Intelligent fire-fighting fire prediction method and system, medium and program product

Abstract

The invention provides an intelligent fire-fighting fire prediction method and system, a medium and a program product, and relates to the technical field of electric digital data processing. The method comprises the following steps: firstly, acquiring real-time smoke monitoring data of a smoke alarm sensor in a target building, and once the data reaches a preset threshold value, immediately determining fire occurrence and locking a target fire source range; thirdly, calculating a fire risk value of the range by means of a fire risk assessment model trained based on multiple factors of a target building and a machine learning algorithm, and determining that a high-risk fire source point exists when the fire risk value is higher than a threshold value; then, combining meteorological data to obtain multi-dimensional environment characteristic data, and inputting the high-risk fire source point and the meteorological data into a fire spreading prediction model trained by a deep learning algorithm to obtain a fire spreading prediction condition; and finally, the situation is sent to a management end for alarming, and according to the method, rapid and accurate pre-judgment of the fire situation is achieved by means of multi-source data and an intelligent model.

Description

Technical Field

[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a fire intelligent prediction method, system, medium and program product. Background Art

[0002] In modern society, the number and size of buildings are increasing. Fire, as a common and serious disaster, poses a huge threat to the life and property safety of people in buildings. Timely and accurate detection of fires and taking effective countermeasures are crucial to reducing fire losses. Therefore, how to efficiently monitor and predict fires has become an important research direction in the field of building safety.

[0003] At present, building fire protection mainly adopts the method of installing smoke alarms and fire-fighting facilities in buildings. These smoke alarms will sound an alarm when the smoke concentration exceeds the preset threshold, and notify the building management personnel to take corresponding fire-fighting measures. After receiving the alarm, the management personnel usually need to judge the development of the fire based on experience and decide on the rescue plan.

[0004] However, this method of relying solely on smoke alarms and human experience has its limitations. Due to the complex structure of buildings, factors such as building materials and firefighting facilities in different areas will affect the spread of fires. In addition, the changes in external weather conditions make it difficult to accurately predict the development of fires based on human experience alone, which can easily lead to inaccurate rescue measures and affect the effectiveness of fire control. Summary of the invention

[0005] The present application provides a fire intelligent prediction method, system, medium and program product for intelligently predicting fires in buildings.

[0006] In the first aspect, the present application provides a fire intelligent prediction method for fire fighting, which is applied to an intelligent prediction system, and the method includes: obtaining real-time smoke monitoring data sent by a smoke alarm sensor pre-installed in a target building; when the real-time smoke monitoring data reaches a preset smoke data threshold, determining that a fire has occurred in the target building, and determining the target fire source range where the fire has occurred based on the real-time smoke monitoring data; obtaining a fire risk value for the target fire source range based on a pre-set fire risk assessment model, the fire risk assessment model is pre-trained based on the structural layout of different areas in the target building, the flammability of building materials, the equipment of fire-fighting facilities, and the corresponding fire risk value annotations combined with a machine learning algorithm, and the fire risk assessment model is pre-trained based on the structural layout of different areas in the target building, the flammability of building materials, the equipment of fire-fighting facilities, and the corresponding fire risk value annotations combined with a machine learning algorithm. The model is used to evaluate the fire risk values ​​of different areas of the target building; if the fire risk value is higher than the preset risk threshold, it is judged that there is a high-risk fire source point within the target fire source range; the current multi-dimensional environmental characteristic data is obtained in combination with meteorological data, and the multi-dimensional environmental characteristic data at least includes meteorological combination data; the high-risk fire source point and the meteorological combination data are input into the preset fire spread prediction model to obtain the fire spread prediction situation. The fire spread prediction model is trained in advance by a deep learning algorithm based on a large number of fire experimental data and actual fire case review data under different meteorological scenarios. The fire spread prediction model is used to predict the fire spread path, speed and smoke diffusion direction in real time; the fire spread prediction situation is sent to the management end for alarm.

[0007] By adopting the above technical solutions, the system uses real-time monitoring data from smoke alarm sensors to identify fires early, and combines fire risk assessment models to analyze the characteristics of different areas of the building to promptly identify high-risk fire sources. At the same time, the system integrates meteorological combination data into the fire spread prediction model, which is trained based on a large amount of experimental and case data and can accurately predict the fire spread path, speed and smoke diffusion direction. This multi-dimensional data fusion and deep learning prediction method significantly improves the accuracy and timeliness of fire warnings, provides a scientific decision-making basis for fire prevention and control, and effectively reduces the risk of casualties and property losses caused by fires.

[0008] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, it also includes: using a geographic information system to visualize the target building and the environment within a set surrounding range, and the visualization includes marking the location of the high-risk fire source point and the fire spread prediction situation on an electronic map; automatically screening the fire rescue station that is closest and has corresponding rescue capabilities according to a preset emergency rescue plan; and sending the location of the high-risk fire source point, the fire spread prediction situation and the multi-dimensional environmental characteristic data to the fire rescue station.

[0009] By adopting the above technical solutions, the geographic information system is used to realize the visualization of fire conditions, intuitively presenting the location of high-risk fire sources and the prediction of fire spread. The system automatically selects the nearest fire station with corresponding rescue capabilities and sends fire-related information to it. This visualization combined with intelligent screening has greatly improved the response speed and decision-making efficiency of fire rescue, enabling rescue personnel to understand the fire situation at the first time and formulate the best rescue plan, effectively improving the success rate and efficiency of fire rescue.

[0010] In combination with some embodiments of the first aspect, in some embodiments, when the real-time smoke monitoring data reaches a preset smoke data threshold, it is determined that a fire has occurred in the target building, and after the step of determining the target fire source range where the fire has occurred based on the real-time smoke monitoring data, it also includes: obtaining a thermal imaging image of the fire area through a thermal imaging device installed in the target building; locating the boundary and position of the fire source through the thermal imaging image and the real-time smoke monitoring data; and correcting the target fire source range based on the boundary and position of the fire source.

