Fire-fighting equipment intelligent warehouse management system based on Internet of Things

Abstract

The invention discloses a fire-fighting equipment intelligent warehouse management system based on the Internet of Things, and relates to the technical field of warehouse management. A cross-regional data fusion module comprises a real-time fire monitoring unit, an environment situation sensing unit and an equipment state monitoring unit; the real-time fire monitoring unit acquires temperature, smoke and fire point coordinate data of a current area and adjacent areas with the radius of 50km in real time by deploying an NB-loT fire detection cluster and a satellite remote sensing receiving terminal, and performs area correlation calculation. According to the method, the correlation degree is calculated through the factors such as the area distance and the fire frequency, the potential relation between the areas is mined, the spatial attenuation coefficient is introduced to quantify the fire influence between the adjacent areas, cross-area fire data collaborative analysis is achieved, the problem of data islands is solved, and the fire occurrence probability in the subsequent different areas is predicted according to the analysis result, so that the fire risk is reduced. Therefore, the usage amount of the fire-fighting equipment can be conveniently predicted according to the fire occurrence probability, and the systematicness and efficiency of fire-fighting equipment management are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to an intelligent warehouse management system for fire-fighting equipment based on the Internet of Things. Background Art

[0002] Traditional fire-fighting equipment warehouse management relies on historical experience when purchasing, lacks scientific support from regional fire dynamics data, and the inventory volume does not match the regional risk level. There is a regional equipment shortage or surplus, and it is impossible to predict the consumption trend of fire-fighting equipment in the future, which has certain defects.

[0003] The defects of the existing fire equipment storage management system are:

[0004] 1. In the patent document CN113610416A, the fire-fighting equipment is managed in detail mainly through the management system of the Internet of Things to prevent the omission of fire-fighting equipment and facilitate the reserve management of fire-fighting equipment. However, it does not take into account that the existing fire-fighting equipment storage management system is prone to excess or shortage of fire-fighting equipment when purchasing fire-fighting equipment;

[0005] 2. In the application document CN118798856A, the main consideration is how to help managers fully understand the operation status of the park by presenting the analysis results and operation and maintenance guidance information to users in a graphical and three-dimensional visual way. It does not take into account that the existing fire equipment storage management system is not convenient for reasonably allocating fire equipment according to fire risks, and has poor practicality and flexibility;

[0006] 3. In the application document CN118840045A, the main consideration is how to take specific response management measures according to the warning level to reduce the supply chain risk, and it does not take into account the problem of low resource utilization in the existing fire equipment storage management system. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent storage management system for fire-fighting equipment based on the Internet of Things to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent warehouse management system for fire-fighting equipment based on the Internet of Things, comprising a cross-regional data fusion module, a dynamic risk assessment module, an equipment demand prediction module and an intelligent procurement decision module, wherein the cross-regional data fusion module is used to obtain fire data and regional conditions, the dynamic risk assessment module performs risk assessment based on the data obtained by the cross-regional data fusion module and outputs risk results, the equipment demand prediction module predicts fire-fighting equipment demand based on the data of the cross-regional data fusion module and the results of the dynamic risk assessment module, and the intelligent procurement decision module makes procurement decisions with reference to regional conditions, risk results and fire-fighting equipment demand;

[0009] The cross-regional data fusion module includes a real-time fire monitoring unit, an environmental situation awareness unit and an equipment status monitoring unit. The real-time fire monitoring unit acquires the temperature, smoke and fire point coordinate data of the current area and adjacent areas within a radius of 50km in real time by deploying an NB-IoT fire detection cluster and a satellite remote sensing receiving terminal, and performs regional correlation calculations.

[0010] Regional correlation calculation includes building a regional correlation calculation model ,in =3KM, is the area spacing, is the number of fires in area i in 30 days, is the number of fires in area j in 30 days.

[0011] Preferably, the dynamic risk assessment module uses a spatiotemporal convolutional neural network to receive a five-dimensional feature matrix, output a dynamic risk map, and annotate the fire probability values ​​for the next 7 days with a 500m×500m grid;

[0012] The input features of the five-dimensional feature matrix are: historical fire data, building structure characteristics, spatial topological relationships, equipment consumption characteristics, and real-time environmental parameters;

[0013] Historical fire data: contains not only the historical fire records of the current area, but also the conduction risk values ​​of adjacent areas;

[0014] Building structure characteristics: obtained by multiplying the density of high-rise buildings by the flammable material coefficient;

[0015] Spatial topological relationship: Inter-region distance attenuation coefficient = 1 / (1+0.2d 2 ), d is the area spacing;

[0016] Equipment consumption characteristics: amount of fire extinguishing agent used / unit fire area;

[0017] Real-time environmental parameters: wind speed × humidity × lightning warning level.

