Big-data-driven urban fire prevention and control intelligent decision-making method and system

Abstract

The invention relates to the field of computer science and technology and data science and big data technology, and discloses a big data-driven urban fire prevention and control intelligent decision-making method and system, the system comprises a data acquisition processing end, a decision-making analysis end and a decision-making execution end, and multi-source data fusion is realized through frequency consistency judgment and an interpolation algorithm; constructing an urban fire static and dynamic combined data matrix, and dynamically matching an optimal prevention and control decision algorithm type based on an eagle optimization algorithm; prevention and control scheme data including evacuation path planning, a resource allocation scheme and a fire behavior control strategy are generated, and a closed-loop feedback mechanism real-time correction scheme is established; and the fire-fighting management platform and the Internet of Things sensor network are connected through a data bus. The three technical bottlenecks of multi-source data fusion distortion, decision algorithm solidification and response closed loop deficiency are broken through, the fire behavior prediction error is reduced, the scheme generation efficiency is improved, and the major fire casualty rate is reduced.

Description

Technical Field

[0001] This invention relates to the fields of computer science and technology and data science and big data technology, specifically to a big data-driven intelligent decision-making method and system for urban fire prevention and control. Background Technology

[0002] Computer Science and Technology is a long-established and widely recognized interdisciplinary field, encompassing computer hardware, software, networks, and theory. It primarily studies the design and manufacture of computers, including the fundamental theories, skills, and methods of computer software and hardware, as well as the development and maintenance of computer systems and software. Big Data Science and Technology, on the other hand, is a new engineering discipline that has emerged in recent years due to the explosive growth of data volume and the increasing demand for data analysis. It focuses on big data as its research object and aims to cultivate high-level applied technical personnel who master data science theories and methods and possess skills in big data acquisition, cleaning, storage, processing and analysis, visualization, and application. This major integrates knowledge from multiple disciplines such as statistics and computer science, making it a typical interdisciplinary field.

[0003] In recent years, urban fire prevention and control systems have gradually introduced big data technology, but there are still shortcomings in areas such as multi-source heterogeneous data fusion, dynamic decision optimization, and closed-loop response. Multi-source data fusion distortion problem: Existing systems use a single frequency to collect static building data and dynamic sensor data. Due to the difference in sampling frequency, the generated heat map is offset. Typical cases show that when dynamic data from temperature sensors is directly fused with historical static fire data, the fire prediction error rate increases.

[0004] Defects of rigid decision-making algorithms: Mainstream platforms rely on preset decision trees to generate prevention and control plans, which cannot adapt to complex fire scene environments. Statistics show that in high-rise building fires, the failure to dynamically match the optimal algorithm leads to delays in resource allocation.

[0005] Lack of closed-loop feedback mechanism: Although the current international standard ISO 24681-2020 specifies the implementation process of the plan, it lacks: a real-time data feedback channel, a prediction-actual response deviation analysis mechanism, and an emergency enhancement module for sudden fire events.

[0006] This background technology reveals technical deficiencies through quantitative data and benchmarks against three core innovations: frequency adaptation, Osprey optimization, and three-dimensional fire scene model, laying a necessary foundation for the technical solution in the claims. Summary of the Invention

[0007] (a) Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a big data-driven intelligent decision-making method and system for urban fire prevention and control, which solves the problems mentioned in the background section.

[0009] (II) Technical Solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: a big data-driven intelligent decision-making method and system for urban fire prevention and control, wherein the method includes the following steps: S1. Collect static basic data and dynamic monitoring data of urban fires; S2. Based on the static basic data and dynamic monitoring data, perform data sampling frequency measurement and processing to generate static basic data frequency data and dynamic monitoring data frequency data; S3. Based on the static basic data frequency data and the dynamic monitoring data frequency data, perform sampling frequency consistency judgment processing to generate frequency consistency judgment data; when the frequencies are the same, directly execute step S5. S4. When the frequencies are different, the sampling frequency of the static basic data is adjusted based on the static basic data, the frequency data of the static basic data and the frequency data of the dynamic monitoring data to generate static basic data adjustment data. S5. Based on the static basic data, the static basic data adjustment data, and the dynamic monitoring data, perform combined processing of urban fire static and dynamic data to construct an urban fire static and dynamic combined data matrix. S6. Based on the urban fire static and dynamic combination data matrix and the standard static and dynamic combination data corresponding to different fire prevention and control decision algorithms, perform fire prevention and control decision algorithm type matching processing to generate target fire prevention and control decision algorithm type feature data. S7. Combine the static and dynamic combination data matrix of urban fire with the characteristic data of the target fire prevention and control decision algorithm type to form fire prevention and control decision summary data, and call the corresponding decision algorithm to process and generate urban fire prevention and control plan data. Specifically, S6 employs the Osprey optimization algorithm for fire prevention and control decision-making algorithm type matching, including: initializing the Osprey population position, updating the position during the exploration phase, optimizing the position during the development phase, and iteratively outputting the optimal matching algorithm type.