[0011] By adopting the above technical solutions, combined with the thermal imaging images and smoke monitoring data obtained by thermal imaging equipment, the system can accurately locate the boundaries and positions of the fire source. Thermal imaging technology can penetrate the smoke to directly observe the fire source, and verify it with the smoke monitoring data. The combination of the two data greatly improves the accuracy of fire source positioning. By dynamically correcting the range of the target fire source, the system can more accurately assess the development trend of the fire, provide more reliable basic data for subsequent fire spread prediction, and improve the reliability of the entire early warning system.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after the step of obtaining the fire risk value of the target fire source range according to a pre-set fire risk assessment model, it also includes: collecting device data of Internet of Things devices in the target building, and the Internet of Things devices include at least temperature sensor data, electrical equipment and ventilation equipment; integrating the device data with the real-time smoke monitoring data, and making real-time dynamic adjustments to the fire risk value corresponding to the target fire source range.

[0013] By adopting the above technical solutions, the system collects and integrates data from IoT devices such as temperature sensors, electrical equipment and ventilation equipment in the building, and conducts comprehensive analysis of these data with smoke monitoring data. This multi-source data fusion analysis method enables the system to more comprehensively perceive changes in the fire environment and dynamically adjust the fire risk value in real time, thereby improving the accuracy and timeliness of risk assessment and providing a more reliable basis for fire warning.

[0014] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, it also includes: correlating the current fire scene with historical fire cases; by comparing the development process and results of historical fire cases in a scene matching the fire scene, the fire spread prediction situation is adjusted accordingly, and the fire scene includes various characteristics and meteorological data of the target fire source range.

[0015] By adopting the above technical solution, the system will analyze the correlation between the current fire scene and historical cases, and optimize the prediction results by comparing the historical fire development process and results under similar scenes. This intelligent analysis method based on historical experience makes full use of the empirical data of existing fire cases, which can effectively improve the accuracy of fire spread prediction, make the prediction results more in line with the actual situation, and provide a more reliable reference for fire prevention and control decisions.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, it also includes: determining the distance between the location of the target building and the target area; determining the spread time of the fire to spread to the target area based on the fire spread prediction situation and the distance; after receiving a spread time viewing instruction sent by the user terminal, sending the spread time to the user terminal for display.

[0017] By adopting the above technical solutions, the system can accurately calculate the time it takes for a fire to spread to the target area based on the fire spread prediction results and spatial distance information. When the user needs to view it, the system can provide this key information immediately. This precise time prediction function buys valuable time for evacuation and rescue, allowing relevant personnel to take necessary protective measures in advance, effectively reducing the harm caused by the fire and improving the efficiency of emergency response.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of sending the fire spread prediction situation to the management end for alarm, it also includes: converting the fire-related information into voice commands and broadcasting them through the broadcast settings in the target building, the fire-related information includes the target fire source range and the fire spread prediction situation.

[0019] By adopting the above technical solution, the system converts fire-related information into voice commands and broadcasts them through the broadcasting system in the building. This information conversion and automatic broadcasting mechanism enables fire information to be quickly and clearly conveyed to all personnel in the building, improving the efficiency and coverage of information transmission. At the same time, the form of voice broadcast is easier for personnel to receive and understand, effectively improving the actual effect of fire warning and winning precious time for personnel evacuation.

[0020] In a second aspect, the present application provides an intelligent prediction system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the intelligent prediction system to perform the method described in the first aspect and any possible implementation of the first aspect.

[0021] In a third aspect, the present application provides a computer-readable storage medium comprising instructions, which, when executed on an intelligent prediction system, causes the intelligent prediction system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, the present application provides a computer program product, which, when executed on an intelligent prediction system, enables the intelligent prediction system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The use of a machine learning-based fire risk assessment model to intelligently analyze buildings, combined with a deep learning-trained fire spread prediction model for multi-dimensional prediction, effectively solves the technical problem in the existing technology that a single smoke alarm cannot accurately assess fire risks and is difficult to scientifically predict the development of fires. It also realizes the intelligence of fire warning and the accuracy of prediction, greatly improves the scientific decision-making ability of fire prevention and control, and significantly reduces the risk of casualties and property losses caused by fires.

[0024] 2. The use of geographic information systems to achieve visual display of fire conditions and the use of technical means to intelligently screen the best fire stations based on preset emergency rescue plans effectively solves the technical problems of non-intuitive fire information transmission and lack of scientific basis for rescue station selection in existing technologies, thereby achieving intuitive presentation of the fire situation and optimal deployment of rescue forces, significantly improving the response speed and decision-making efficiency of fire rescue.

[0025] 3. Due to the use of technical means for intelligent calculation based on fire spread prediction results and spatial distance information, and accurate prediction of fire spread time, the technical problems in the existing technology that the fire spread time cannot be accurately predicted and sufficient time is difficult to reserve for evacuation are effectively solved, thereby achieving accurate prediction of fire spread time, providing reliable time guarantee for personnel evacuation and emergency response, and greatly improving the initiative and effectiveness of fire prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of the fire intelligent prediction method in the embodiment of the present application; Figure 2 This is another flow chart of the fire intelligent prediction method in the embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a physical device of the intelligent prediction system in the embodiment of the present application. DETAILED DESCRIPTION

[0027] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0028] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0029] For ease of understanding, the following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of the fire intelligent prediction method in the embodiment of the present application.