[0018] Preferably, the equipment demand prediction module includes establishing a multimodal equipment mapping model, and the establishment of the multimodal equipment mapping model includes a fire intensity-equipment type matching rule base, a burned area-equipment quantity conversion algorithm and a regional priority weight calculator, where is the real-time fire area, is the wind speed influence coefficient, The number of basic devices.

[0019] Preferably, the intelligent procurement decision module includes an inventory-demand balance monitoring unit, a supplier credit assessment and a blockchain smart contract, the inventory-demand balance monitoring unit includes an inventory intelligent early warning module, and the inventory intelligent early warning module purchases fire-fighting equipment by setting a dual threshold trigger mechanism;

[0020] Supplier credit assessment through formula Conduct credit assessment on suppliers, including is the indicator weight, is the standardized score, It is the sum of the weighted scores of the supplier on 20 indicators. The number of transactions with the supplier.

[0021] Preferably, the environmental situation awareness unit includes meteorological coupling analysis and population thermal tracking. The meteorological coupling analysis constructs a fire spread probability model by integrating wind speed, humidity and lightning location data. The population thermal tracking obtains the population density change trend through the operator base station to calculate the evacuation channel occupancy coefficient.

[0022] The equipment status monitoring unit includes the use of smart shelves and RFID technology to monitor the inventory and expiration date of fire-fighting equipment. The equipment status monitoring unit also includes the use of sensor technology to analyze the health status of fire-fighting equipment.

[0023] Preferably, the spatiotemporal convolutional neural network adopts a dual-channel architecture;

[0024] The first channel processes spatial features: a 3×3 convolution kernel is used to extract the fire propagation pattern between regions, and a graph attention mechanism is introduced to calculate the regional correlation;

[0025] The second channel processes time features: bidirectional LSTM is used to capture holiday / seasonal cycle patterns and embed time decay factors ,in is the number of days since the last fire.

[0026] Preferably, the multimodal device mapping model further includes special scenario response strategies, which include high-rise building scenarios, chemical park scenarios, and forest border areas;

[0027] High-rise building scenario: Automatically increase the ladder truck reserve weight to the baseline value (h / 30) 2 times, where h is the average building height;

[0028] Chemical Park Scenario: Setting the Minimum Foam Fire Extinguishing Agent Inventory , where V is the storage volume of hazardous chemicals;

[0029] Forest border area: Configure a mobile fire water tank network with node spacing meeting , where v is the spreading speed.

[0030] Preferably, the dual threshold trigger mechanism is as follows:

[0031] Yellow warning: Inventory quantity is less than 120% of the forecast demand, and a replenishment suggestion sheet is generated;

[0032] Red alert: Inventory level < 80% of forecast demand, emergency procurement agreement initiated.

[0033] Preferably, the blockchain smart contract writes the predicted demand data into the Hyperledger Fabric chain code to automatically generate a procurement contract;

[0034] Blockchain smart contract execution conditions: When the regional fire probability is >35% and the inventory reaches the red alert, the advance payment release is automatically triggered;

[0035] When the risk value of the adjacent area is > 0.7, the strategic reserve will be increased by 20% compulsorily;

[0036] Blockchain smart contract automatic triggering conditions:

[0037] When three adjacent regions issue yellow warnings at the same time, inter-provincial joint procurement is initiated;

[0038] When the equipment price index breaks through the upper Bollinger Band, switch to futures purchasing mode.

[0039] Preferably, the cross-regional data fusion module realizes data communication between different regions by establishing a regional data sharing channel.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention integrates and analyzes the collected data from different regions, calculates the correlation based on factors such as the regional spacing and the number of fires, explores the potential connections between regions, and introduces the spatial attenuation coefficient to quantify the impact of fires between adjacent regions, thereby realizing collaborative analysis of cross-regional fire data and solving the problem of data islands. The present invention predicts the probability of subsequent fires in different regions based on the results of the analysis, thereby facilitating the prediction of the use of fire-fighting equipment based on the probability of fire. To a certain extent, it can improve the systematicness and efficiency of fire-fighting equipment management, improve the scientificity, accuracy and efficiency of fire-fighting equipment management, and provide strong support for fire prevention and control and emergency rescue.

