A smart fire safety prediction linkage management method and platform

Abstract

The present invention relates to the field of intelligent fire protection technology, and in particular to a smart fire safety prediction linkage management method and platform. The method comprises the following steps: collecting circuit data, performing circuit abnormality feature analysis based on the circuit data to generate circuit abnormality feature data; collecting environmental infrared spectrum monitoring data; performing combustible gas leakage feature analysis based on environmental infrared spectrum monitoring to generate combustible gas leakage feature data; collecting fire hydrant water test operation monitoring data; performing fire hydrant abnormality feature analysis based on fire hydrant water test operation monitoring data to generate fire hydrant abnormality feature data; performing data linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data to generate abnormal fire protection linkage data; performing fire safety prediction demand assessment on abnormal fire protection linkage data to generate fire safety prediction demand data. The present invention realizes efficient fire safety prediction management through linkage analysis of fire protection data.

Description

Technical Field

[0001] The present invention relates to the field of smart fire protection technology, and in particular to a smart fire safety prediction linkage management method and platform. Background Art

[0002] Smart fire protection is to use advanced technologies such as the Internet of Things, big data, and artificial intelligence to upgrade the fire protection system to an intelligent level, and achieve intelligent early warning, accurate monitoring, and efficient response to fires. Through the comprehensive application of sensors, monitoring equipment, and data analysis platforms, smart fire protection can collect and analyze fire-related data in real time, provide fire hazard predictions, emergency response suggestions, and decision-making support, thereby improving the scientificity and effectiveness of fire protection management and protecting people's lives and property. However, the traditional smart fire safety prediction management method does not respond to fire-related information processing in a timely manner, resulting in a lag in fire safety prediction management, and the linkage management effect of fire safety prediction is poor, which cannot guarantee the accuracy and efficiency of fire prevention and control. Summary of the invention

[0003] Based on this, the present invention provides a smart fire safety prediction linkage management method and platform to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a smart fire safety prediction linkage management method includes the following steps:

[0005] Step S1: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data and circuit temperature data; performing circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; performing local temperature anomaly analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature anomaly data; performing circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature anomaly data to generate circuit abnormality feature data;

[0006] Step S2: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; performing combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; performing combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data;

[0007] Step S3: using the fire hydrant monitor to monitor the fire hydrant water test operation of the fire hydrant equipment, and generating fire hydrant water test operation monitoring data; performing abnormal vibration mode analysis of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generating fire hydrant pipeline abnormal vibration mode data; performing abnormal analysis of the fire hydrant water test pressure according to the fire hydrant water test operation monitoring data, and generating fire hydrant water test pressure abnormality data; performing fire hydrant abnormal feature analysis based on the fire hydrant pipeline abnormal vibration mode data and the fire hydrant water test pressure abnormality data, and generating fire hydrant abnormal feature data;

[0008] Step S4: Based on the rule engine technology, the circuit abnormal feature data, the combustible gas leakage feature data and the fire hydrant abnormal feature data are processed in a data linkage manner to generate abnormal fire linkage data; a fire safety prediction demand assessment of a single abnormal feature is performed on the abnormal fire linkage data to generate fire safety prediction demand data; the fire safety prediction demand data is fed back to the terminal to execute the fire safety prediction intelligent repair operation.

[0009] The present invention collects circuit data, including circuit distribution data, circuit current data and circuit temperature data, through an electrical appliance monitoring detector, to fully understand the circuit operation status and achieve accurate data acquisition. The circuit residual current distribution analysis is performed according to the circuit distribution data and the circuit current data, which is helpful to identify potential overload and leakage problems in the circuit. The circuit temperature data is analyzed for local temperature anomaly according to the circuit residual current distribution data, and the hot spots that may exist in the circuit are accurately located to prevent fires caused by local overheating. The circuit abnormality feature analysis is performed based on the circuit residual current distribution data and the local circuit temperature anomaly data, and circuit faults and fire hazards are warned in advance. The data information of the problematic circuit is extracted, and the circuit abnormality weight feature is assigned based on the temperature anomaly of the problematic circuit, so as to improve the safety management level of the electrical appliance. The infrared spectrum monitoring equipment is used to perform environmental infrared spectrum monitoring processing on the preset outdoor monitoring area, and the changes in gas composition in the environment are monitored in real time. The characteristic absorption peak analysis of the combustible gas spectrum is performed on the environmental infrared spectrum monitoring data, and the combustible gas composition and concentration in the environment are accurately identified. The combustible gas leakage feature analysis of the outdoor monitoring area is performed based on the combustible gas spectrum absorption peak data, and the combustible gas leakage risk is discovered and warned in time, so as to prevent fire accidents caused by gas leakage and improve the effectiveness of environmental safety monitoring. Fire hydrant monitors are used to monitor and process the fire hydrant water test operation of fire hydrant equipment to ensure that the fire hydrant equipment can work normally under normal conditions. According to the water test operation monitoring data, abnormal vibration mode analysis of fire hydrant pipelines is carried out to identify potential vibration problems of fire hydrant pipelines and prevent pipeline rupture or damage. Based on the water test operation monitoring data, abnormal analysis of fire hydrant water test pressure is carried out to timely discover and warn of insufficient or excessive water pressure problems to ensure stable and reliable water pressure during fire extinguishing. Fire hydrant abnormal feature analysis is carried out based on the pipeline abnormal vibration mode data and water test pressure abnormal data to comprehensively evaluate the operating status of fire hydrant equipment and improve the maintenance and management level of fire hydrant systems. Based on the rule engine technology, circuit abnormal feature data, combustible gas leakage feature data and fire hydrant abnormal feature data are linked and processed to achieve comprehensive analysis of multi-source data and improve the accuracy and comprehensiveness of fire warning. Abnormal fire linkage data is used to conduct fire safety prejudgment demand assessment of single abnormal features, accurately identify and evaluate fire safety hazards, and provide a scientific basis for subsequent response measures. Feedback the fire safety prediction demand data to the terminal, perform fire safety prediction intelligent repair operations, automatically handle hidden dangers and equipment maintenance, improve the response speed and repair efficiency of the fire protection system, and enhance the overall fire safety protection. Therefore, the intelligent fire safety prediction linkage management method of the present invention analyzes the abnormal related information in the fire protection data, so as to immediately respond to the processing of fire safety response, ensure the timeliness of fire safety prediction management, and effectively implement the linkage management of fire safety prediction, so as to ensure the accuracy and efficiency of fire prevention and control.