[0030] S101, obtaining real-time smoke monitoring data sent by a smoke alarm sensor pre-installed in a target building; After the intelligent prediction system is started, the first task is to establish a stable and reliable connection with the smoke alarm sensors in the target building. This is the basis for obtaining real-time smoke monitoring data. These sensors are distributed in various key areas of the building, such as corridors, stairwells, offices, warehouses, etc. Their layout has been carefully designed to ensure that any possible signs of smoke can be fully and timely monitored. Take a large commercial complex as an example. Its internal space is complex and contains many shops, restaurants, entertainment venues, etc. In order to achieve no-dead-angle monitoring, smoke alarm sensors will be reasonably arranged according to the functions and structural characteristics of different areas. In the store, sensors will be installed on the ceiling near the possible location of the fire source; in the corridor, one is set at a certain distance to ensure that the smoke can be quickly detected. The intelligent prediction system uses advanced wireless communication protocols, such as ZigBee, Wi-Fi or Bluetooth, to ensure that the data collected by the sensors can be quickly and accurately transmitted back to the system. These protocols have the characteristics of low power consumption, high reliability and strong anti-interference ability, which can effectively avoid data loss or transmission delay. At the same time, in order to deal with possible communication failures, the system also sets up multiple backup mechanisms. For example, when a problem occurs in one communication method, the sensor will automatically switch to the backup communication channel to ensure the continuity of data transmission.

[0031] S102: When the real-time smoke monitoring data reaches a preset smoke data threshold, it is determined that a fire has occurred in the target building, and a target fire source range of the fire is determined based on the real-time smoke monitoring data; The intelligent prediction system continuously receives data from smoke alarm sensors and analyzes the data in real time. The preset smoke data threshold is an important basis for judging whether a fire has occurred. This threshold is not fixed, but is set comprehensively based on multiple factors such as the type and purpose of the building and fire safety standards. For ordinary residential buildings, considering that daily activities are less likely to generate smoke, the preset threshold will be relatively low so that fire hazards can be discovered as early as possible; for some industrial plants, since a certain amount of smoke may be generated during the production process, the preset threshold will be increased accordingly to avoid unnecessary panic to residents due to misjudgment.

[0032] Once the monitoring data reaches the preset threshold, the system will quickly start the fire confirmation procedure. To prevent false alarms, the system will combine the data of multiple sensors for cross-verification. If multiple adjacent sensors detect that the smoke concentration exceeds the standard at the same time, or a sensor detects a sharp increase in smoke concentration in a short period of time, the system will be more confident that a fire has occurred. In a high-rise office building, if the smoke alarm sensors on multiple floors detect that the smoke concentration exceeds the threshold almost at the same time, the intelligent prediction system can quickly determine that a fire has occurred in the office building.

[0033] After confirming that a fire has occurred, the system will then determine the scope of the target fire source. This process is mainly based on the changing characteristics of the smoke monitoring data and the layout information of the sensors. The system will analyze the distribution of smoke concentrations, and areas with higher concentrations are usually closer to the fire source. If the smoke concentration detected by several sensors is significantly higher than that of other sensors, and these sensors are relatively concentrated in space, then the area where these sensors are located is very likely to be the source of the fire. At the same time, the system will also consider the direction and speed of smoke spread. By analyzing the time sequence in which different sensors detect smoke, combined with the ventilation system and spatial structure of the building, the location and scope of the fire source can be more accurately inferred. In a well-ventilated factory workshop, smoke may spread rapidly along the ventilation ducts. The system will comprehensively consider these factors to accurately determine the scope of the fire source and provide precise guidance for subsequent rescue operations.

[0034] In some embodiments, thermal imaging devices are installed in advance at various key locations of the target building. These devices use infrared thermal imaging technology to detect infrared rays emitted by objects and convert them into visible thermal imaging images. Once a fire occurs, the thermal imaging device will immediately start working, scan the fire area, and obtain a thermal imaging image of the fire area. Since the temperature of the fire source area is significantly higher than the surrounding environment when a fire occurs, the fire source area will show a different color and brightness from the surrounding area in the thermal imaging image, thereby clearly showing the heat distribution of the fire. After obtaining the thermal imaging image, the intelligent prediction system will conduct a comprehensive analysis of it with the real-time smoke monitoring data. The smoke monitoring data can roughly determine the area where the fire occurs, while the thermal imaging image can more accurately locate the fire source. In the thermal imaging image, the area with the highest temperature is the core position of the fire source. Along the direction of the fire source core to the direction where the temperature gradually decreases, the system can determine the boundary of the fire source. By analyzing the shape, size and position of the high-temperature area in the thermal imaging image, combined with information such as the smoke concentration distribution and diffusion direction in the smoke monitoring data, the system can accurately determine the specific location and boundary range of the fire source. After locating the boundaries and position of the fire source, the system will correct the target fire source range that was previously initially determined based on the smoke monitoring data. If the fire source range previously determined based solely on smoke monitoring data is not accurate enough, the combined analysis of thermal imaging images and smoke monitoring data can make the fire source range more accurate. The system will feed back the corrected target fire source range to the entire fire prediction system, providing more reliable data support for subsequent fire risk assessment, fire spread prediction, and rescue decisions.

[0035] S103, obtaining the fire risk value of the target fire source range according to a preset fire risk assessment model, wherein the fire risk assessment model is obtained in advance based on the structural layout of different areas in the target building, the flammability of building materials, the configuration of fire-fighting facilities and the corresponding fire risk value annotations combined with machine learning algorithm training, and the fire risk assessment model is used to assess the fire risk values ​​of different areas of the target building; The fire risk assessment model is a core component of the intelligent prediction system. When building the model, a large amount of data related to the target building needs to be collected. In terms of structural layout, it includes information such as the number of floors of the building, the area of ​​each floor, the functional division of each area, the location of passages and stairs, etc. Different structural layouts will affect the spread speed of the fire and the difficulty of evacuation. For example, in a building with a long and narrow layout and only one evacuation passage, once a fire occurs, the fire will easily spread along the passage, and the evacuation of personnel will also be greatly hindered, and the fire risk will be relatively high.