[0042] 2. The present invention fuses multi-dimensional features for analysis through a spatiotemporal convolutional neural network, deeply explores the fire propagation pattern and periodic laws from the spatial and temporal dimensions, generates a high-precision dynamic risk map, and can accurately and dynamically assess fire risks, thereby facilitating the reasonable deployment of fire-fighting equipment according to fire risks.

[0043] 3. The present invention constructs a multimodal equipment mapping model by comprehensively considering multiple factors through the equipment demand prediction module, including a fire intensity-equipment type matching rule base, a burned area-equipment quantity conversion algorithm, etc., to scientifically and accurately estimate the fire-fighting equipment demand in different scenarios, improve the rationality and effectiveness of equipment configuration, avoid resource waste or shortage, and at the same time, for special scenarios such as high-rise buildings, chemical parks, and forest border areas, corresponding equipment demand strategies and risk response measures are formulated, which can effectively meet the fire-fighting needs in diversified scenarios.

[0044] 4. The present invention evaluates suppliers through multiple indicators to ensure procurement quality and service. At the same time, it flexibly sets procurement trigger conditions based on regional warnings and price fluctuations, combines dynamic attenuation weight algorithm and equipment demand asymmetric response model to achieve intelligent adjustment of procurement strategies, reduce procurement costs, improve resource utilization efficiency, and ensure that fire-fighting equipment can be replenished in a timely and reasonable manner under different circumstances. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a system diagram of the present invention; DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] See also Figure 1, an embodiment provided by the present invention: an intelligent warehouse management system for fire-fighting equipment based on the Internet of Things, comprising a cross-regional data fusion module, a dynamic risk assessment module, an equipment demand prediction module and an intelligent procurement decision module, the cross-regional data fusion module is used to obtain fire data and regional conditions, the dynamic risk assessment module performs risk assessment based on the data obtained by the cross-regional data fusion module and outputs risk results, the equipment demand prediction module predicts fire-fighting equipment demand based on the data of the cross-regional data fusion module and the results of the dynamic risk assessment module, and the intelligent procurement decision module makes procurement decisions with reference to regional conditions, risk results and fire-fighting equipment demand;

[0048] The cross-regional data fusion module includes a real-time fire monitoring unit, an environmental situation awareness unit, and an equipment status monitoring unit. The real-time fire monitoring unit acquires the temperature, smoke, and fire point coordinate data of the current area and adjacent areas within a radius of 50km in real time by deploying an NB-IoT fire detection cluster and a satellite remote sensing receiving terminal, and performs regional correlation calculations.

[0049] Regional correlation calculation includes building a regional correlation calculation model ,in =3KM, is the area spacing, is the number of fires in area i in 30 days, is the number of fires in area j in 30 days.

[0050] The environmental situation awareness unit includes meteorological coupling analysis and population thermal tracking. The meteorological coupling analysis builds a fire spread probability model by integrating wind speed, humidity and lightning location data. The population thermal tracking obtains the population density change trend through the operator's base station to calculate the evacuation channel occupancy coefficient.

[0051] The equipment status monitoring unit includes the use of smart shelves and RFID technology to monitor the inventory and expiration date of fire-fighting equipment. The equipment status monitoring unit also includes the use of sensor technology to analyze the health status of fire-fighting equipment.

[0052] Furthermore, the NB-IoT fire detection cluster is widely deployed in various regions. It has the characteristics of high precision and low power consumption, and can monitor the changes of fire-related parameters such as temperature and smoke in the region in real time. The satellite remote sensing receiving terminal is connected to the satellite communication system, which can obtain macro data such as the coordinates of the fire points in adjacent areas within a radius of 50km, providing the system with a wider range of fire information. The regional correlation calculation model comprehensively considers the distance factors between regions and the frequency of fire occurrence factors. This shows that the closer the distance, the higher the correlation between regions, and the influence of distance decays exponentially. This reflects that there is a high degree of correlation between areas with similar fire occurrences. This model can accurately quantify the degree of correlation between different areas, providing an important basis for subsequent risk assessment and decision-making. For example, in a city and its surrounding areas, if two adjacent towns are close to each other, If the number of recent fires is similar, then their correlation under the model will be high, which means that in terms of fire prevention and control and equipment management, these two areas need to be considered in a coordinated manner. When a fire occurs in one area, the other area may be affected, and the system can respond in advance and allocate resources.