[0010] This specification provides a smart fire safety prediction linkage management platform for executing the smart fire safety prediction linkage management method as described above. The smart fire safety prediction linkage management platform includes:

[0011] A circuit abnormality feature analysis module is used to collect circuit data of circuit equipment using an electrical appliance monitoring detector to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data, and circuit temperature data; perform circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; perform local temperature abnormality analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature abnormality data; perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data;

[0012] The combustible gas leakage characteristic analysis module is used to use infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; perform combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; perform combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data;

[0013] The fire hydrant abnormal feature module is used to use the fire hydrant monitor to monitor and process the fire hydrant water test operation of the fire hydrant equipment, and generate the fire hydrant water test operation monitoring data; to analyze the abnormal vibration mode of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generate the fire hydrant pipeline abnormal vibration mode data; to analyze the abnormal pressure of the fire hydrant water test according to the fire hydrant water test operation monitoring data, and generate the fire hydrant water test pressure abnormal data; to analyze the abnormal characteristics of the fire hydrant based on the abnormal vibration mode data of the fire hydrant pipeline and the abnormal data of the fire hydrant water test pressure, and generate the fire hydrant abnormal feature data;

[0014] The fire linkage safety intelligent processing module is used to perform data linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data based on rule engine technology to generate abnormal fire linkage data; perform fire safety prediction demand assessment on the abnormal fire linkage data with a single abnormal feature to generate fire safety prediction demand data; and feed back the fire safety prediction demand data to the terminal to execute fire safety prediction intelligent repair operations.

[0015] The beneficial effect of the present application is that the present invention uses electrical monitoring detectors to collect detailed data on circuit equipment, including circuit distribution, current, temperature and other information, so that it can deeply understand the operating status of the circuit system and discover potential circuit problems and safety hazards early; analyze the characteristics of combustible gas leakage through infrared spectroscopy monitoring equipment, timely discover possible gas leakage sources, and reduce fire risks; monitor the water test operation of fire hydrants, analyze pipeline vibration and water pressure anomalies, and ensure the normal function of fire hydrants. Through detailed analysis of circuit current data, including residual current characteristic analysis, current waveform monitoring, high-order harmonic analysis, etc., it is possible to accurately identify potential anomalies in the circuit and prevent the occurrence of electrical fires; through infrared spectroscopy and chemical group analysis, combustible gas leaks in outdoor environments can be accurately identified and quantified to improve the accuracy of fire prevention; through vibration pattern analysis and pressure anomaly detection, it is ensured that fire hydrants can work normally in emergency situations and enhance fire extinguishing capabilities. The rule engine technology is used to link the abnormal circuit, combustible gas leakage and fire hydrant abnormal data to achieve a comprehensive evaluation of multiple abnormal data; the long short-term memory neural network algorithm is used to analyze the time series characteristics of abnormal fire linkage data to improve the ability to predict fire risks; a fire safety prediction strategy gradient model is constructed, and the prediction strategy is optimized through strategy functions and reward analysis to improve the accuracy and efficiency of fire safety warnings. The fire safety prediction demand data generated based on the intelligent model can evaluate and respond to the safety needs of a single abnormal feature in real time, feed back the prediction demand data to the terminal, perform intelligent repair operations, and take timely measures to repair, ensuring the safety and reliability of the fire protection system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the steps of a smart fire safety prediction linkage management method of the present invention;

[0017] Figure 2 for Figure 1 Detailed implementation steps of step S1 in FIG.

[0018] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0019] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] To achieve this, please refer to Figures 1 to 3 The present invention provides a smart fire safety prediction linkage management method, comprising the following steps:

[0024] Step S1: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data and circuit temperature data; performing circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; performing local temperature anomaly analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature anomaly data; performing circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature anomaly data to generate circuit abnormality feature data;

[0025] Step S2: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; performing combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; performing combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data;

[0026] Step S3: using the fire hydrant monitor to monitor the fire hydrant water test operation of the fire hydrant equipment, and generating fire hydrant water test operation monitoring data; performing abnormal vibration mode analysis of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generating fire hydrant pipeline abnormal vibration mode data; performing abnormal analysis of the fire hydrant water test pressure according to the fire hydrant water test operation monitoring data, and generating fire hydrant water test pressure abnormality data; performing fire hydrant abnormal feature analysis based on the fire hydrant pipeline abnormal vibration mode data and the fire hydrant water test pressure abnormality data, and generating fire hydrant abnormal feature data;

[0027] Step S4: Based on the rule engine technology, the circuit abnormal feature data, the combustible gas leakage feature data and the fire hydrant abnormal feature data are processed in a data linkage manner to generate abnormal fire linkage data; a fire safety prediction demand assessment of a single abnormal feature is performed on the abnormal fire linkage data to generate fire safety prediction demand data; the fire safety prediction demand data is fed back to the terminal to execute the fire safety prediction intelligent repair operation.