[0036] The flammability of building materials is also an important factor in the assessment. Different building materials, such as wood, plastic, and steel, have very different combustion characteristics. Wood and plastic are flammable materials that tend to burn quickly and release a lot of heat and toxic gases in a fire; while steel, although not easy to burn, loses strength at high temperatures, making the building structure unstable. The system keeps a detailed record of the building materials used in each area of ​​the building, and classifies and evaluates them according to their flammability. For areas that use a large amount of flammable decoration materials, such as conference rooms with wooden ceilings and synthetic carpets, the fire risk assessment value will be increased accordingly.

[0037] The firefighting facilities should not be ignored. Firefighting facilities include fire extinguishers, fire hydrants, automatic sprinkler systems, fire alarm systems, etc. The system will count the types, quantities and distribution locations of firefighting facilities in each area, and evaluate whether they meet the fire safety standards. If the number of fire hydrants on a certain floor is insufficient or some fire extinguishers are expired and not replaced, the fire risk value of that floor will increase. In addition, the maintenance status of firefighting facilities will also affect the evaluation results. Firefighting facilities that are regularly maintained and in good operation can effectively reduce fire risks.

[0038] After collecting this data, the model is trained using a machine learning algorithm in combination with the corresponding fire risk value annotations. Machine learning algorithms can automatically learn rules and patterns from large amounts of data, thereby improving the accuracy of the assessment. For example, by analyzing a large amount of historical fire case data, the model can learn the probability of fire and the degree of harm that may be caused under different structural layouts, building materials, and fire-fighting facilities. After repeated training and optimization, the fire risk assessment model can accurately assess the fire risk values ​​of different areas of the target building. After determining the range of the target fire source, the intelligent prediction system inputs the relevant data within the range into the fire risk assessment model to quickly obtain the fire risk value of the area, providing an important basis for subsequent judgments on whether there are high-risk fire sources and the formulation of corresponding fire response strategies.

[0039] In some embodiments, the fire risk value of the target fire source range can be adjusted dynamically in real time by collecting data from various IoT devices in the target building and integrating it with real-time smoke monitoring data, so as to more accurately assess the fire risk. The specific process is as follows: the intelligent prediction system will establish data connections with various IoT devices in the target building and collect the data generated by them in real time. Among them, the temperature sensor is one of the important data acquisition devices. It is distributed in various areas of the building and continuously monitors the ambient temperature. Once the temperature rises abnormally, it may indicate a fire hazard. These temperature data will be transmitted to the intelligent prediction system in a timely manner. Data collection of electrical equipment is also critical, including information such as the operating status of the equipment (whether it is overloaded, short-circuited, etc.), changes in power consumption, etc. The system integrates the collected temperature sensor data, electrical equipment data, ventilation equipment data and real-time smoke monitoring data. These data reflect the possibility of fire and the trend of fire development from different angles. Smoke monitoring data can directly indicate whether a fire has occurred and the initial scale of the fire; temperature data can assist in judging the development stage and severity of the fire. High temperature areas often mean that the fire is more intense; electrical equipment data helps to find potential triggers of the fire; ventilation equipment data can provide a basis for evaluating the speed of fire spread and the direction of smoke diffusion. The system will use data fusion technology to correlate and comprehensively analyze the data from different devices to form a data set that comprehensively reflects the fire situation. Based on the integrated data, the system will make real-time dynamic adjustments to the fire risk value corresponding to the target fire source range. The fire risk assessment model will recalculate the risk value based on the new data input. If the temperature sensor detects that the temperature in a certain area has risen sharply, the smoke concentration is also increasing, and there are signs of failure in the electrical equipment, and the ventilation equipment is not conducive to smoke exhaust, then the system will combine this information to increase the fire risk value of the area. On the contrary, if the ventilation equipment can effectively exhaust the smoke and the temperature rise trend is controlled, the fire risk value may be reduced accordingly. This real-time dynamic adjustment mechanism makes the fire risk assessment more in line with the actual situation and provides a more accurate basis for subsequent fire warnings and decisions, so that effective prevention and control measures can be taken in time to reduce the losses caused by fire.

[0040] S104: If the fire risk value is higher than the preset risk threshold, it is determined that there is a high-risk fire source point within the target fire source range; After obtaining the fire risk value of the target fire source range, the intelligent prediction system will immediately compare it with the preset risk threshold. The preset risk threshold is not set arbitrarily, but is determined after a large amount of fire research, historical data statistical analysis and the experience of fire safety experts. These data cover the risk situations of different types of buildings and different fire scenarios. Through the mining and analysis of massive data, a boundary value that can reasonably distinguish between ordinary fire risks and high-risk fire risks is found.

[0041] When the fire risk value is higher than the preset risk threshold, the system determines that there are high-risk fire sources within the target fire source range. This determination process is based on the scientific assessment results of the fire risk assessment model, which comprehensively considers many factors such as the structural layout of different areas in the target building, the flammability of building materials, and the equipment of fire-fighting facilities. In areas with complex structural layouts and narrow passages, once a fire occurs, the fire spreads quickly and it is difficult to evacuate people, and the fire risk value will increase accordingly; areas using flammable building materials, such as places with a large number of wooden structures or flammable decoration materials, are prone to fueling the fire and will also increase the fire risk value; and areas with insufficient or ineffective fire-fighting facilities cannot promptly and effectively extinguish and control the fire when a fire occurs, which will also cause the fire risk value to exceed the normal range.