[0053] The environmental situation awareness unit provides comprehensive and in-depth environmental information for fire prevention and control through meteorological coupling analysis and population thermal tracking. Meteorological coupling analysis integrates wind speed, humidity and lightning location data to build a fire spread probability model, which can accurately predict the spread trend of fire under different meteorological conditions. For example, in dry, windy and thunderous weather, the model can quickly assess the possibility of fire occurrence and spread, and provide a basis for the fire department to deploy firefighting forces in advance. Population thermal tracking uses the operator's base station to obtain the trend of population density changes and calculate the occupancy coefficient of the evacuation channel, so that firefighters can understand the difficulty of evacuation in advance when a fire occurs, so as to reasonably plan evacuation routes and rescue plans to maximize the safety of personnel.

[0054] The equipment status monitoring unit uses smart shelves and RFID technology to achieve real-time monitoring of the inventory and validity period of fire-fighting equipment, ensuring that the number of fire-fighting equipment is sufficient and in an available state, avoiding the impact of equipment expiration or shortage on fire-fighting and rescue work. At the same time, sensor technology is used to analyze the health status of fire-fighting equipment, timely discover potential equipment failures, arrange maintenance and replacement in advance, and ensure that the equipment operates normally at any time. For example, sensors are used to monitor the pressure of fire extinguishers and the integrity of fire hoses, so that firefighters can know the status of the equipment clearly.

[0055] When in use, the area can be divided based on the fire station, and the occurrence of fires in different areas and the use of fire-fighting equipment can be collected. Then the use of fire-fighting equipment in the subsequent period of time can be predicted, and fire-fighting equipment can be reasonably purchased based on the predicted results.

[0056] See also Figure 1 , an embodiment provided by the present invention: an intelligent storage management system for fire fighting equipment based on the Internet of Things, a dynamic risk assessment module adopts a spatiotemporal convolutional neural network to receive a five-dimensional feature matrix, outputs a dynamic risk map, and annotates the fire probability value in the next 7 days with a 500m×500m grid;

[0057] The input features of the five-dimensional feature matrix are: historical fire data, building structure characteristics, spatial topological relationships, equipment consumption characteristics, and real-time environmental parameters;

[0058] Historical fire data: contains not only the historical fire records of the current area, but also the conduction risk values ​​of adjacent areas;

[0059] Building structure characteristics: obtained by multiplying the density of high-rise buildings by the flammable material coefficient;

[0060] Spatial topological relationship: Inter-region distance attenuation coefficient = 1 / (1+0.2d 2 ), d is the area spacing;

[0061] Equipment consumption characteristics: amount of fire extinguishing agent used / unit fire area;

[0062] Real-time environmental parameters: wind speed × humidity × lightning warning level.

[0063] The spatiotemporal convolutional neural network adopts a dual-channel architecture;

[0064] The first channel processes spatial features: a 3×3 convolution kernel is used to extract the fire propagation pattern between regions, and a graph attention mechanism is introduced to calculate the regional correlation;

[0065] The second channel processes time features: bidirectional LSTM is used to capture holiday / seasonal cycle patterns and embed time decay factors ,in is the number of days since the last fire.

[0066] Furthermore, historical fire data not only includes historical fire records in the current area, but also includes risk values ​​transmitted from adjacent areas. These historical data are an important basis for systematic analysis of fire occurrence patterns and trends. The transmission risk values ​​of adjacent areas take into account the impact of fire propagation between regions, making risk assessment more comprehensive. Building structure characteristics are obtained by multiplying the density of high-rise buildings by the flammable material coefficient. The density of high-rise buildings reflects the concentration of high-rise buildings in the region. Due to their height and structural characteristics, fire prevention and control of high-rise buildings is more difficult. The flammable material coefficient is related to the flammability of the materials used in the building. The combination of the two can intuitively reflect the impact of building structure on fire risk. For example, in a commercial area, high-rise buildings are dense and some buildings use flammable decorative materials. Then the building structure characteristic value of the area will be higher, and the fire risk will increase accordingly.

[0067] Real-time environmental parameters: Real-time environmental factors such as wind speed, humidity, and lightning warning level have a significant impact on the occurrence and spread of fire. The greater the wind speed, the faster the fire may spread; when the humidity is low, the environment is more prone to fire; a high lightning warning level means that the possibility of lightning causing a fire is increased. The system obtains these parameters in real time and incorporates them into the risk assessment model, which can adjust the fire risk prediction in time according to environmental changes.

[0068] The spatial topological relationship is reflected by the distance attenuation coefficient between regions. The closer the distance between regions, the higher the degree of mutual influence. In the fire risk assessment, the system can more accurately analyze the spatial correlation between different regions and provide strong support for the fire propagation risk assessment. The equipment consumption characteristics are measured by the amount of fire extinguishing agent / unit fire area, which reflects the actual use of firefighting equipment in past fires. It can intuitively reflect the complexity of fires in different regions and the demand for firefighting equipment. For example, in a fire in a chemical park, due to the special nature of the fire, a large amount of fire extinguishing agent is required, and the equipment consumption characteristic value will be high, making the demand for firefighting equipment in the area under future fire risks greater.