[0028] The present invention collects circuit data, including circuit distribution data, circuit current data and circuit temperature data, through an electrical appliance monitoring detector, to fully understand the circuit operation status and achieve accurate data acquisition. The circuit residual current distribution analysis is performed according to the circuit distribution data and the circuit current data, which is helpful to identify potential overload and leakage problems in the circuit. The circuit temperature data is analyzed for local temperature anomaly according to the circuit residual current distribution data, and the hot spots that may exist in the circuit are accurately located to prevent fires caused by local overheating. The circuit abnormality feature analysis is performed based on the circuit residual current distribution data and the local circuit temperature anomaly data, and circuit faults and fire hazards are warned in advance. The data information of the problematic circuit is extracted, and the circuit abnormality weight feature is assigned based on the temperature anomaly of the problematic circuit, so as to improve the safety management level of the electrical appliance. The infrared spectrum monitoring equipment is used to perform environmental infrared spectrum monitoring processing on the preset outdoor monitoring area, and the changes in gas composition in the environment are monitored in real time. The characteristic absorption peak analysis of the combustible gas spectrum is performed on the environmental infrared spectrum monitoring data, and the combustible gas composition and concentration in the environment are accurately identified. The combustible gas leakage feature analysis of the outdoor monitoring area is performed based on the combustible gas spectrum absorption peak data, and the combustible gas leakage risk is discovered and warned in time, so as to prevent fire accidents caused by gas leakage and improve the effectiveness of environmental safety monitoring. Fire hydrant monitors are used to monitor and process the fire hydrant water test operation of fire hydrant equipment to ensure that the fire hydrant equipment can work normally under normal conditions. According to the water test operation monitoring data, abnormal vibration mode analysis of fire hydrant pipelines is carried out to identify potential vibration problems of fire hydrant pipelines and prevent pipeline rupture or damage. Based on the water test operation monitoring data, abnormal analysis of fire hydrant water test pressure is carried out to timely discover and warn of insufficient or excessive water pressure problems to ensure stable and reliable water pressure during fire extinguishing. Fire hydrant abnormal feature analysis is carried out based on the pipeline abnormal vibration mode data and water test pressure abnormal data to comprehensively evaluate the operating status of fire hydrant equipment and improve the maintenance and management level of fire hydrant systems. Based on the rule engine technology, circuit abnormal feature data, combustible gas leakage feature data and fire hydrant abnormal feature data are linked and processed to achieve comprehensive analysis of multi-source data and improve the accuracy and comprehensiveness of fire warning. Abnormal fire linkage data is used to conduct fire safety prejudgment demand assessment of single abnormal features, accurately identify and evaluate fire safety hazards, and provide a scientific basis for subsequent response measures. Feedback the fire safety prediction demand data to the terminal, perform fire safety prediction intelligent repair operations, automatically handle hidden dangers and equipment maintenance, improve the response speed and repair efficiency of the fire protection system, and enhance the overall fire safety protection. Therefore, the intelligent fire safety prediction linkage management method of the present invention analyzes the abnormal related information in the fire protection data, so as to immediately respond to the processing of fire safety response, ensure the timeliness of fire safety prediction management, and effectively implement the linkage management of fire safety prediction, so as to ensure the accuracy and efficiency of fire prevention and control.

[0029] As an embodiment of the present invention, refer to Figure 1 The above is a schematic diagram of the steps of a smart fire safety prediction linkage management method of the present invention. In this embodiment, the smart fire safety prediction linkage management method includes the following steps:

[0030] Step S1: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data and circuit temperature data; performing circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; performing local temperature anomaly analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature anomaly data; performing circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature anomaly data to generate circuit abnormality feature data;

[0031] In the embodiment of the present invention, an electrical appliance monitoring detector, such as a digital multimeter, an oscilloscope, a current sensor, a temperature sensor, etc., is used to collect data from circuit equipment, and the detector is installed at key nodes of the circuit, such as a distribution box, a circuit breaker, a cable connector, etc., to collect circuit distribution data (such as circuit wiring diagram, load distribution, etc.), circuit current data (such as current intensity, waveform, etc.), and circuit temperature data (such as cable temperature, equipment surface temperature, etc.). The current data and circuit distribution data are used for calculation to determine the distribution of the residual current in the circuit, the current distribution and load model are used to analyze the distribution of the current, and an algorithm or software tool (such as MATLAB, LabVIEW) is used to calculate the distribution data of the residual current at each node and region, generate circuit residual current distribution data, and characterize the residual current situation at each point in the circuit. The circuit residual current distribution data is used to analyze the abnormal area in the circuit temperature data, and the area that may cause overheating is determined based on the residual current distribution data. Combined with the measurement results of the temperature sensor, the regional temperature abnormality analysis is performed to generate local circuit temperature abnormality data, indicating the temperature abnormal area in the circuit. Combined with the circuit residual current distribution data and the local circuit temperature anomaly data for comprehensive analysis, the anomaly detection algorithm (such as statistical analysis, machine learning model) is used to identify the abnormal characteristics in the circuit, analyze the abnormal patterns of the current and temperature data in the circuit, determine the potential fault points and risk areas, and generate circuit abnormality feature data to characterize the abnormal states and potential problems in the circuit, such as leakage anomaly, leakage temperature anomaly and other conditions and the corresponding abnormal nodes.

[0032] Step S2: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; performing combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; performing combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data;

[0033] In the embodiment of the present invention, infrared spectrum monitoring equipment, such as infrared spectrometer or Fourier infrared spectrometer, is used to perform environmental monitoring on a preset outdoor monitoring area. The infrared spectrum monitoring equipment is installed at a suitable position in the monitoring area to ensure that the entire area can be covered. Environmental monitoring is performed, and infrared spectrum data, including absorption peaks, spectral intensity, etc., are recorded to generate environmental infrared spectrum monitoring data for subsequent analysis. The environmental infrared spectrum monitoring data is analyzed to extract the spectral characteristics of the combustible gas. Spectral analysis software, such as FTIR analysis tools or Fourier frequency domain data analysis, is used to identify the characteristic absorption peaks of the combustible gas, determine the position and intensity of the absorption peaks, analyze their association with the combustible gas, generate the spectral absorption peak data of the combustible gas, and characterize the characteristics of the combustible gas in the environment. Leakage characteristic analysis is performed based on the spectral absorption peak data of combustible gas, and the spectral absorption peak data is compared with a preset gas concentration model. The gas concentration model is a mathematical model that predicts the gas concentration corresponding to the spectral intensity by training a supervised learning model based on historical spectral characteristics and historical gas concentration characteristics. The location and concentration of the leakage source are determined, and data processing tools (such as gas leak detection algorithms) are used to analyze the combustible gas leakage characteristics of the monitored area, generate combustible gas leakage characteristic data, and describe the combustible gas leakage situation in the outdoor monitoring area, such as combustible gas composition, combustible gas concentration and other characteristics.