[0042] After the system determines that there is a high-risk fire source, on the one hand, the system will focus on marking and recording the relevant information of the high-risk fire source, including the specific location of the fire source, the area involved, the initial assessment of the degree of danger, etc. This information will be integrated into the system's fire emergency database to provide key basis for subsequent rescue decisions and actions. On the other hand, the system will send high-risk fire source warning information to relevant personnel and departments to notify them to make preparations for response.

[0043] S105, obtaining current multi-dimensional environmental characteristic data in combination with meteorological data, where the multi-dimensional environmental characteristic data at least includes meteorological combination data; The intelligent prediction system obtains meteorological data mainly through docking with professional meteorological data service platforms. These platforms have a wide network of meteorological monitoring stations that can collect basic meteorological data such as atmospheric temperature, humidity, wind speed, wind direction, and air pressure in real time. The system obtains the latest meteorological information from the meteorological data service platform regularly at set time intervals through a standard data interface. After obtaining the meteorological data, the system will integrate it with the relevant information of the target building to generate multi-dimensional environmental characteristic data. Meteorological combination data is an important component of it, which comprehensively considers the relationship and synergy between multiple meteorological factors. For example, wind speed and wind direction directly affect the direction and speed of fire spread. In windy weather, the fire will spread rapidly along the wind direction and expand the scope of the fire; and a sudden change in wind direction may cause the fire to turn, bringing great difficulties to rescue work. Temperature and humidity will also affect the fire. A high temperature and dry environment will make flammable materials easier to burn and the fire will develop more rapidly; on the contrary, a low temperature and high humidity environment may suppress the fire to a certain extent, but it may also cause smoke accumulation and increase the risk of suffocation.

[0044] In addition to meteorological combination data, multi-dimensional environmental characteristic data may also include other relevant environmental factors such as geographic information around the building, environmental temperature gradient, air flow pattern, etc. These data can more comprehensively and accurately reflect the actual environmental conditions when the fire occurs through the organic combination with meteorological data. The system will use geographic information system (GIS) technology to spatially analyze and integrate information such as the geographical location of the building, the surrounding topography, and the distribution of buildings with meteorological data. By analyzing the impact of the terrain around the building, such as mountains and rivers, on wind direction and wind speed, the prediction model for fire spread is further optimized. After generating multi-dimensional environmental characteristic data, the system will store it in a special database and update it in real time. These data will serve as important input parameters for subsequent fire spread prediction models, providing strong support for more accurate predictions of fire spread paths, speeds, and smoke diffusion directions.

[0045] S106, inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation. The fire spread prediction model is trained in advance by a deep learning algorithm based on a large number of fire experiment data and actual fire case review data under different meteorological scenarios. The fire spread prediction model is used to predict the fire spread path, speed and smoke diffusion direction in real time; After the intelligent prediction system obtains the information of high-risk fire sources and meteorological data, it begins to prepare for input into the fire spread prediction model. The system will standardize these data and convert data of different formats and magnitudes into a unified and standardized form so that the model can process them efficiently. For the location coordinates of the fire source, the system will convert them according to the spatial coordinate system required by the model; for the temperature, humidity and other values ​​in the meteorological data, they will be normalized according to a specific ratio so that they are distributed within a reasonable range, ensuring the consistency of the weights of each data feature in model training and prediction.

[0046] After completing data preprocessing, the intelligent prediction system inputs the processed data into the preset fire spread prediction model. The model is built based on a deep learning algorithm and usually uses convolutional neural networks (CNN), recurrent neural networks (RNN) or their variants, such as long short-term memory networks (LSTM), gated recurrent units (GRU), etc. These network structures can effectively capture the spatiotemporal characteristics and complex relationships in the data.

[0047] During the model training phase, a large amount of fire experiment data under different meteorological scenarios and actual fire case review data were used to train the model. The fire experiment data was generated in a strictly controlled environment and contained data on the fire development process under various meteorological conditions, fire source types, building materials and other factors; the actual fire case review data was a detailed record and analysis of real fire events, which has a strong practical reference value. By learning from these massive data, the model can gradually grasp the laws and patterns of fire spread under different conditions. During the training process, the model continuously adjusts its own parameters, such as weights and biases in the neural network, to minimize the error between the predicted results and the actual situation. This process is achieved with the help of the back propagation algorithm, which guides the update direction and amplitude of the parameters by calculating the gradient of the prediction error to the model parameters. After multiple iterative training, the prediction performance of the model is continuously improved.

[0048] After inputting high-risk fire source and meteorological combination data, the model begins to predict the fire spread path, speed and smoke diffusion direction in real time based on the knowledge learned during training. The model will consider the initial energy of the fire source, combustion characteristics and the impact of meteorological conditions on the fire. When predicting the fire spread path, the model will judge the possible direction of fire expansion based on wind speed and wind direction information, combined with the structural layout of the building, taking into account the blocking or guiding effect of obstacles on the fire; when predicting the fire spread speed, it will comprehensively consider factors such as temperature, humidity, and flammability of building materials. High temperature and dry environment and flammable materials will accelerate the spread of fire, while low temperature and high humidity environment may slow down the fire; for the direction of smoke diffusion, the model will combine wind direction, building ventilation system and spatial structure to make judgments to ensure the accuracy of the prediction.

[0049] After the prediction is completed, the system will post-process the results output by the model. The predicted fire spread path, speed, smoke diffusion direction and other information will be visualized and converted to intuitive and easy-to-understand graphics or charts for subsequent analysis and display. At the same time, the system will also perform uncertainty analysis on the prediction results, evaluate the reliability of the prediction, and provide confidence intervals for the prediction results through multiple simulation predictions or ensemble learning methods, providing more comprehensive information support for subsequent decision-making.