[0069] The first channel of the spatiotemporal convolutional neural network uses a 3×3 convolution kernel to extract the inter-regional fire propagation pattern, which is used to capture the propagation law and characteristics of fire between different regions. At the same time, the graph attention mechanism is introduced to calculate the regional correlation, so that the model can pay more attention to other regions that are closely related to the current region, and further optimize the understanding of fire propagation. For example, when analyzing urban fire risks, the graph attention mechanism can focus on areas adjacent to and highly correlated with high-incidence fire areas, and take preventive measures in advance. The second channel uses a bidirectional LSTM to capture holiday / seasonal cycle laws. LSTM can process time series data, remember past information and use it for current decision-making. In fire risk assessment, holidays and seasonal changes are often related to the frequency and characteristics of fires. For example, during holidays, the flow of people increases, commercial activities are frequent, and the risk of fire may increase. In winter, due to the dry climate and increased use of fire and electricity, fires are also more frequent. Bidirectional LSTM can analyze time series from both the past and future directions to more accurately capture these cycle laws. In addition, the time decay factor is embedded. As time goes by, the impact of the last fire on the current risk gradually weakens, allowing the model to dynamically adapt to time changes.

[0070] The dynamic risk map generated by the output layer marks the fire probability values ​​for the next 7, 15 or 30 days in units of 500m×500m grids. This high-precision grid division makes the risk assessment results more detailed. Managers can clearly understand the fire risk situation in each small area, and provide accurate guidance for the formulation of targeted prevention and control measures and resource allocation. For example, in urban planning, fire-fighting facilities can be reasonably arranged according to the risk map, and fire stations can be added or more fire-fighting equipment can be stored in grid areas with higher fire probabilities.

[0071] See also Figure 1 , an embodiment provided by the present invention: an intelligent storage management system for fire fighting equipment based on the Internet of Things, the equipment demand prediction module includes establishing a multimodal equipment mapping model, and establishing a multimodal equipment mapping model includes a fire intensity-equipment type matching rule base, a burned area-equipment quantity conversion algorithm and a regional priority weight calculator, where is the real-time fire area, is the wind speed influence coefficient, The number of basic devices.

[0072] The multimodal device mapping model also includes special scenario response strategies, including high-rise building scenarios, chemical park scenarios, and forest border areas;

[0073] High-rise building scenario: Automatically increase the ladder truck reserve weight to the baseline value (h / 30) 2 times, where h is the average building height;

[0074] Chemical Park Scenario: Setting the Minimum Foam Fire Extinguishing Agent Inventory , where V is the storage volume of hazardous chemicals;

[0075] Forest border area: Configure a mobile fire water tank network with node spacing meeting , where v is the spreading speed.

[0076] Furthermore, the fire intensity-equipment type matching rule library: the system has pre-established a detailed matching rule library to clarify the type of fire-fighting equipment to be used according to the different fire intensities. For example, for small fires, only small fire-fighting equipment such as fire extinguishers are needed, while for large-scale fires, large equipment such as fire trucks and ladder trucks are required, ensuring that the appropriate equipment can be accurately selected in different fire situations.

[0077] Burnt area-equipment quantity conversion algorithm: It comprehensively considers the real-time burned area and wind speed influence coefficient. The real-time burned area reflects the scale of the fire, and the wind speed influence coefficient considers the impact of wind speed on the spread of fire and equipment demand. Through integral calculation, the number of equipment required in a period of time can be dynamically calculated, and then added with the number of basic equipment to obtain the final equipment demand forecast value. For example, in a forest fire, as the burned area expands and the wind speed changes, the system can adjust the equipment demand forecast in real time to provide accurate equipment support for firefighting operations.

[0078] Regional priority weight calculator: Taking into account the particularity of sensitive areas such as hospitals and schools, a regional priority weight calculator is set. These sensitive areas are densely populated and difficult to evacuate. Once a fire occurs, the consequences are serious. Therefore, they will be given higher weights in the equipment demand forecast. For example, in a city, if the risk of fire in the school area increases, the system will automatically increase the fire equipment demand forecast in the area to give priority to ensuring fire safety in these areas.