[0034] Step S3: using the fire hydrant monitor to monitor the fire hydrant water test operation of the fire hydrant equipment, and generating fire hydrant water test operation monitoring data; performing abnormal vibration mode analysis of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generating fire hydrant pipeline abnormal vibration mode data; performing abnormal analysis of the fire hydrant water test pressure according to the fire hydrant water test operation monitoring data, and generating fire hydrant water test pressure abnormality data; performing fire hydrant abnormal feature analysis based on the fire hydrant pipeline abnormal vibration mode data and the fire hydrant water test pressure abnormality data, and generating fire hydrant abnormal feature data;

[0035] In the embodiment of the present invention, a fire hydrant monitor, such as a pressure sensor, a flow meter, a vibration sensor, etc., is used to monitor the water test operation of the fire hydrant equipment. The fire hydrant monitor is installed and calibrated on the designated fire hydrant equipment, and is intelligently connected with the terminal to perform the fire hydrant water test operation, record the operation data of the fire hydrant under different flow and pressure conditions, monitor and record the water flow, pressure change and pipeline vibration parameters of the fire hydrant, and generate fire hydrant water test operation monitoring data, including fire hydrant water test data and fire hydrant pipeline vibration data. The fire hydrant pipeline vibration data is used for analysis to identify the vibration mode, and the fire hydrant pipeline vibration data is analyzed in the time domain and frequency domain (such as FFT analysis). Clustering algorithms (such as K-means, DBSCAN) are used to group the vibration data, identify common vibration modes, generate abnormal vibration mode data of the fire hydrant pipeline, and describe the abnormal characteristics of the vibration mode. Perform pressure anomaly analysis on the fire hydrant water test data, calculate the pressure loss according to the fluid dynamics model (such as Darcy-Weisbach equation, Hazen-Williams equation), compare the actual pressure data with the expected pressure value, identify pressure anomalies, perform pressure anomaly statistics and trend analysis, determine the degree of anomaly and possible causes, generate fire hydrant water test pressure anomaly data, and indicate the pressure problems that occurred in the fire hydrant during the water test. Comprehensively analyze the abnormal vibration pattern data of the fire hydrant pipeline and the abnormal pressure data of the fire hydrant water test, combine the vibration pattern data and the pressure anomaly data, and apply multidimensional data analysis technology (such as principal component analysis PCA, anomaly detection algorithm) to identify and extract abnormal characteristics of fire hydrant equipment, including the combined impact of vibration and pressure anomalies, such as pressure unfavorable nodes and structural unfavorable nodes of fire hydrants, etc., generate fire hydrant abnormal characteristic data, and characterize the abnormal conditions and potential problems of fire hydrant equipment.

[0036] Step S4: Based on the rule engine technology, the circuit abnormal feature data, the combustible gas leakage feature data and the fire hydrant abnormal feature data are processed in a data linkage manner to generate abnormal fire linkage data; a fire safety prediction demand assessment of a single abnormal feature is performed on the abnormal fire linkage data to generate fire safety prediction demand data; the fire safety prediction demand data is fed back to the terminal to execute the fire safety prediction intelligent repair operation.

[0037] In an embodiment of the present invention, rule engine technology is used to perform linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data, and the rules of the rule engine (such as IF-THEN rules) are set, including the trigger conditions of circuit abnormality, combustible gas leakage and fire hydrant abnormality and their associated logic, and define how to combine data from different sources together, such as identifying abnormal nodes of circuit leakage and abnormal temperature conditions, whether there will be an association with combustible gas leakage, and whether the fire hydrant can perform fire fighting. The rule engine performs data matching and logical judgment according to the set rules, identifies the correlation between abnormal events, and applies the rule engine to fuse different data to generate comprehensive abnormal fire fighting linkage data, which comprehensively reflects the linkage abnormal status of circuits, combustible gas and fire hydrant equipment. The fire safety prediction demand assessment of a single abnormal feature of abnormal fire linkage data is performed, and a fire safety prediction model, such as a policy gradient algorithm, is used to analyze each action in the abnormal fire linkage data, that is, a separate circuit abnormality, a separate combustible gas abnormality, etc., as a strategy analysis decision, so as to evaluate the impact of different abnormal features on fire safety, generate fire safety prediction demand data, and generate fire safety prediction demand data for the priority of formulating corresponding safety prediction measures. The fire safety prediction demand data is transmitted to the terminal system (such as the control center, the automation control system), and the terminal system performs intelligent repair operations (such as automatic alarm, equipment adjustment, starting fire protection equipment, etc.) according to the prediction demand data, and implements intelligent repair measures to ensure that the fire safety system can respond quickly and perform appropriate processing when an abnormality occurs.

[0038] Preferably, step S1 comprises the following steps:

[0039] Step S11: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data;

[0040] Step S12: performing residual current characteristic analysis according to the circuit current data to generate residual current characteristic data;

[0041] Step S13: performing circuit residual current distribution identification according to the circuit distribution data and the residual current characteristic data to generate circuit residual current distribution data;

[0042] Step S14: extracting target circuit temperature data of the circuit residual current distribution from the circuit temperature data according to the circuit residual current distribution data to obtain target circuit temperature data;

[0043] Step S15: performing spatial temperature distribution analysis on the target circuit temperature data to generate the target circuit spatial temperature distribution data;

[0044] Step S16: performing temperature gradient analysis according to the target circuit spatial temperature distribution data to generate target circuit temperature gradient data;

[0045] Step S17: performing local temperature anomaly analysis according to the target circuit temperature gradient data to generate local circuit temperature anomaly data;

[0046] Step S18: Perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data.

[0047] The present invention uses an electrical appliance monitoring detector to collect circuit data from circuit equipment, ensures comprehensive monitoring of the circuit state, and provides detailed circuit operation information, including current, temperature and circuit distribution, to provide basic data support for subsequent analysis. Performing residual current feature analysis based on circuit current data helps to identify abnormal current conditions in the circuit, such as leakage or overload, discover circuit hidden dangers in advance, and prevent fire risks caused by electrical appliance failures. Performing circuit residual current distribution identification based on circuit distribution data and residual current feature data, by detailed identification of the residual current distribution in the circuit, helps to accurately locate potential electrical problems and improve the accuracy of circuit safety monitoring. Extracting target circuit temperature data of the circuit residual current distribution from circuit temperature data based on the circuit residual current distribution data, by extracting key temperature data, focusing on the circuit area with problems, and providing key data support for subsequent temperature analysis. Performing spatial temperature distribution analysis on the target circuit temperature data, generating target circuit spatial temperature distribution data, identifying abnormal temperature areas by analyzing the temperature distribution in the circuit, preventing electrical fires caused by local overheating, and improving the safety of circuit operation. Perform temperature gradient analysis based on the spatial temperature distribution data of the target circuit. By analyzing the temperature change gradient, identify abnormal temperature fluctuations in the circuit, and further improve the early warning capability of electrical hidden dangers. Perform local temperature anomaly analysis based on the temperature gradient data of the target circuit. By analyzing the local temperature anomaly in detail, accurately locate the problem area in the circuit, provide timely early warning information, and prevent electrical failures caused by temperature anomalies. Perform circuit anomaly feature analysis based on the circuit residual current distribution data and local circuit temperature anomaly data. By comprehensively analyzing the current and temperature data, generate circuit anomaly feature data, and comprehensively identify potential hidden dangers in the circuit, such as circuits with residual current problems, and temperature anomaly information of circuits with residual current problems, to provide a basis for subsequent maintenance and repair, and improve the overall safety of the circuit system.