[0050] In some embodiments, the current fire scene can be associated with historical fire cases for analysis, so as to optimize and adjust the fire spread prediction, make the prediction results more consistent with the actual situation, and provide a more reliable reference for fire prevention and control decisions. Specifically, the system will extract the key features of the current fire scene, covering various attributes and meteorological data of the target fire source range. The characteristics of the target fire source range include its location, area size, internal structure layout, types of building materials involved and their flammability. Meteorological data includes wind speed, wind direction, temperature, humidity, air pressure, etc. The system combines these features into a unique fire scene identifier. Subsequently, the system searches in a huge database of historical fire cases to find historical cases that match the current fire scene identifier. Each historical fire case in the database pre-stores detailed scene information. The system will use a specific algorithm to compare the characteristics of the current scene with historical cases one by one to find cases with high similarity. Once a matching historical fire case is found, the system will deeply compare the development process and results of the current fire with these historical cases. For example, analyze how the fire spreads under similar target fire source ranges and meteorological conditions in historical cases, the changes in the spread speed and path, and the final impact and loss. If in historical cases, under similar high temperature and dry weather conditions, and when the fire source is in a well-ventilated but complexly structured area, the fire spreads quickly and easily forms three-dimensional combustion, then the system will consider increasing the estimate of the fire spread speed in the current fire spread prediction, and pay attention to the possibility of three-dimensional combustion. Based on the comparison results, the system will adjust the fire spread situation based on the preliminary prediction of the model. If it is found that the current fire scene is highly similar to a historical case in key features, and the fire in the historical case spreads rapidly in a short period of time, the system will adjust the current predicted fire spread speed and range to make the prediction result closer to the actual possible development trend.

[0051] In some embodiments, the intelligent prediction system uses positioning technology and geographic information data to determine the location of the target building. This may involve technologies such as the global positioning system (GPS) and base station positioning, combined with detailed map data in the geographic information system (GIS) to obtain the precise geographic coordinates of the target building. For the target area, the system will also clarify its specific geographic location information. Through these coordinate information, the system uses a suitable distance calculation algorithm, such as the Euclidean distance algorithm, the network distance algorithm that takes into account the actual road conditions, etc., to calculate the distance between the target building and the target area. After obtaining the distance information, the system calculates the spread time in combination with the previously predicted fire spread. The fire spread prediction includes key data such as the fire spread path and speed. Assuming that the system predicts that the fire will spread along a specific path at a certain speed, by dividing the distance between the target building and the target area by the predicted fire spread speed, the time required for the fire to spread to the target area can be roughly determined. If the predicted fire spread speed is 50 meters per hour and the distance between the target building and the target area is 200 meters, then the preliminary calculation shows that the time it takes for the fire to spread to the target area is about 4 hours. The system will also take into account various influencing factors during the spread of fire, such as changes in speed when encountering different building materials and obstacles, the dynamic impact of meteorological conditions on the fire, etc., and correct the calculated time to obtain a more accurate spread time.

[0052] When the user side, such as building management personnel, fire command center and other relevant personnel, wants to know the time when the fire spreads to the target area, they will send a spread time viewing instruction to the intelligent prediction system. After receiving the instruction, the system will quickly obtain the previously determined spread time data of the fire spreading to the target area from the storage or calculation module. The system formats the time data and converts it into an easy-to-understand format, such as "The fire is expected to spread to the target area in X hours and X minutes", and then sends the information to the user side through the network communication module. After the user side receives the information, the fire spread time will be displayed on the corresponding interface, so that relevant personnel can intuitively obtain this key information and make preparations in advance, such as organizing personnel evacuation and deploying rescue resources.

[0053] S107: Send the fire spread prediction to the management terminal for alarm.

[0054] After obtaining the fire spread forecast, the intelligent prediction system will integrate and encrypt the information. The fire spread path, speed, smoke diffusion direction and related risk assessment information are summarized into a complete data packet to ensure that the management end receives comprehensive and organized fire information. To ensure the security of data transmission, the system uses advanced encryption algorithms, such as AES (Advanced Encryption Standard), to encrypt the data packet to prevent the information from being stolen or tampered with during transmission. The system sends the encrypted fire spread forecast to the management end through the network communication module.

[0055] After receiving the data, the management end first performs a decryption operation to restore the original fire spread prediction information. The alarm system of the management end will issue various forms of alarm prompts based on the received information. The sound and light alarm emits a strong sound and light signal to attract the attention of the management personnel; at the same time, a striking alarm window pops up on the monitoring interface of the management end, displaying the location of the fire, spread trend and other key information in an intuitive way.

[0056] In some embodiments, the system can convert the collected target fire source range and fire spread prediction into clear and easy-to-understand voice commands. During the conversion process, natural language processing technology is used to process the information. The complex fire location information is converted into popular language descriptions, such as "the fire occurred in the east office area on the third floor of the building, near the stairwell"; for the fire spread prediction, it will be expressed as "the fire is expected to spread to the fourth floor within 10 minutes, and mainly spread to the west corridor, accompanied by a lot of thick smoke". By organizing the language reasonably, the voice commands are ensured to be concise and clear, so that people in the building can quickly understand the severity and potential dangers of the fire. After completing the voice command conversion, the system uses the pre-installed broadcast settings in the target building to broadcast. These broadcasting devices are distributed in various public areas of the building, such as corridors, halls, stairwells, etc., to ensure that the sound can cover every corner of the building. This ensures that people can get the latest fire dynamics in a timely manner.

[0057] In the embodiments of the present application, due to the use of real-time monitoring data of smoke alarm sensors, a fire risk assessment model based on multi-factor training, a multi-dimensional environmental feature analysis combined with meteorological data, a fire spread prediction model trained by deep learning, and a complete alarm and information transmission mechanism, it is possible to accurately determine the occurrence of fire and the scope of the fire source in the early stage of the fire, accurately assess the fire risk and discover high-risk fire sources, accurately predict the spread of fire by combining multiple factors, and promptly transmit information to the management end and relevant personnel, effectively solving the technical problems in the prior art that a single reliance on smoke alarms cannot accurately assess fire risks, is difficult to scientifically predict the development of fires, and has untimely and incomplete fire information transmission, thereby realizing the intelligence of fire warning and the precision of prediction, greatly improving the scientific decision-making ability of fire prevention and control, and significantly reducing the risk of casualties and property losses caused by fires.