[0079] High-rise building scenario: As high-rise building fires are difficult to extinguish, the system automatically increases the ladder truck reserve weight to the baseline value (h / 30) 2times, this is because the higher the high-rise building is, the greater the demand for ladder trucks. This way, it can ensure that there are enough ladder trucks for rescue when a fire occurs in a high-rise building.

[0080] Chemical Park Scenario: Chemical parks store a large amount of hazardous chemicals, which poses a high risk of fire. Therefore, a minimum amount of foam fire extinguishing agent is set. , which makes it convenient to determine the minimum amount of foam fire extinguishing agent required based on the reserves of hazardous chemicals to meet the basic needs of fire fighting in chemical parks.

[0081] Forest border area: Fire spreads quickly in forest border areas. Mobile fire water tank network is configured, and the node spacing meets By rationally arranging mobile fire water tanks, water can be quickly provided when a fire occurs, effectively controlling the spread of the fire.

[0082] See also Figure 1 , an embodiment provided by the present invention: an intelligent warehouse management system for fire-fighting equipment based on the Internet of Things, wherein the intelligent procurement decision module includes an inventory-demand balance monitoring unit, a supplier credit evaluation and a blockchain smart contract, the inventory-demand balance monitoring unit includes an inventory intelligent early warning module, and the inventory intelligent early warning module purchases fire-fighting equipment by setting a dual threshold trigger mechanism;

[0083] Supplier credit assessment through formula Conduct credit assessment on suppliers, including is the indicator weight, is the standardized score, It is the sum of the weighted scores of the supplier on 20 indicators. The number of transactions with the supplier.

[0084] The dual threshold trigger mechanism is as follows:

[0085] Yellow warning: Inventory quantity is less than 120% of the forecast demand, and a replenishment suggestion sheet is generated;

[0086] Red alert: Inventory level < 80% of forecast demand, emergency procurement agreement initiated.

[0087] The blockchain smart contract writes the predicted demand data into the Hyperledger Fabric chain code to automatically generate a procurement contract;

[0088] Blockchain smart contract execution conditions: When the regional fire probability is >35% and the inventory reaches the red alert, the advance payment release is automatically triggered;

[0089] When the risk value of the adjacent area is > 0.7, the strategic reserve will be increased by 20% compulsorily;

[0090] Blockchain smart contract automatic triggering conditions:

[0091] When three adjacent regions issue yellow warnings at the same time, inter-provincial joint procurement is initiated;

[0092] When the equipment price index breaks through the upper Bollinger Band, switch to futures purchasing mode.

[0093] Furthermore, the inventory intelligent early warning module adopts a dual-threshold trigger mechanism to accurately manage the procurement process of fire-fighting equipment. This mechanism fully considers the dynamic relationship between inventory and predicted demand, and can respond promptly according to different inventory levels, ensuring the flexibility and timeliness of inventory management.

[0094] Yellow warning: When the inventory level is lower than 120% of the forecast demand, the system will issue a yellow warning and generate a replenishment suggestion. This threshold setting is designed to remind the procurement department to pay attention to inventory changes in advance, provide buffer time for the normal procurement process, and avoid excessive inventory consumption. For example, if a region predicts that 100 fire extinguishers will be needed in the future based on fire risk assessment, when the inventory drops below 120, the system will automatically generate a replenishment suggestion to prompt the purchase of an appropriate amount of fire extinguishers to maintain inventory levels.

[0095] Red alert: Once the inventory level is less than 80% of the forecast demand, the system immediately initiates a red alert and triggers an emergency procurement agreement. This is an emergency measure when the inventory is at risk of severe shortage, ensuring that firefighting equipment can be quickly replenished in an emergency. Taking the fire extinguisher mentioned above as an example, when the inventory level drops below 80, the system will quickly initiate an emergency procurement procedure and negotiate with suppliers to speed up the delivery to ensure the smooth progress of firefighting work.

[0096] Supplier credit assessment comprehensively considers the supplier's performance in 20 indicators, covering product quality, delivery time, price, after-sales service and other aspects. Each indicator has a corresponding weight, reflecting its importance in the overall assessment. The standardized score is a quantification of the supplier's actual performance in each indicator. By calculating the sum of weighted scores, the supplier's comprehensive performance under conventional indicators can be obtained. At the same time, the number of supplier transactions is introduced. The more transactions, the longer the supplier's cooperation history with the system and the higher the stability, which will also have a positive impact on its credit assessment. This evaluation method helps to screen out suppliers with good credit and reliability, and provide guarantees for the procurement quality and supply stability of fire equipment.