[0048] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation step flow chart of step S1 in the embodiment, step S1 in this embodiment includes:

[0049] Step S11: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data;

[0050] In an embodiment of the present invention, electrical appliance monitoring detectors, such as digital multimeters, oscilloscopes, current sensors, temperature sensors, etc., are used to intelligently connect the electrical appliance monitoring detectors with terminals through the Internet of Things technology to collect data on circuit equipment. Detectors are installed at key nodes of the circuit, such as distribution boxes, circuit breakers, cable connectors, etc., to collect circuit distribution data (such as circuit wiring diagrams, load distribution, etc.), circuit current data (such as current intensity, waveform, etc.), and circuit temperature data (such as cable temperature, equipment surface temperature, etc.).

[0051] Step S12: performing residual current characteristic analysis according to the circuit current data to generate residual current characteristic data;

[0052] In an embodiment of the present invention, input current and output current values ​​are extracted from the circuit current data. These data identify the residual current portion by calculating the current difference, usually including the portion of current imbalance. Then, the current waveform data is acquired using current waveform monitoring technology, and the current fundamental frequency signal is further analyzed to extract the high-order harmonic characteristics of the current. Intermittent residual current data analysis is performed based on these high-order harmonic data to identify possible current anomalies and residual current characteristics. The high-order harmonic data reflects the specific conditions of the residual current to generate residual current characteristic data, including the current intensity of the residual current (leakage current), the change trend, and the abnormal characteristics of the current waveform.

[0053] Step S13: performing circuit residual current distribution identification according to the circuit distribution data and the residual current characteristic data to generate circuit residual current distribution data;

[0054] In the embodiment of the present invention, the residual current distribution model of the circuit is established by combining the circuit distribution data and the residual current characteristic data. Through the data fusion technology, the characteristics of the current distribution are mapped to the circuit distribution diagram, and the distribution of the residual current in the circuit is analyzed using statistical and fitting methods to identify the areas in the circuit where there is significant residual current. These areas are identified based on the current intensity, change trend and abnormal characteristics of the current waveform of the residual current (leakage current) to generate the circuit residual current distribution data, which reflects the specific distribution of the residual current in the circuit.

[0055] Step S14: extracting target circuit temperature data of the circuit residual current distribution from the circuit temperature data according to the circuit residual current distribution data to obtain target circuit temperature data;

[0056] In an embodiment of the present invention, the circuit residual current distribution data is associated with the circuit temperature data to identify the circuit area corresponding to the residual current distribution. The relevant part of the circuit temperature data is extracted through data matching and interpolation methods, and the temperature data in the residual current distribution area is extracted to obtain the target circuit temperature data, which involves extracting and converting the temperature data of the current distribution area into the temperature data of the circuit residual current area, thereby ensuring the accuracy and relevance of the target circuit temperature data.

[0057] Step S15: performing spatial temperature distribution analysis on the target circuit temperature data to generate the target circuit spatial temperature distribution data;

[0058] In an embodiment of the present invention, a temperature distribution map is created using the temperature data of the target circuit. The temperature distribution of each area in the circuit is plotted using spatial analysis tools, such as heat map generation and interpolation analysis. The temperature data is processed using a spatial analysis algorithm to identify the concentrated areas and distribution patterns of the temperature. During the analysis, a spatial statistical method is used to determine the distribution characteristics of the temperature data in the circuit and identify heat sources or abnormal points. Finally, the spatial temperature distribution data of the target circuit is generated, clearly showing the temperature changes at different locations in the leakage area of ​​the circuit.

[0059] Step S16: performing temperature gradient analysis according to the target circuit spatial temperature distribution data to generate target circuit temperature gradient data;

[0060] In an embodiment of the present invention, a temperature field model is created using the spatial temperature distribution data of the target circuit, and a gradient calculation method is applied, which generally includes a differential or convolution operation, to calculate the temperature gradient of each point. The temperature gradient represents the rate and intensity of temperature change in space. By analyzing the temperature gradient data, the temperature change trend and temperature non-uniformity in the circuit can be identified. The temperature gradient analysis generally includes calculating the temperature change rate and identifying the hot spots or potential problem points of the circuit through the temperature gradient map, and generating the target circuit temperature gradient data, which describes the temperature change rate and intensity at different positions in the circuit.

[0061] Step S17: performing local temperature anomaly analysis according to the target circuit temperature gradient data to generate local circuit temperature anomaly data;

[0062] In an embodiment of the present invention, temperature gradient data is analyzed to identify areas with significant abnormal temperature changes, and an anomaly detection algorithm, such as a threshold-based method, cluster analysis, or a machine learning model, is applied to find local areas that deviate significantly from the normal temperature gradient distribution. These abnormal areas are areas where the temperature gradient is abnormally large, or areas where the gradient change is not as expected. These abnormal areas are analyzed in detail to determine whether there is local overheating or insufficient cooling, and local circuit temperature anomaly data is generated, which details the specific location and degree of the temperature anomaly in the circuit.

[0063] Step S18: Perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data.

[0064] In an embodiment of the present invention, the circuit residual current distribution data and the local circuit temperature anomaly data are comprehensively analyzed, and the two data sets are combined through data fusion technology to identify the abnormal area in the circuit, and the key features of the circuit anomaly are extracted using feature extraction methods, such as statistical analysis or pattern recognition technology. For example, by analyzing the correlation between current anomalies and temperature anomalies, possible fault points or abnormal patterns are identified. Comprehensive analysis tools, such as multidimensional analysis or association rule mining, are used to further verify and evaluate the accuracy and impact of abnormal features. Ultimately, circuit abnormality feature data is generated, which integrates the comprehensive features of current and temperature anomalies in the circuit to guide subsequent maintenance and repair work.

