or spread rate, explosion risk area, and toxic gas diffusion rate; distance-related data includes, but is not limited to: distance to the exit or path distance, number of turns, and effective distance affected by passage width; terrain and building structure data includes, but is not limited to: slope, whether stairs are available, whether doors are open, whether passages are blocked, and building fire resistance rating; personnel location and density data includes, but is not limited to: current personnel location, personnel density, and personnel movement speed; exit capacity and evacuation direction data includes, but is not limited to: exit width, maximum exit throughput, and whether it conflicts with the smoke diffusion direction.

[0025] S3: Input real-time multi-source heterogeneous data into the optimal evacuation route planning model and dynamically output the optimal evacuation route for personnel in each factory area. S4: The optimal evacuation route is sent to the mobile positioning devices of the corresponding personnel in the factory area, and the personnel in the factory area carry out emergency evacuation according to their respective optimal evacuation routes; and the optimal evacuation route is continuously and dynamically updated during emergency evacuation.

[0026] In this embodiment, one-button evacuation physical buttons are installed in the plant monitoring center and key management offices. These buttons are designed to be waterproof, dustproof, and prevent accidental activation, ensuring stable and reliable operation in various emergency situations. The one-button evacuation physical buttons trigger the optimal evacuation route planning model to plan evacuation routes within the plant during emergencies; simultaneously, they trigger all emergency lighting systems within the plant, including but not limited to corridor lights and exit indicator lights, ensuring sufficient lighting in evacuation routes; they also trigger the voice broadcasting system, using speakers distributed throughout the plant to broadcast evacuation messages clearly and in real time, such as "Emergency situation, please evacuate immediately according to the indicated directions"; and they trigger the access control system, unlocking all relevant access points to ensure unobstructed evacuation routes and minimize decision-making and response time. The optimal evacuation route planning model, emergency lighting system, voice broadcasting system, and access control system are integrated into an intelligent safety and environmental protection system.

[0027] The mobile positioning devices worn by personnel within the factory area interact with UWB positioning base stations deployed around the factory area via UWB positioning tags, enabling high-precision real-time tracking of personnel and equipment. This allows for accurate tracking of each person's location coordinates, direction of movement, and speed, with errors controllable to the centimeter level. This provides accurate data support for the precise issuance of evacuation notices and personalized guidance.

[0028] The mobile positioning device is a smart bracelet equipped with a vibration module. Upon receiving an evacuation notification, the smart bracelet immediately activates strong vibration, enabling it to detect danger immediately and respond promptly to evacuation commands, even in noisy production environments. The smart bracelet's screen displays the current location information in real time and indicates the direction of the nearest evacuation route using arrows and text. Through its built-in voice module, the smart bracelet can receive and play key information transmitted via sound columns, assisting personnel in quickly locating their position. If personnel within the factory area discover any abnormalities, including but not limited to blocked passages or injuries to other personnel, they can report them to the central control system via the SOS emergency button on the smart bracelet. Upon receiving the feedback, the central control system promptly dispatches rescue forces for targeted handling.

[0029] If the location shows that personnel have deviated from the route, are stranded, or have triggered the SOS function, the system automatically sends an early warning message containing the location of the open flame and the personnel's location to the management personnel, facilitating rapid intervention and guidance, and further improving the safety and efficiency of evacuation in fire scenarios. The evacuation method of this invention can also guide personnel according to their job positions and locations. Personnel in the production area evacuate laterally along the equipment safety passages without open flame hazards, while personnel working at heights are prioritized to transfer via dedicated vertical evacuation passages (such as escape ladders). The wristband will simultaneously push evacuation batch, assembly point information, and open flame warning prompts for the area, avoiding congestion and blockage.

[0030] In this embodiment, the intelligent safety and environmental protection system uses UWB positioning data as its core to capture the distribution density and movement trajectory of personnel in real time. Simultaneously, it combines the terrain and building layout of the evacuation area with danger signals such as open flames and dense smoke from video surveillance to quickly match the optimal evacuation route with the "shortest distance + lowest risk," and simultaneously pushes this information to the mobile positioning devices worn by personnel within the factory area. For example, when the system detects sudden situations such as congestion on a certain route (e.g., positioning shows people lingering at a stairwell for more than 30 seconds) or fire spread (e.g., video identifies an open flame in a corridor), it will recalculate within 10 seconds and provide dual reminders through a pop-up window on the wristband (with a new route arrow) and a factory-wide broadcast voice announcement (e.g., "Corridor 3 is blocked, please evacuate via staircase 5") to guide personnel to avoid obstacles in a timely manner. For example, when it is detected that the east corridor of a workshop is impassable due to thick smoke from a fire, the system will immediately lock the UWB location of 15 people in that area, replan their route through the west auxiliary passage to safety exit No. 2, and simultaneously notify the management personnel to pay close attention to the evacuation progress of this group of people to ensure that the whole process is efficient and smooth.