[0011] Preferably, according to claim 3, the big data-driven intelligent decision-making method and system for urban fire prevention and control, wherein S3 specifically includes: S31. Collect static basic data through the urban fire management platform, including building structure data, fire protection facility distribution data, historical fire data, and population density data; S32. Real-time dynamic monitoring data is collected through an Internet of Things (IoT) sensor network deployed in urban building clusters, including temperature data, smoke concentration data, gas concentration data, and video surveillance data.

[0012] Preferably, in the big data-driven intelligent decision-making method and system for urban fire prevention and control according to claim 4, step S4 includes: S41. Import static basic data and dynamic monitoring data into the urban fire prevention and control decision-making platform; S42. Use a bidirectional search algorithm to search for data sampling frequency information in the decision-making platform based on frequency keywords, and generate static basic data frequency data. and dynamic monitoring data frequency data All units are Hertz.

[0013] Preferably, according to claim 5, the big data-driven intelligent decision-making method and system for urban fire prevention and control, step S5 includes: S51, Obtain and ; S52, to and Perform numerical comparison: like The output frequency is the same as the judgment flag; like The output frequency is different, which is a judgment indicator; in This is a preset threshold.

[0014] Preferably, in the big data-driven intelligent decision-making method and system for urban fire prevention and control according to claim 6, step S6 includes: S61. When frequencies are inconsistent, first check the static basic data. Normalize it to map it to the range [0,1]: in, This refers to the normalized static basic data; Using linear interpolation algorithm sampling frequency Adjust to Generate static basic data and adjust data ,satisfy: in, This represents a linear interpolation function. Normalization ensures that data of different dimensions remain consistent during the interpolation process, eliminating the risk of fusion distortion.

[0015] Preferably, according to claim 7, the big data-driven intelligent decision-making method and system for urban fire prevention and control, step S7 includes: S71. Unify the time scale, firstly, for the frequency-adjusted static basic data... and dynamic monitoring data Normalization is performed to map each feature value to the range [0,1]. in, and These are the normalized data; Subsequently, a combined static and dynamic data matrix of urban fires was constructed. : in, This is a static and dynamic combined data matrix of urban fires, with dimensions of [missing information]. , It is the feature number of static data. It is the feature number of dynamic data. It refers to the number of time points. This is the static baseline data after frequency adjustment. For dynamic monitoring data, normalization processing is emphasized to be completed before matrix construction to eliminate the influence of dimensions and improve the reliability of feature fusion.

[0016] Preferably, according to claim 8, the big data-driven intelligent decision-making method and system for urban fire prevention and control, step S8 includes: S81. Establish a standard library matrix for fire prevention and control decision-making algorithms: in, This is the matrix of the standard library for fire prevention and control decision-making algorithms, with dimension 1. h is the number of algorithm types, and f is the standard feature dimension for each algorithm type. Indicates the first Standard static and dynamic combined data characteristics corresponding to each algorithm type; S82, using the Osprey optimization algorithm to... and To perform matching, including: S821. Initialization: Randomly generate the locations of the osprey population and normalize them to the range [0,1]. in, , For the search space boundary, As the lower limit, The upper limit, ∈[0,1] is a random number. The position is the normalized position; normalization eliminates the influence of boundary dimensions. S822, Exploration Phase: Update positions according to standard exploration logic, simulating random search behavior: Update location, where, , For the search space boundary, ∈[0,1], The population location is randomly selected and normalized to [0,1]. S823, Development Phase: Optimize the position according to standard development logic, simulate local fine-grained search, and base the current optimal position. : Optimize the location, among which, To find the current optimal position, normalize to [0,1]. , For boundary parameters, ∈[0,1], As the attenuation factor, This represents the number of iterations. S824, Iterative Output and Highest matching Corresponding algorithm type identifier ,in The target fire prevention and control decision-making algorithm type feature data.