[0065] Preferably, step S12 comprises the following steps:

[0066] Step S121: Analyze the current difference between input and output according to the circuit current data to generate current difference data;

[0067] Step S122: monitoring current waveform data according to circuit current data to generate current waveform data;

[0068] Step S123: performing current fundamental frequency signal analysis according to the current waveform data to generate current fundamental frequency data;

[0069] Step S124: performing current high-order harmonic characteristic analysis on the current waveform data according to the current fundamental frequency data to generate current high-order harmonic characteristic data;

[0070] Step S125: performing intermittent residual current data analysis according to the current high-order harmonic characteristic data to generate intermittent residual current data;

[0071] Step S126: Analyze the residual current characteristics of the intermittent residual current data and the current difference data to generate residual current characteristic data.

[0072] The present invention performs input and output current difference analysis based on circuit current data, detects leakage or imbalance problems that may exist in the circuit, and quickly identifies abnormalities by comparing the difference between input and output currents to ensure the safety of circuit operation. Current waveform data monitoring is performed based on circuit current data. By monitoring the current waveform, detailed information on current changes is provided, which helps to identify the laws and abnormalities of current fluctuations and supports further current feature analysis. Current fundamental frequency signal analysis is performed based on current waveform data, the fundamental frequency components in the current signal are extracted, the basic characteristics of the current are analyzed, the main frequency components in the current are identified, and basic data is provided for subsequent high-order harmonic analysis. Current waveform data is analyzed for current high-order harmonic characteristics based on current fundamental frequency data, the high-order harmonic components in the current are analyzed, the abnormal frequency components in the current are identified, potential fault sources in electrical equipment are detected, and abnormal conditions of electrical equipment are detected in advance. Intermittent residual current data analysis is performed based on current high-order harmonic feature data, intermittent abnormalities in the current are identified, and detailed information on intermittent leakage in the circuit is provided, which helps to accurately detect intermittent faults of electrical equipment. Residual current characteristic analysis is performed based on current difference data and intermittent residual current data. In-depth residual current characteristic analysis is performed based on the current difference data and intermittent residual current data. Comprehensive current abnormality characteristics are provided to support rapid location and diagnosis of circuit faults and improve the level of electrical safety management.

[0073] In an embodiment of the present invention, current data of each part of the circuit is collected, and the input current and output current are recorded respectively. The difference calculation is performed on these data to determine the difference between the current inflow and outflow, such as calculating the direct difference between the input current and the output current or using statistical methods such as mean difference, and the current imbalance area is identified by analyzing the current difference data, which reflects the problem or load abnormality in the circuit, generates current difference data, and provides the current difference situation of each point in the circuit. The current data of the circuit is recorded using high-frequency sampling technology to form time series data, and these data are digitized to generate current waveform data. During the monitoring process, filtering technology needs to be applied to remove noise and smooth the waveform data. By analyzing the current waveform data, the change of the current over time, including amplitude, periodicity and any irregular fluctuations, is observed, and the current waveform data is generated to reflect the actual waveform and characteristics of the current signal. Perform Fourier transform (FFT) or other frequency domain analysis methods on the current waveform data to extract the fundamental frequency component of the current signal. The fundamental frequency is the main periodic component in the current waveform and represents the basic frequency of the current waveform, that is, it reflects the conventional current waveform range. During the analysis process, it is necessary to determine the amplitude and frequency of the fundamental frequency and compare it with the standard frequency. Through fundamental frequency analysis, the stability and frequency consistency of the current waveform are evaluated, and current fundamental frequency data is generated, which contains the frequency and amplitude information of the fundamental frequency. Use Fourier transform to perform frequency domain decomposition on the current waveform data to extract the fundamental frequency and its higher harmonic components. Higher harmonics refer to frequency components that are integer multiples of the fundamental frequency. By analyzing the amplitude and phase of these harmonics, the nonlinear characteristics of the current waveform can be identified, and the amplitude, phase and total harmonic distortion (THD) of each harmonic component can be calculated to generate current higher harmonic characteristic data to describe the higher harmonic components in the current signal and their effects. Collect current higher harmonic characteristic data and monitor intermittent fluctuations of the current signal. Use time series analysis methods, such as sliding window technology, to identify the intermittent characteristics of current fluctuations. During the analysis process, it is necessary to determine the periodicity, duration, and amplitude changes of current fluctuations. Thus, intermittent residual current conditions in the circuit are detected, indicating that there is an intermittent load imbalance or fault in the circuit to generate intermittent residual current data to characterize the intermittent characteristics of the current signal. Combine the current difference data with the intermittent residual current data, analyze the residual current characteristics in the circuit through data fusion technology, and use statistical analysis methods, such as regression analysis or correlation analysis, to evaluate the characteristics and patterns of the residual current. Identify the relationship between the current difference and the intermittent residual current, as well as the impact of these characteristics on circuit performance, generate residual current characteristic data, and describe in detail the characteristics of the residual current in the circuit and its impact.

[0074] Preferably, step S125 includes the following steps:

[0075] Step S1251: performing current harmonic amplitude characteristic analysis on current high-order harmonic characteristic data to generate current harmonic amplitude characteristic data;

[0076] Step S1252: performing current harmonic phase characteristic analysis on the current high-order harmonic characteristic data to generate current harmonic phase characteristic data;

[0077] Step S1253: performing current harmonic distortion characteristic analysis based on the current harmonic amplitude characteristic data and the current harmonic phase characteristic data to generate current harmonic distortion characteristic data;

[0078] Step S1254: performing intermittent residual current data analysis based on the current harmonic distortion characteristic data to generate intermittent residual current data.

[0079] The present invention performs current harmonic amplitude characteristic analysis on current high-order harmonic characteristic data to help identify the amplitude of each harmonic in the current, thereby revealing the harmonic problems caused by load imbalance or nonlinear load in electrical equipment, and providing key data for further analyzing the impact of harmonics on equipment performance. Current harmonic phase characteristic analysis is performed on current high-order harmonic characteristic data to determine the phase angle of current harmonics. By analyzing the phase relationship of different harmonic components, possible phase mismatch or time delay problems in the equipment are identified, and the operating stability of the electrical system is optimized. Current harmonic distortion characteristic analysis is performed based on current harmonic amplitude characteristic data and current harmonic phase characteristic data, comprehensively considering the amplitude and phase information of the harmonics, analyzing the harmonic distortion characteristics in the current, thereby identifying possible harmonic distortion problems in the equipment, helping to evaluate the health status of electrical equipment, and optimizing power quality. Intermittent residual current data analysis is performed based on the current harmonic distortion characteristic data to identify intermittent residual current in the current. By analyzing the impact of harmonic distortion on residual current, a detailed understanding of intermittent current anomalies is provided, which helps to promptly discover possible intermittent leakage problems in the electrical system and improve the safety and reliability of electrical equipment.