[0097] Blockchain smart contracts use Hyperledger Fabric chain code technology to automatically write predicted demand data into and generate procurement contracts. This approach ensures that the contract generation process is transparent and cannot be tampered with, improves the credibility and execution efficiency of the contract. Once the predicted demand is determined, the system will automatically generate contract terms according to preset rules, clarify key information such as the type, quantity, price, and delivery time of the purchased equipment, and avoid errors and disputes that may be caused by manual operations.

[0098] Prepayment release: When the probability of fire in a region exceeds 35% and the inventory reaches a red alert, the blockchain smart contract will automatically trigger the release of prepayments. This mechanism takes into account the urgency of fire risks and the severity of inventory shortages, ensuring that suppliers can obtain financial support in a timely manner in an emergency, speeding up production and supply to meet firefighting needs. For example, if the probability of fire in a certain area rises to 40% in the near future, and the inventory of firefighting equipment is less than 80% of the predicted demand, the system will automatically release prepayments to suppliers to promote the rapid progress of the procurement process.

[0099] When the risk value of the adjacent area is greater than 0.7, the system will force a 20% increase in strategic reserves. The high risk of the adjacent area may mean that the fire has a tendency to spread. Increasing strategic reserves can improve the region's ability to respond to potential fire threats. If a large-scale fire occurs in an adjacent city and the risk value reaches 0.8, the intelligent procurement decision-making module in the region will automatically adjust the procurement plan and increase the fire equipment reserves by 20% to cope with the possible spread of the fire.

[0100] Interprovincial joint procurement: When three adjacent regions issue a yellow warning at the same time, the system activates the interprovincial joint procurement mechanism. This cooperation mode can integrate resources from different regions, improve procurement efficiency, and reduce procurement costs. For example, if the inventory of fire equipment in three adjacent regions or provinces is close to the warning level at the same time, through interprovincial joint procurement, each region can negotiate with suppliers to obtain more favorable prices and better supply conditions.

[0101] Switching to futures purchasing mode: When the equipment price index breaks through the upper Bollinger Band, the system automatically switches to the futures purchasing mode, which can avoid the cost increase caused by future price increases to a certain extent. For example, the market price of fire-fighting equipment fluctuates greatly. When the price index exceeds the upper Bollinger Band, the system will immediately sign a futures purchasing contract with the supplier to ensure that the required equipment is obtained at the current relatively low price.

[0102] The various trigger conditions and automatic execution functions of blockchain smart contracts can respond quickly to different risk situations and market changes, optimize procurement strategies, reduce procurement costs, and provide strong material support for firefighting work.

[0103] See also Figure 1, an embodiment provided by the present invention: an intelligent storage management system for fire fighting equipment based on the Internet of Things, a cross-regional data fusion module realizes data communication between different regions by establishing a regional data sharing channel.

[0104] Furthermore, the cross-regional data fusion module achieves data communication between different regions by establishing regional data sharing channels, integrating fire-related data scattered in various regions, and providing comprehensive and accurate data support for subsequent risk assessments, equipment demand forecasts, and procurement decisions, thereby improving the coordination and effectiveness of the entire fire management system.

[0105] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An intelligent storage management system for fire fighting equipment based on the Internet of Things, characterized by: It includes a cross-regional data fusion module, a dynamic risk assessment module, an equipment demand prediction module and an intelligent procurement decision module. The cross-regional data fusion module is used to obtain fire data and regional conditions. The dynamic risk assessment module performs risk assessment based on the data obtained by the cross-regional data fusion module and outputs risk results. The equipment demand prediction module predicts the demand for fire equipment based on the data of the cross-regional data fusion module and the results of the dynamic risk assessment module. The intelligent procurement decision module makes procurement decisions based on regional conditions, risk results and fire equipment demand; The cross-regional data fusion module includes a real-time fire monitoring unit, an environmental situation awareness unit and an equipment status monitoring unit. The real-time fire monitoring unit acquires the temperature, smoke and fire point coordinate data of the current area and adjacent areas within a radius of 50km in real time by deploying an NB-IoT fire detection cluster and a satellite remote sensing receiving terminal, and performs regional correlation calculations. Regional correlation calculation includes building a regional correlation calculation model ,in =3KM, is the area spacing, is the number of fires in area i in 30 days, is the number of fires in area j in 30 days.