[0017] Preferably, in the big data-driven intelligent decision-making method and system for urban fire prevention and control according to claim 9, step S9 includes: S91. Constructing a summary data set for fire prevention and control decisions: in, Data is collected for fire prevention and control decision-making. Indicates will and This data structure will be combined into a tuple. and These are linked together and used as input for decision-making algorithms. S92, Call Corresponding decision algorithm processing The fire characteristic parameters are used to generate prevention and control plan data, including evacuation route planning, rescue resource allocation plans, and fire control strategies. in, For data related to the prevention and control plan, This is the function to execute the decision algorithm.

[0018] Preferably, the big data-driven intelligent decision-making method for urban fire prevention and control according to any one of claims 1-8 further includes, after generating the prevention and control plan data: S11. Establish a scheme execution feedback channel to obtain the actual response data generated when the fire terminal executes the prevention and control scheme data in real time; S12. Perform deviation analysis between the actual response data and the predicted response data. If the deviation exceeds the safety threshold... This triggers the dynamic scheme correction mechanism: S121: Use the Osprey optimization algorithm to re-match the decision algorithm type; S122: Update rescue resource allocation parameters based on the revised algorithm type; S123: Generate dynamic correction scheme data The original solution An optimized version based on the existing version was then pushed to the fire alarm terminal.

[0019] Preferably, the dynamic plan correction mechanism based on the generated prevention and control plan data further includes: S13. When actual response data indicates that the fire spread rate exceeds the predicted value γ%: S131: Activate multi-source data enhancement acquisition and increase the temperature sensor deployment density to α times the original deployment; S132: Reconstructing a static-dynamic combined data matrix based on enhanced data: in, This is the static baseline data after frequency adjustment. For enhanced dynamic monitoring data; S133: Processed using a spatiotemporal convolutional neural network Output a heat map (Ht) predicting the spread of the fire. S134: Overlay Ht onto the original fire prevention and control plan to form a three-dimensional fire control decision model: Three-dimensional fire control decision model

[0020] Ht represents the heat map of fire spread. The original prevention and control plan, This is the space addition operator. This is a function for generating three-dimensional decision models.

[0021] (III) Beneficial Effects

[0022] Compared with existing technologies, this invention provides a big data-driven intelligent decision-making method and system for urban fire prevention and control, which has the following beneficial effects: 1. In this invention, by setting up a data acquisition and processing terminal, when making urban fire prevention and control decisions, static basic data and dynamic monitoring data are collected in real time, and the sampling rate of static data is dynamically adjusted based on frequency consistency judgment to ensure the accuracy of multi-source heterogeneous data fusion. At the same time, a system monitoring mechanism is established during the data acquisition process. When the data frequency deviation exceeds the threshold, frequency correction is triggered in real time, which can eliminate the data distortion problem caused by sensor sampling differences, ensure the reliability of fire feature parameter extraction, and improve the accuracy of prevention and control decisions from the data source.

[0023] 2. In this invention, by setting up a decision analysis terminal, when generating fire prevention and control strategies, intelligent matching of decision algorithms is achieved based on the static and dynamic combined data matrix and the Osprey optimization algorithm. The optimal decision model is dynamically selected through a collaborative optimization mechanism in the exploration and development phases. This enables the system to adaptively select algorithm types such as principal point iterative correction and improved DTW according to the actual fire scene characteristics, and to correct algorithm selection deviations in real time in complex and ever-changing fire scenarios. This ensures that the transformation process from data to decision is accurate and efficient, and improves the response speed of scheme generation.