[0080] In the embodiment of the present invention, the current harmonic amplitude characteristic analysis is performed on the current high-order harmonic characteristic data, including the amplitude information of different harmonics, such as using the frequency domain data after Fourier transformation to extract the amplitude of each harmonic component, including determining the amplitude of the main harmonic frequency and calculating its ratio to the fundamental frequency component, using the peak detection method to identify the maximum amplitude of each harmonic, and performing statistical analysis to generate current harmonic amplitude characteristic data to reflect the harmonic amplitude characteristics in the current waveform and provide detailed information about the stability and quality of the current waveform. The current harmonic phase characteristic analysis is performed on the current high-order harmonic characteristic data, and the phase information in the current high-order harmonic characteristic data is processed, such as using the Fourier transformation result to extract the phase angle of each harmonic component, including calculating the phase difference of each harmonic, and evaluating its impact on the overall current waveform. Phase demodulation technology is used to obtain an accurate phase angle, and phase stability analysis is performed to generate current harmonic phase characteristic data, which reflects the phase information and changes of each harmonic component, and helps understand the phase consistency and interference of the current waveform. Combine the current harmonic amplitude characteristic data and the current harmonic phase characteristic data to calculate the total harmonic distortion (THD) of the current, which is an important indicator to measure the degree of harmonic distortion in the current waveform. Apply the nonlinear distortion model to analyze the current harmonic distortion characteristics of the current harmonic amplitude characteristic data and the current harmonic phase characteristic data. During the analysis process, it is necessary to consider the contribution of each harmonic amplitude and its phase to the overall distortion to evaluate the impact of harmonic distortion and generate current harmonic distortion characteristic data, which reflects the distortion of the current signal and its impact on circuit performance. Use the current harmonic distortion characteristic data to analyze the intermittent residual current phenomenon in the current signal. Use time series analysis methods such as sliding window technology or spectrum analysis to identify intermittent patterns in the current signal. During the analysis process, it is necessary to evaluate the fluctuation of current distortion and determine whether it is related to the intermittent residual current. Use statistical methods to calculate the amplitude and duration of the intermittent current to generate intermittent residual current data, which reflects the intermittent anomaly in the current signal and its possible impact.

[0081] Preferably, step S2 comprises the following steps:

[0082] Step S21: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data;

[0083] Step S22: Analyze the characteristic absorption peaks of the combustible gas spectrum on the environmental infrared spectrum monitoring data to generate the combustible gas spectrum absorption peak data;

[0084] Step S23: performing a combustible gas chemical group analysis based on the combustible gas spectrum absorption peak data to generate combustible gas chemical group data;

[0085] Step S24: performing a combustible gas copolymer analysis in the outdoor monitoring area according to the combustible gas chemical group data to generate combustible gas copolymer data;

[0086] Step S25: performing absorbance analysis on the spectral absorption peak data of the combustible gas to generate spectral absorbance data of the combustible gas;

[0087] Step S26: performing a characteristic analysis of the combustible gas concentration in the outdoor monitoring area according to the combustible gas spectral absorbance data to generate combustible gas concentration data;

[0088] Step S27: Based on the leaked combustible gas concentration data and the leaked combustible gas copolymer data, a combustible gas leakage characteristic analysis is performed in the outdoor monitoring area to generate combustible gas leakage characteristic data.

[0089] The present invention uses infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area, provides detailed information on gas components in the outdoor environment, lays a foundation for subsequent gas analysis, and ensures comprehensive monitoring of potential combustible gases in the environment. The environmental infrared spectrum monitoring data is analyzed for characteristic absorption peaks of the combustible gas spectrum to identify the characteristic wavelengths of the combustible gas in the environment. By analyzing the spectral absorption peaks, the existence and concentration of the combustible gas are accurately detected, providing key data for leak detection. According to the combustible gas spectrum absorption peak data, the chemical group analysis of the combustible gas can determine the specific chemical groups in the gas, help identify different types of combustible gases, further understand the chemical properties of the gas, and enhance the ability to identify the type of gas. According to the combustible gas chemical group data, the copolymer analysis of the combustible gas in the outdoor monitoring area is performed to identify the copolymer components in the gas, understand the complexity of the gas and the potential chemical reactions, and help to more accurately assess the danger and leakage risk of the combustible gas. The absorbance analysis of the combustible gas spectrum absorption peak data provides the light absorption intensity information of the combustible gas at a specific wavelength, which helps to accurately measure the gas concentration and improve the sensitivity and accuracy of gas detection. The characteristic analysis of the combustible gas concentration in the outdoor monitoring area based on the combustible gas spectral absorbance data can determine the actual concentration level of the gas, help identify potential gas leakage areas and concentration changes, and improve the real-time monitoring capability of gas leakage. The characteristic analysis of combustible gas leakage in the outdoor monitoring area is performed based on the leaked combustible gas concentration data and the leaked combustible gas copolymer data. The detailed characteristic analysis of combustible gas leakage is provided by considering the gas concentration and copolymer information, supporting the rapid location of the leakage source and taking corresponding safety measures to effectively prevent the risk of fire and explosion.