[0058] After combining the above content, the following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the fire intelligent prediction method in the embodiment of the present application.

[0059] S201, using a geographic information system to visualize the target building and the environment within a set surrounding area, wherein the visualization includes marking the location of high-risk fire sources and the fire spread prediction on an electronic map; After completing the fire spread prediction, the intelligent prediction system calls the Geographic Information System (GIS) function. The system first obtains the geographic data of the target building and its surrounding area from the internal database or relevant data interface. These data include topographic information, road distribution, building location and outline, etc. The system integrates these geographic data with fire-related information.

[0060] For the location of high-risk fire source points, the system obtains the previously determined fire source point coordinate information and adapts it to the coordinate system of the electronic map through the coordinate conversion algorithm. Then, the high-risk fire source points are marked on the electronic map layer with eye-catching marks (such as flashing red icons) to ensure that they are easy to identify on the map. For the fire spread prediction, the system converts the predicted fire spread path into line graphics on the electronic map. According to the different stages and directions of the fire spread, lines of different colors and widths are used to represent it. For example, a thick red line represents the main path of the fire spread, and a thin yellow line represents a possible secondary spread path. For the speed of fire spread, the system displays it by adding dynamic elements or marking values ​​on the path lines, so that the viewer can intuitively understand the speed of the fire. For the direction of smoke diffusion, the system superimposes arrow graphics on the map. The direction of the arrow is the direction of smoke diffusion, and the density of the arrow can be used to represent the relative size of the smoke concentration.

[0061] When performing visualization, the system will select a suitable electronic map base map based on actual needs and data characteristics. For urban construction areas, select a city map base map that contains detailed street and building information; for remote areas or industrial plants, select a map base map that can clearly display the topography and main facilities. The system will also add interactive functions to the visualization interface to facilitate user operation. Users can view fire conditions in different areas by zooming and panning the map, and can also click on marks and graphics on the map to obtain more detailed information, such as the specific location description of high-risk fire sources, and the estimated time for the fire to spread in a certain area. In addition, the system will update the visualization content in real time. As the fire develops and new forecast data is generated, the marks, lines and arrows on the map will be adjusted in a timely manner to ensure that users always see the latest fire situation.

[0062] S202, automatically selecting the nearest fire rescue site with corresponding rescue capabilities according to the preset emergency rescue plan; After displaying the fire-related information, the intelligent prediction system begins to perform the task of screening fire rescue sites. The system first reads the preset emergency rescue plan, which contains detailed information on each fire rescue site, such as the site location coordinates, the type and quantity of rescue equipment equipped, the type and scale of fire that can be dealt with, etc. At the same time, the system also obtains the accurate location information of the target building.

[0063] The system uses a distance calculation algorithm to calculate the distance between each fire rescue station and the target building based on the location of the target building. Common distance calculation methods include the Euclidean distance algorithm and the Manhattan distance algorithm. The system will select the appropriate algorithm based on the actual situation. During the calculation process, the system fully considers factors such as road conditions and traffic congestion to correct the distance. If a road is often congested during a certain period of time, the system will appropriately increase the distance weight to the target building through this road to more accurately reflect the actual rescue time required.

[0064] When selecting fire rescue stations with corresponding rescue capabilities, the system matches the rescue capabilities of each fire rescue station based on the fire spread prediction and the characteristics of the target fire source range. If the fire involves hazardous chemicals, the system will select fire rescue stations equipped with professional hazardous chemical handling equipment and relevant trained personnel; if the fire is large, the system will give priority to stations equipped with large fire-fighting equipment (such as fire trucks, fire helicopters) and sufficient personnel. The system will also consider the real-time status of the fire rescue station, such as whether other rescue missions are being performed, whether the equipment is in an available state, and other information. If most of the equipment at a station is under maintenance or the personnel have been transferred to other rescue sites, the priority of the station in screening will be reduced.

[0065] The system sorts the fire rescue sites in order of distance from near to far and rescue capability matching from high to low. In the sorting process, the weights of multiple factors will be comprehensively considered, such as the distance factor may account for 40% of the weight, the rescue capability matching degree accounts for 50% of the weight, and the real-time status accounts for 10% of the weight. Through this weighted calculation method, it is ensured that the selected fire rescue sites can not only quickly reach the fire scene, but also have the ability to deal with the current fire. Finally, the system determines the fire rescue site with the closest distance and corresponding rescue capability as the priority recommended site to provide strong support for subsequent rescue operations.

[0066] S203: Send the high-risk fire source location, the fire spread prediction and the multi-dimensional environmental feature data to the fire rescue site.

[0067] After the intelligent prediction system determines the fire rescue site, it integrates and processes the location of high-risk fire sources, fire spread predictions, and multi-dimensional environmental feature data. For the location of high-risk fire sources, the system converts the coordinate information previously marked on the electronic map into a detailed geographic location description, including specific street addresses, building names, floors, and other information, so that fire rescue personnel can find the fire source more accurately. For the fire spread prediction, the system organizes the predicted path, speed, smoke diffusion direction and other information into a structured data format, such as JSON format, to make the information easier to transmit and parse. For multi-dimensional environmental feature data, the system summarizes meteorological data (such as wind speed, wind direction, temperature, humidity), geographic information around the building (such as topography, water source location), etc. to ensure the integrity of the data.