[0024] 3. In this invention, by setting up a decision execution terminal, during the scheme implementation phase, the algorithm matching results are fused with real-time fire data in three dimensions to generate a prevention and control scheme that includes evacuation route planning, dynamic resource allocation, and fire level control strategies. By establishing a scheme execution feedback channel, a dynamic correction mechanism is triggered in real time when the actual response deviates from the predicted value, enabling the system to automatically increase the data acquisition density and reconstruct the decision model in response to sudden changes in fire conditions. This ensures the reliability of closed-loop control from scheme generation to terminal execution, minimizing the risk of casualties and property losses. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the steps of the big data-driven intelligent decision-making method and system for urban fire prevention and control of the present invention. Figure 2 This is a schematic diagram of the overall architecture of the big data-driven intelligent decision-making method and system for urban fire prevention and control of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1-2The big data-driven intelligent decision-making method and system for urban fire prevention and control includes the following steps: S1. Collect static basic data and dynamic monitoring data of urban fires; S2. Based on static basic data and dynamic monitoring data, perform data sampling frequency measurement and processing to generate static basic data frequency data and dynamic monitoring data frequency data. S3. Based on the static basic data frequency data and the dynamic monitoring data frequency data, perform sampling frequency consistency judgment processing to generate frequency consistency judgment data; when the frequencies are the same, directly execute step S5. S4. When the frequencies are different, the sampling frequency of the static basic data is adjusted based on the static basic data, the frequency data of the static basic data, and the frequency data of the dynamic monitoring data to generate the static basic data adjustment data. S5. Based on the static basic data, the adjusted data, and the dynamic monitoring data, the static and dynamic data of urban fires are combined and processed to construct a static and dynamic combined data matrix of urban fires. S6. Based on the urban fire static and dynamic combination data matrix and the standard static and dynamic combination data corresponding to different fire prevention and control decision algorithms, perform fire prevention and control decision algorithm type matching processing to generate target fire prevention and control decision algorithm type feature data. S7. Combine the static and dynamic combination data matrix of urban fires with the characteristic data of the target fire prevention and control decision algorithm type to form the summary data of fire prevention and control decision, and call the corresponding decision algorithm to process and generate urban fire prevention and control plan data. Among them, S6 uses the Osprey optimization algorithm for fire prevention and control decision-making algorithm type matching, including: initializing the Osprey population position, updating the position in the exploration phase, optimizing the position in the development phase, and iteratively outputting the optimal matching algorithm type; S31. Collect static basic data through the urban fire management platform, including building structure data, fire protection facility distribution data, historical fire data, and population density data; S32. Real-time collection of dynamic monitoring data, including temperature data, smoke concentration data, gas concentration data and video surveillance data, through an Internet of Things sensor network deployed in urban building clusters; S41. Import static basic data and dynamic monitoring data into the urban fire prevention and control decision-making platform; S42. Use a bidirectional search algorithm to search for data sampling frequency information in the decision-making platform based on frequency keywords, and generate static basic data frequency data. and dynamic monitoring data frequency data All units are Hertz; S51, Obtain and ; S52, to and Perform numerical comparison: like The output frequency is the same as the judgment flag; like The output frequency is different, which is a judgment indicator; in The preset threshold; S61. When frequencies are inconsistent, first check the static basic data. Normalize it to map it to the range [0,1]: in, This refers to the normalized static basic data; Using linear interpolation algorithm sampling frequency Adjust to Generate static basic data and adjust data ,satisfy: in, This represents a linear interpolation function. Normalization ensures that data of different dimensions remain consistent during the interpolation process, eliminating the risk of fusion distortion. S71. Unify the time scale, firstly, for the frequency-adjusted static basic data... and dynamic monitoring data Normalization is performed to map each feature value to the range [0,1]. in, and These are the normalized data; Subsequently, a combined static and dynamic data matrix of urban fires was constructed. : in, This is a static and dynamic combined data matrix of urban fires, with dimensions of [missing information]. , It is the feature number of static data. It is the feature number of dynamic data. It refers to the number of time points. This is the static baseline data after frequency adjustment. For dynamic monitoring data, normalization processing is emphasized to be completed before matrix construction to eliminate the influence of dimensions and improve the reliability of feature fusion. S81. Establish a standard library matrix for fire prevention and control decision-making algorithms: in, This is the matrix of the standard library for fire prevention and control decision-making algorithms, with dimension 1. h is the number of algorithm types, and f is the standard feature dimension for each algorithm type. Indicates the first Standard static and dynamic combined data characteristics corresponding to each algorithm type; S82, using the Osprey optimization algorithm to... and To perform matching, including: S821. Initialization: Randomly generate the locations of the osprey population. in, , For the search space boundary, ∈[0,1]; S822, Exploration Phase: Note: Update location, where, , For the search space boundary, ∈[0,1], This is the current optimal position; S823, Development Phase: Note: Optimize the location, among which, , For boundary parameters, ∈[0,1], This represents the number of iterations. S821. Initialization: Randomly generate the locations of the osprey population and normalize them to the range [0,1]. in, , For the search space boundary, As the lower limit, The upper limit, ∈[0,1] is a random number. The position is the normalized position; normalization eliminates the influence of boundary dimensions. S822, Exploration Phase: Update positions according to standard exploration logic, simulating random search behavior: Update location, where, , For the search space boundary, ∈[0,1], The population location is randomly selected and normalized to [0,1]. S823, Development Phase: Optimize the position according to standard development logic, simulate local fine-grained search, and base the current optimal position. : Optimize the location, among which, To find the current optimal position, normalize to [0,1]. , For boundary parameters, ∈[0,1], As the attenuation factor, This represents the number of iterations. S91. Constructing a summary data set for fire prevention and control decisions: in, Data is collected for fire prevention and control decision-making. Indicates will and This data structure will be combined into a tuple. and These are linked together and used as input for decision-making algorithms. S92, Call Corresponding decision algorithm processing The fire characteristic parameters are used to generate prevention and control plan data, including evacuation route planning, rescue resource allocation plans, and fire control strategies. in, For data related to the prevention and control plan, This is the function to execute the decision algorithm. S11. Establish a scheme execution feedback channel to obtain the actual response data generated when the fire terminal executes the prevention and control scheme data in real time; S12. Perform deviation analysis between the actual response data and the predicted response data. If the deviation exceeds the safety threshold... This triggers the dynamic scheme correction mechanism: S121: Use the Osprey optimization algorithm to re-match the decision algorithm type; S122: Update rescue resource allocation parameters based on the revised algorithm type; S123: Generate dynamic correction scheme data The original solution An optimized version based on the existing one, which is then pushed to the fire alarm terminal; S13. When actual response data indicates that the fire spread rate exceeds the predicted value γ%: S131: Activate multi-source data enhancement acquisition and increase the temperature sensor deployment density to α times the original deployment; S132: Reconstructing a static-dynamic combined data matrix based on enhanced data: in, This is the static baseline data after frequency adjustment. For enhanced dynamic monitoring data; S133: Processed using a spatiotemporal convolutional neural network Output a heat map (Ht) predicting the spread of the fire. S134: Overlay Ht onto the original fire prevention and control plan to form a three-dimensional fire control decision model: Three-dimensional fire control decision model