[0031] This technology, centered on emergency evacuation control in factory areas, has broad application prospects in many fields with high requirements for personnel safety and emergency management due to its high efficiency, precision, and intelligence.

[0032] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for emergency evacuation control of a factory area based on a multi-dimensional information fusion algorithm, characterized in that, Includes the following steps: Construct an optimal evacuation route planning model based on factory evacuation under emergency conditions and train and validate it; The optimal evacuation route planning model is obtained based on a combination of multi-source heterogeneous data and a dynamic weighting mechanism; the multi-source heterogeneous data includes risk data, distance data, terrain and building structure data, personnel location and density data, and exit capacity and evacuation direction data; The dynamic weighting mechanism is constructed based on the importance determination rules of multi-source heterogeneous data under emergency conditions; Multi-source heterogeneous data on real-time evacuation of the factory area in emergency situations is obtained. The risk data, terrain and building structure data, exit capacity and evacuation direction data are obtained by comprehensively acquiring data based on factory area monitoring, sensors and electronic drawings of factory buildings. The personnel location and density data are obtained by comprehensively acquiring data based on UWB positioning base stations and factory area monitoring. All personnel in the factory area are wearing mobile positioning devices. Real-time multi-source heterogeneous data is input into the optimal evacuation route planning model, and the optimal evacuation route corresponding to personnel in each factory area is dynamically output. The optimal evacuation route is distributed to the mobile positioning devices of the corresponding personnel within the factory area, and the personnel within the factory area carry out emergency evacuation according to their respective optimal evacuation routes; and the optimal evacuation route is continuously and dynamically updated during emergency evacuation.

2. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 1, characterized in that: The optimal evacuation route planning model is expressed as follows: ; The path that minimizes the sum of the weighted costs of all edges; in, In the formula, Weights for distance-based data; The actual distance to the edge; Weights for risk-related data; Risk level; Weights for terrain and building structure data; Due to the complexity of the terrain; Weights are assigned to export capacity and evacuation direction data. The degree of congestion in the passageway; Weights are assigned to personnel location and density data; For personnel density; It is the sum of the weighted costs of one side.

3. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 1, characterized in that: When initially setting the weights for multi-source heterogeneous data, the weights for risk-related data are 40%-60%, distance-related data are 20%-30%, terrain and building structure data are 10%-20%, exit capacity and evacuation direction data are 5%-10%, and personnel location and density data are 10%-20%.

4. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 3, characterized in that: The risk-related data includes, but is not limited to: open flame identification results, smoke concentration or spread rate, explosion risk area, and toxic gas diffusion rate; distance-related data includes, but is not limited to: distance to the exit or path distance, number of turns, and effective distance affected by passage width; terrain and building structure data includes, but is not limited to: slope, whether stairs are available, whether doors are open, whether passages are blocked, and building fire resistance rating; personnel location and density data includes, but is not limited to: current personnel location, personnel density, and personnel movement speed; exit capacity and evacuation direction data includes, but is not limited to: exit width, maximum exit throughput, and whether it conflicts with the smoke diffusion direction.

5. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 4, characterized in that: The dynamic weighting mechanism based on the importance determination rules of multi-source heterogeneous data in emergency situations includes, but is not limited to: If the smoke concentration exceeds the preset smoke concentration threshold, the weight of risk-related data will be increased to 80%. If a channel is blocked, then the cost of that side is set to infinity; If the number of people queuing at a certain exit exceeds the maximum capacity of that exit, the weighting of the exit capacity and evacuation direction data for that exit will be reduced. If the population density in a certain area is too high, the distance data for that area will be reduced to prevent more people from entering the congested area.

6. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 1, characterized in that: It also includes setting up one-click evacuation physical buttons in the factory monitoring center and key management offices. These one-click evacuation physical buttons are used to trigger the optimal evacuation route planning model to plan evacuation routes in the factory area in an emergency; at the same time, they trigger all emergency lighting systems in the factory area, including but not limited to passageway lights and exit indicator lights; at the same time, they trigger the voice broadcasting system to broadcast evacuation voice messages in real time through speakers distributed in various corners of the factory area; and at the same time, they trigger the access control system to unlock all relevant access control systems and unblock personnel evacuation channels.

7. The emergency evacuation control method for a factory area based on a multi-dimensional information fusion algorithm according to claim 6, characterized in that: The mobile positioning devices worn by personnel within the factory area interact with UWB positioning base stations deployed around the factory area via UWB positioning tags. The mobile positioning devices are smart bracelets equipped with vibration modules. The smart bracelet screen can display the current location information in real time, and indicate the direction of the nearest evacuation route in the form of arrows and text. Through the built-in voice module of the smart bracelet, it can receive and play key information transmitted by the sound column to help personnel quickly locate their own location. If personnel within the factory area discover any abnormal situations in their surroundings, including but not limited to blocked passages or injuries to other personnel, they can report these situations to the central control system via the SOS emergency button on their smart bracelets. Upon receiving the feedback, the central control system will promptly dispatch rescue forces for targeted handling.

Images