2. According to the Internet of Things-based intelligent storage management system for firefighting equipment according to claim 1, it is characterized by: The dynamic risk assessment module uses a spatiotemporal convolutional neural network to receive a five-dimensional feature matrix, output a dynamic risk map, and annotate the fire probability value in the next 7 days with a 500m×500m grid; The input features of the five-dimensional feature matrix are: historical fire data, building structure characteristics, spatial topological relationships, equipment consumption characteristics, and real-time environmental parameters; Historical fire data: contains not only the historical fire records of the current area, but also the conduction risk values ​​of adjacent areas; Building structure characteristics: obtained by multiplying the density of high-rise buildings by the flammable material coefficient; Spatial topological relationship: Inter-region distance attenuation coefficient = 1 / (1+0.2d 2 ), d is the area spacing; Equipment consumption characteristics: amount of fire extinguishing agent used / unit fire area; Real-time environmental parameters: wind speed × humidity × lightning warning level.

3. According to the Internet of Things-based intelligent storage management system for firefighting equipment according to claim 1, it is characterized by: The equipment demand prediction module includes establishing a multimodal equipment mapping model, which includes a fire intensity-equipment type matching rule base and a burned area-equipment quantity conversion algorithm. and a regional priority weight calculator, where is the real-time fire area, is the wind speed influence coefficient, The number of basic devices.

4. According to the Internet of Things-based intelligent storage management system for firefighting equipment according to claim 1, it is characterized by: The intelligent procurement decision module includes an inventory-demand balance monitoring unit, a supplier credit assessment and a blockchain smart contract. The inventory-demand balance monitoring unit includes an inventory intelligent early warning module. The inventory intelligent early warning module purchases fire-fighting equipment by setting a dual threshold trigger mechanism. Supplier credit assessment through formula Conduct credit assessment on suppliers, including is the indicator weight, is the standardized score, It is the sum of the weighted scores of the supplier on 20 indicators. The number of transactions with the supplier.

5. According to the Internet of Things-based intelligent storage management system for firefighting equipment according to claim 1, it is characterized by: The environmental situation awareness unit includes meteorological coupling analysis and population thermal tracking. The meteorological coupling analysis constructs a fire spread probability model by integrating wind speed, humidity and lightning location data. The population thermal tracking obtains the population density change trend through the operator base station to calculate the evacuation channel occupancy coefficient. The equipment status monitoring unit includes the use of smart shelves and RFID technology to monitor the inventory and expiration date of fire-fighting equipment. The equipment status monitoring unit also includes the use of sensor technology to analyze the health status of fire-fighting equipment.

6. According to claim 2, the intelligent storage management system for fire fighting equipment based on the Internet of Things is characterized by: The spatiotemporal convolutional neural network adopts a dual-channel architecture; The first channel processes spatial features: a 3×3 convolution kernel is used to extract the fire propagation pattern between regions, and a graph attention mechanism is introduced to calculate the regional correlation; The second channel processes time features: bidirectional LSTM is used to capture holiday / seasonal cycle patterns and embed time decay factors ,in is the number of days since the last fire.

7. According to claim 3, the intelligent storage management system for fire fighting equipment based on the Internet of Things is characterized by: The multimodal device mapping model also includes special scenario response strategies, which include high-rise building scenarios, chemical park scenarios, and forest border areas; High-rise building scenario: Automatically increase the ladder truck reserve weight to the baseline value (h / 30) 2 times, where h is the average building height; Chemical Park Scenario: Setting the Minimum Foam Fire Extinguishing Agent Inventory , where V is the storage volume of hazardous chemicals; Forest border area: Configure a mobile fire water tank network with node spacing meeting , where v is the spreading speed.

8. According to claim 4, the intelligent storage management system for fire fighting equipment based on the Internet of Things is characterized by: The dual threshold trigger mechanism is as follows: Yellow warning: Inventory quantity is less than 120% of the forecast demand, and a replenishment suggestion sheet is generated; Red alert: Inventory level < 80% of forecast demand, emergency procurement agreement initiated.

9. The intelligent storage management system for fire fighting equipment based on the Internet of Things according to claim 8 is characterized by: The blockchain smart contract writes the predicted demand data into the Hyperledger Fabric chain code to automatically generate a procurement contract; Blockchain smart contract execution conditions: When the regional fire probability is >35% and the inventory reaches the red alert, the advance payment release is automatically triggered; When the risk value of the adjacent area is > 0.7, the strategic reserve will be increased by 20% compulsorily; Blockchain smart contract automatic triggering conditions: When three adjacent regions issue yellow warnings at the same time, inter-provincial joint procurement is initiated; When the equipment price index breaks through the upper Bollinger Band, switch to futures purchasing mode.

10. The intelligent storage management system for fire fighting equipment based on the Internet of Things according to claim 1 is characterized in that: The cross-regional data fusion module realizes data communication between different regions by establishing a regional data sharing channel.

Images