[0028] Ht represents the heat map of fire spread. The original prevention and control plan, This is the space addition operator. This is a function for generating three-dimensional decision models. Specific Implementation

[0029] Specific Implementation Example 1: System Overall Structure This system, called the Urban Fire Prevention and Control Intelligent Decision System, comprises three core parts: a data acquisition and processing end, a decision analysis end, and a decision execution end. These three parts share a system monitoring and alarm module. The data acquisition and processing end is responsible for acquiring urban fire-related information in real time. It connects to the urban fire management platform and a widely distributed IoT sensor network, collecting static basic data such as building structure information and historical fire records, as well as dynamic monitoring data such as temperature readings and smoke concentration. When the update frequency of static data is inconsistent with the frequency of dynamic monitoring, the system adjusts the sampling frequency and generates new adjusted data to unify the time scale. The decision analysis end uses this data to construct a combined data matrix, combines it with the Osprey optimization algorithm to analyze the optimal fire prevention and control decision algorithm type, and outputs feature data. The decision execution end receives this data, combines it into a summary of fire prevention and control decision information, and finally generates a prevention and control plan to be pushed to the fire terminal for execution. At the same time, the system monitoring and alarm module monitors the operating status of each end in real time to ensure stable system operation. When an anomaly is detected, such as data loss or processing delay, an alarm is immediately issued to notify the operator for manual handling. The entire system is deployed on the urban fire prevention and control decision platform, using a data bus to integrate the fire management platform and the IoT sensor network to achieve efficient data transmission and coordination.