[0068] After receiving the data, the receiving system of the fire rescue site first performs a decryption operation to restore the original data content. Then, the data is imported into the fire rescue command system. In the command system, the data will be displayed to rescuers in an intuitive way, such as re-marking the location of high-risk fire sources on electronic maps, displaying the fire spread prediction in the form of charts, and presenting multi-dimensional environmental feature data on the information panel. Rescuers can quickly understand the actual situation at the fire scene based on these data and formulate scientific and reasonable rescue plans.

[0069] In the embodiment of the present application, due to the use of technical means to visualize fire information in a geographic information system, screen fire rescue sites based on preset emergency rescue plans, and accurately transmit key data to the sites, it is possible to intuitively present the fire situation, quickly determine suitable rescue sites, and provide comprehensive and accurate fire-related information, effectively solving the problems of unintuitive presentation of fire information, unreasonable allocation of rescue sites, and untimely and incomplete acquisition of rescue information in the prior art, thereby achieving efficient and precise fire rescue, greatly improving the response speed and success rate of fire rescue, and effectively protecting the life safety and property safety of people in the building.

[0070] The following describes the intelligent prediction system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of a physical device structure of the intelligent prediction system in an embodiment of the present application.

[0071] It should be noted that Figure 3 The structure of the intelligent prediction system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0072] like Figure 3 As shown, the intelligent prediction system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0073] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0074] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0075] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0077] Specifically, the intelligent prediction system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the fire intelligent prediction method provided in the above embodiment is implemented.

[0078] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the intelligent prediction system described in the above embodiment; or may exist independently without being assembled into the intelligent prediction system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the intelligent prediction system, the intelligent prediction system implements the fire intelligent prediction method provided in the above embodiment.

[0079] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0080] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0081] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A fire intelligent prediction method, applied to an intelligent prediction system, characterized in that: The method comprises: Obtain real-time smoke monitoring data from pre-installed smoke alarm sensors in the target building; When the real-time smoke monitoring data reaches a preset smoke data threshold, determining that a fire has occurred in the target building, and determining a target fire source range where the fire has occurred based on the real-time smoke monitoring data; Obtaining the fire risk value of the target fire source range according to a preset fire risk assessment model, wherein the fire risk assessment model is obtained in advance based on the structural layout of different areas in the target building, the flammability of building materials, the configuration of fire-fighting facilities and the corresponding fire risk value annotations combined with machine learning algorithm training, and the fire risk assessment model is used to assess the fire risk values ​​of different areas of the target building; If the fire risk value is higher than the preset risk threshold, it is determined that there is a high-risk fire source point within the target fire source range; Acquire current multi-dimensional environmental characteristic data in combination with meteorological data, wherein the multi-dimensional environmental characteristic data at least includes meteorological combination data; The high-risk fire source point and the meteorological combination data are input into a preset fire spread prediction model to obtain a fire spread prediction situation. The fire spread prediction model is trained in advance by a deep learning algorithm based on a large number of fire experiment data and actual fire case review data under different meteorological scenarios. The fire spread prediction model is used to predict the fire spread path, speed and smoke diffusion direction in real time; The fire spread prediction situation is sent to the management end for alarm.

2. The method according to claim 1, characterized in that After the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, the method further includes: Using a geographic information system to visualize the target building and the environment within a set surrounding area, the visualization includes marking the location of high-risk fire sources and the fire spread prediction on an electronic map; Automatically select the nearest fire rescue site with corresponding rescue capabilities according to the preset emergency rescue plan; The high-risk fire source location, the fire spread prediction and the multi-dimensional environmental characteristic data are sent to the fire rescue site.

3. The method according to claim 1, characterized in that When the real-time smoke monitoring data reaches a preset smoke data threshold, after determining that a fire has occurred in the target building and determining the target fire source range of the fire according to the real-time smoke monitoring data, the method further includes: Acquire thermal imaging images of the fire area by installing thermal imaging equipment in the target building; Locating the boundary and position of the fire source through the thermal imaging image and the real-time smoke monitoring data; Modify the target fire source range based on the boundaries and location of the fire source.

4. The method according to claim 1, characterized in that: After the step of obtaining the fire risk value of the target fire source range according to the preset fire risk assessment model, the method further includes: Collecting device data of IoT devices in the target building, wherein the IoT devices include at least temperature sensor data, electrical equipment, and ventilation equipment; After the equipment data and the real-time smoke monitoring data are integrated, the fire risk value corresponding to the target fire source range is dynamically adjusted in real time.

5. The method according to claim 1, characterized in that After the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, the method further includes: Correlate and analyze current fire scenarios with historical fire cases; By comparing the development process and results of historical fire cases under scenarios matching the fire scenario, the fire spread prediction is adjusted accordingly. The fire scenario includes various characteristics and meteorological data of the target fire source range.

6. The method according to claim 1, characterized in that After the step of inputting the high-risk fire source point and the meteorological combination data into a preset fire spread prediction model to obtain a fire spread prediction situation, the method further includes: Determine the distance between the location of the target building and the target area; Determine the time it takes for the fire to spread to the target area based on the fire spread prediction and the distance; After receiving the spreading time viewing instruction sent by the user terminal, the spreading time is sent to the user terminal for display.

7. The method according to claim 1, characterized in that After the step of sending the fire spread prediction situation to the management end for alarm, the method further includes: The fire-related information is converted into voice commands and broadcasted through the broadcasting settings in the target building, wherein the fire-related information includes the target fire source range and the fire spread prediction.

8. An intelligent prediction system, characterized in that: The intelligent prediction system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the intelligent prediction system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the intelligent prediction system, the intelligent prediction system is caused to execute the method as described in any one of claims 1-7.

10. A computer program product, characterized in that When the computer program product runs on an intelligent prediction system, the intelligent prediction system is enabled to execute the method according to any one of claims 1 to 7.

Images