[0030] Specific Implementation Example 2: System Module Refinement In the specific implementation of the data acquisition and processing end, the urban fire data acquisition and processing module is divided into a static data acquisition unit, a dynamic data acquisition unit, and a frequency adaptive adjustment unit. The static data acquisition unit extracts detailed building structure drawings, historical fire databases, and fire protection facility location maps from the fire management platform. The dynamic data acquisition unit obtains real-time monitoring data, including temperature fluctuations and smoke concentration changes, from temperature sensors and smoke sensors distributed throughout the city. The frequency adaptive adjustment unit measures the sampling frequency of all data, generates frequency data, and performs consistency judgment. When the frequencies do not match, such as static data being updated daily while dynamic data is updated every second, the unit uses an interpolation algorithm to adjust the frequency of the static data, generating standardized adjusted data. The fire prevention and control decision analysis module includes a data combination unit and an algorithm matching engine. The data combination unit aligns the adjusted static and dynamic data with the time axis to form a static-dynamic combined data matrix. The algorithm matching engine runs the Osprey optimization algorithm to simulate the natural optimization process. This algorithm explores the matching possibilities of different decision algorithms and finally outputs the optimal algorithm type feature data, such as the evacuation route planning algorithm identifier. The fire prevention and control decision execution module includes a scheme generation unit and an execution feedback unit. The scheme generation unit calls the matched algorithm to process the data and generate a prevention and control scheme that includes evacuation routes, allocation of rescue resources, and fire control strategies. The execution feedback unit immediately sends the scheme to the firefighters' terminal equipment and collects execution feedback data to ensure real-time response.

[0031] Specific Implementation Example 3: Basic Method Flow This method is applied to urban fire prevention and control decision-making. The steps include data acquisition, frequency processing, data combination, and decision execution. First, static basic data such as building layout information and dynamic monitoring data such as temperature sensor readings are collected. Then, the sampling frequency of these data is measured to generate frequency data and perform consistency judgment processing. If the frequencies are the same, the adjustment step is skipped. If they are different, such as if the static data updates slowly, the frequency of the static basic data is adjusted to generate adjusted data. Then, the adjusted static data and dynamic data are combined to construct an urban fire static and dynamic combined data matrix. Based on this matrix, the system is matched with standard data corresponding to different fire prevention and control decision-making algorithms. The Osprey optimization algorithm explores the optimization process and outputs target algorithm type feature data. Finally, the combined data matrix and feature data form the fire prevention and control decision summary data. The corresponding decision-making algorithm is called to generate a specific fire prevention and control plan. The core logic of the Osprey optimization algorithm involves initializing the Osprey population to simulate random positions, updating the positions in the exploration phase to discover new areas, optimizing the positions in the development phase to achieve precise matching, and iterating until the optimal algorithm type is found.

[0032] Specific Implementation Example 4: Detailed Method Steps In the specific implementation of the method, the data collection step obtains static basic data, including building structure drawings, fire protection facility distribution maps, historical fire statistics, and population density maps, through the urban fire management platform. Simultaneously, an IoT sensor network is deployed in urban building clusters to collect dynamic monitoring data in real time, such as temperature curves, smoke concentration trends, flammable gas data, and video surveillance images. After this data is imported into the urban fire prevention and control decision-making platform, a bidirectional search algorithm is used to scan the platform's storage using frequency keywords to identify the frequency of the static basic data and the frequency of the dynamic monitoring data, with the unit being Hertz. Frequency consistency judgment processing obtains the frequency values ​​and performs numerical comparisons. If the frequency difference is within a preset threshold, it is determined that the frequencies are the same; otherwise, it is determined that the frequencies are different and an identifier is output. When the frequencies are inconsistent, a linear interpolation algorithm is used to adjust the sampling frequency of the static basic data to match the dynamic frequency, generating adjusted data. This algorithm smoothly adds data. The system fills in the frequency difference at the data points, constructs a static and dynamic combined data matrix, aligns the adjusted static data and the original dynamic data according to the time series to generate a complete matrix representing the comprehensive situation of urban fires, and matches the fire prevention and control decision-making algorithm type. It uses a pre-established algorithm standard library matrix containing multiple standard feature data and employs the Osprey optimization algorithm process, starting from a random initial position, simulating extensive search and updating the position in the exploration phase, and fine-tuning the position in the development phase. Through multiple iterations, it outputs the algorithm type identifier with the highest matching degree as feature data. Finally, the fire prevention and control decision summary data combines this matrix and feature data to form a complete decision input, calls the corresponding algorithm function to process fire feature parameters, generates prevention and control plans such as optimized evacuation route maps, rescue resource deployment lists, and specific strategies for fire control measures, and executes them to ensure urban safety in fire response. All steps are monitored in real time by the system monitoring mechanism, and alarms are triggered in case of abnormalities for manual intervention.

[0033] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A big data-driven intelligent decision-making system for urban fire prevention and control, characterized in that: It includes a data acquisition and processing terminal, a decision analysis terminal, and a decision execution terminal, all of which are equipped with a system monitoring and alarm module. The data acquisition and processing terminal is used to collect static basic data and dynamic monitoring data of urban fires in real time by connecting the fire management platform and the Internet of Things sensor network. When the frequencies are different, the static basic data sampling frequency is adjusted to generate static basic data adjustment data. The decision analysis terminal is used to construct a static and dynamic combined data matrix of urban fire based on static basic data, static basic data adjustment data and dynamic monitoring data, and to generate target fire prevention and control decision algorithm type feature data using the Osprey optimization algorithm. The decision execution terminal is used to combine the static and dynamic combined data matrix of urban fires with the characteristic data of the target fire prevention and control decision algorithm type into fire prevention and control decision summary data. The system monitoring and alarm module is used to monitor the operating status of the data acquisition and processing terminal, the decision analysis terminal, and the decision execution terminal in real time through the system status monitor, issue alarm reminders when the system is operating abnormally, and provide feedback for manual handling. The system is deployed on the urban fire prevention and control decision platform and is connected to the fire management platform and the Internet of Things sensor network via a data bus.

2. The big data-driven intelligent decision-making system for urban fire prevention and control according to claim 1, characterized in that: The urban fire data acquisition and processing module includes: Static data acquisition unit: Extracts building structure and historical fire data from the fire management platform; Dynamic data acquisition unit: Collects dynamic data in real time from temperature sensors and smoke sensors; Frequency adaptive adjustment unit: performs frequency consistency judgment and interpolation adjustment; Based on the collected data, the data sampling frequency is measured to generate static basic data frequency data and dynamic monitoring data frequency data. The sampling frequency consistency is judged based on the frequency data. When the frequencies are different, the sampling frequency of the static basic data is adjusted to generate static basic data adjustment data. The fire prevention and control decision analysis module includes: Data combination unit: Constructs static and dynamic combined data matrices; Algorithm matching engine: Runs the Osprey optimization algorithm to match decision algorithm types; Based on the matrix, the standard static and dynamic combination data corresponding to different fire prevention and control decision algorithms are matched, and the Osprey optimization algorithm is used to generate target fire prevention and control decision algorithm type feature data. The fire prevention and control decision execution module includes: Solution generation unit: Invokes the target algorithm to generate evacuation routes and resource allocation plans; Execution Feedback Unit: Pushes the plan to the fire terminal in real time and collects execution data; The target decision algorithm is invoked to process and generate prevention and control plan data, which includes evacuation route planning, rescue resource allocation plan and fire control strategy, and the plan data is sent to the fire terminal for execution.

3. A big data-driven intelligent decision-making method for urban fire prevention and control, characterized in that: The method includes the following steps: S1. Collect static basic data and dynamic monitoring data of urban fires; S2. Based on the static basic data and dynamic monitoring data, perform data sampling frequency measurement and processing to generate static basic data frequency data and dynamic monitoring data frequency data; S3. Based on the static basic data frequency data and the dynamic monitoring data frequency data, perform sampling frequency consistency judgment processing to generate frequency consistency judgment data; when the frequencies are the same, directly execute step S5. S4. When the frequencies are different, the sampling frequency of the static basic data is adjusted based on the static basic data, the frequency data of the static basic data and the frequency data of the dynamic monitoring data to generate static basic data adjustment data. S5. Based on the static basic data, the static basic data adjustment data, and the dynamic monitoring data, perform combined processing of urban fire static and dynamic data to construct an urban fire static and dynamic combined data matrix. S6. Based on the urban fire static and dynamic combination data matrix and the standard static and dynamic combination data corresponding to different fire prevention and control decision algorithms, perform fire prevention and control decision algorithm type matching processing to generate target fire prevention and control decision algorithm type feature data. S7. Combine the static and dynamic combination data matrix of urban fire with the characteristic data of the target fire prevention and control decision algorithm type to form fire prevention and control decision summary data, and call the corresponding decision algorithm to process and generate urban fire prevention and control plan data. Specifically, S6 employs the Osprey optimization algorithm for fire prevention and control decision-making algorithm type matching, including: initializing the Osprey population position, updating the position during the exploration phase, optimizing the position during the development phase, and iteratively outputting the optimal matching algorithm type.

4. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 3, characterized in that: S3 specifically includes: S31. Collect static basic data through the urban fire management platform, including building structure data, fire protection facility distribution data, historical fire data, and population density data; S32. Real-time dynamic monitoring data is collected through an Internet of Things (IoT) sensor network deployed in urban building clusters, including temperature data, smoke concentration data, gas concentration data, and video surveillance data.

5. The big data-driven intelligent decision-making method and system for urban fire prevention and control according to claim 4, characterized in that: S4 includes: S41. Import static basic data and dynamic monitoring data into the urban fire prevention and control decision-making platform; S42. Use a bidirectional search algorithm to search for data sampling frequency information in the decision-making platform based on frequency keywords, and generate static basic data frequency data. and dynamic monitoring data frequency data All units are Hertz.

6. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 5, characterized in that: S5 includes: S51, Obtain and ; S52, to and Perform numerical comparison: like The output frequency is the same as the judgment flag; like The output frequency is different, which is a judgment indicator; in This is a preset threshold.

7. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 6, characterized in that: S6 includes: S61. When frequencies are inconsistent, first check the static basic data. Normalize it to map it to the range [0,1]: in, This refers to the normalized static basic data; Using linear interpolation algorithm sampling frequency Adjust to Generate static basic data and adjust data ,satisfy: in, This represents a linear interpolation function. Normalization ensures that data of different dimensions remain consistent during the interpolation process, eliminating the risk of fusion distortion.

8. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 7, characterized in that: S7 includes: S71. Unify the time scale, firstly, for the frequency-adjusted static basic data... and dynamic monitoring data Normalization is performed to map each feature value to the range [0,1]. in, and These are the normalized data; Subsequently, a combined static and dynamic data matrix of urban fires was constructed. : in, This is a static and dynamic combined data matrix of urban fires, with dimensions of [missing information]. , It is the feature number of static data. It is the feature number of dynamic data. It refers to the number of time points. This is the static baseline data after frequency adjustment. For dynamic monitoring data, normalization processing is emphasized to be completed before matrix construction to eliminate the influence of dimensions and improve the reliability of feature fusion.

9. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 8, characterized in that: S8 includes: S81. Establish a standard library matrix for fire prevention and control decision-making algorithms: in, This is the matrix of the standard library for fire prevention and control decision-making algorithms, with dimension 1. h is the number of algorithm types, and f is the standard feature dimension for each algorithm type. Indicates the first Standard static and dynamic combined data characteristics corresponding to each algorithm type; S82, using the Osprey optimization algorithm to... and To perform matching, including: S821. Initialization: Randomly generate the locations of the osprey population and normalize them to the range [0,1]. in, , For the search space boundary, As the lower limit, The upper limit, ∈[0,1] is a random number. The position is the normalized position; normalization eliminates the influence of boundary dimensions. S822, Exploration Phase: Update positions according to standard exploration logic, simulating random search behavior: Update location, where, , For the search space boundary, ∈[0,1], The population location is randomly selected and normalized to [0,1]. S823, Development Phase: Optimize the position according to standard development logic, simulate local fine-grained search, and base the current optimal position. : Optimize the location, among which, To find the current optimal position, normalize to [0,1]. , For boundary parameters, ∈[0,1], As the attenuation factor, This represents the number of iterations. S824, Iterative Output and Highest matching Corresponding algorithm type identifier ,in The target fire prevention and control decision-making algorithm type feature data.

10. The big data-driven intelligent decision-making method for urban fire prevention and control according to claim 9, characterized in that: S9 includes: S91. Constructing a summary data set for fire prevention and control decisions: in, Data is collected for fire prevention and control decision-making. Indicates will and This data structure will be combined into a tuple. and These are linked together and used as input for decision-making algorithms. S92, Call Corresponding decision algorithm processing The fire characteristic parameters are used to generate prevention and control plan data, including evacuation route planning, rescue resource allocation plans, and fire control strategies. in, For data related to the prevention and control plan, This is the function to execute the decision algorithm.

Images