[0090] In the embodiment of the present invention, infrared spectrum monitoring equipment is used, such as infrared spectrometer or Fourier infrared spectrometer, and the infrared spectrum monitoring equipment is intelligently connected with the terminal through the Internet of Things technology to perform environmental monitoring on the preset outdoor monitoring area. The infrared spectrum monitoring equipment is installed at a suitable position in the monitoring area to ensure that the entire area can be covered, and environmental monitoring is performed to record infrared spectrum data, including absorption peaks, spectral intensity, spectral reflection position, etc., to generate environmental infrared spectrum monitoring data for subsequent analysis. The environmental infrared spectrum monitoring data is processed, and the absorption peaks of combustible gases are extracted through spectral analysis technology. Specific spectral processing algorithms, such as baseline correction and peak detection, are used to identify characteristic absorption peaks in the infrared spectrum. These absorption peaks correspond to specific combustible gas molecules, such as methane, propane, etc. The specific chemical composition of each absorption peak is determined by comparing the experimental data with the known spectral database. The generated spectral absorption peak data of the combustible gas will provide information about the presence and concentration of the combustible gas. Using the spectral absorption peak data of combustible gases, the chemical groups corresponding to each absorption peak are identified. Chemical spectral analysis techniques, such as Fourier transform infrared spectroscopy (FTIR) or Raman spectroscopy, are applied to analyze each absorption peak. Based on the known chemical group characteristics, the specific group composition of each spectral characteristic peak is determined. The absorption characteristics of different chemical groups need to be considered for quantitative and qualitative analysis. The generated chemical group data of combustible gases will help understand the types of combustible gases and their chemical composition in the outdoor monitoring area. According to the chemical group data of combustible gases, the copolymers in the outdoor monitoring area are analyzed to identify and analyze the copolymers. Using gas chromatography-mass spectrometry (GC-MS) or other separation and analysis techniques, the chemical group data are converted into specific copolymer information. During the analysis process, the chemical reaction of the combustible gas and the possibility of its copolymer formation need to be considered. By comparing the known copolymer data, the combustible copolymer data is generated. The final analysis results will reflect the complex chemical structure and its influence of the combustible gas that may exist in the outdoor monitoring area. The absorbance of the spectral data is measured for the spectral absorption peak data of the combustible gases. The spectral intensity at different wavelengths is calibrated using a spectrometer or an absorbance meter, and the concentration of the combustible gas is determined by measuring the absorbance of the combustible gas at a specific wavelength. The Beer-Lambert Law is applied to correlate the absorbance data with the gas concentration. The generated combustible gas spectral absorbance data will provide quantitative information on the gas concentration and be used for further analysis and processing. The combustible gas concentration characteristics of the outdoor monitoring area are analyzed based on the combustible gas spectral absorbance data, and the absorbance data is input into the established concentration calculation model, which can be adjusted based on laboratory calibration data or field test data. The analysis process includes evaluating the changes in gas concentration at different locations and times, and applying statistical methods to calculate the average value, fluctuation range and spatial distribution characteristics of the gas concentration. The generated combustible gas concentration data will help evaluate the gas concentration level and its potential risks in the outdoor monitoring area.Combine the leaked combustible gas concentration data and combustible gas copolymer data to evaluate the gas leakage. Use the gas diffusion model or computational fluid dynamics model to analyze the diffusion pattern of the gas leakage, consider the gas concentration, flow rate and environmental factors, and identify the location, scale and impact range of the leakage source. Combined with the actual monitoring data for calibration, the final generated combustible gas leakage characteristic data will provide comprehensive feedback on the leakage event.

[0091] Preferably, step S3 comprises the following steps:

[0092] Step S31: using a fire hydrant monitor to perform fire hydrant water test operation monitoring processing on the fire hydrant equipment to generate fire hydrant water test operation monitoring data, wherein the fire hydrant water test operation monitoring data includes fire hydrant water test data and fire hydrant pipeline vibration data;

[0093] Step S32: performing cluster analysis on the fire hydrant pipeline vibration data to generate fire hydrant pipeline cluster vibration data;

[0094] Step S33: performing fire hydrant pipeline vibration pattern analysis according to the fire hydrant pipeline cluster vibration data to generate fire hydrant pipeline vibration pattern data;

[0095] Step S34: extracting abnormal vibration patterns of fire hydrant pipelines according to the vibration pattern data of fire hydrant pipelines, and generating abnormal vibration pattern data of fire hydrant pipelines;

[0096] Step S35: Perform water flow-pressure distribution analysis based on the fire hydrant water test data to generate water flow-pressure distribution data;

[0097] Step S36: performing fire hydrant water test pressure anomaly analysis on the water flow-pressure distribution data based on the fluid dynamics model to generate fire hydrant water test pressure anomaly data;

[0098] Step S37: Perform fire hydrant abnormality feature analysis based on the fire hydrant pipeline abnormal vibration pattern data and the fire hydrant water test pressure abnormality data to generate fire hydrant abnormality feature data.

[0099] The present invention utilizes a fire hydrant monitor to monitor and process the fire hydrant water test operation of the fire hydrant equipment, including the fire hydrant water test data and the fire hydrant pipe vibration data, ensuring comprehensive monitoring of the operation status of the fire hydrant and the vibration of the pipe, providing detailed basic data for subsequent analysis, and helping to detect potential problems with the equipment and pipes. Cluster analysis is performed on the fire hydrant pipe vibration data to identify different types of vibration patterns. By classifying the vibration data, structured information is provided for subsequent vibration pattern analysis, which helps to accurately locate abnormal vibration areas. Fire hydrant pipe vibration pattern analysis is performed based on the clustered vibration data of the fire hydrant pipe. By analyzing different vibration patterns, abnormal vibration behaviors in the pipe are identified to help determine whether there are pipe blockages, looseness or other structural problems. Abnormal vibration patterns of fire hydrant pipes are extracted based on the fire hydrant pipe vibration pattern data, and abnormal vibration patterns are accurately extracted, thereby identifying possible faults or abnormal conditions in the pipe and improving the accuracy of fault diagnosis. Water flow-pressure distribution analysis is performed based on the fire hydrant water test data, which provides a detailed distribution of water flow and pressure during the water test of the fire hydrant, which helps to evaluate the working status and performance of the fire hydrant and ensure that the water flow and pressure are within the normal range. An abnormal analysis of the fire hydrant water test pressure is performed on the water flow-pressure distribution data based on the fluid dynamics model. By simulating and analyzing the dynamic relationship between water flow and pressure, the pressure anomalies that may exist during the water test are identified, which helps to timely discover and solve water pressure problems and ensure the reliability of the fire hydrant system. An abnormal feature analysis of fire hydrants is performed based on the abnormal vibration pattern data of fire hydrant pipelines and the abnormal data of fire hydrant water test pressure. The vibration pattern and pressure abnormality data are comprehensively analyzed to comprehensively evaluate the abnormal features of the fire hydrant system, which helps to accurately locate potential equipment failures or performance problems, provide effective maintenance and repair suggestions, and improve the overall safety and operating efficiency of the fire hydrant system.

[0100] In the embodiment of the present invention, the fire hydrant monitor is installed on the designated fire hydrant equipment to ensure that the sensor and data acquisition system can accurately record the operating status of the equipment, and the fire hydrant monitor is intelligently connected with the terminal through the Internet of Things technology, the monitor is started and the fire hydrant water test operation is performed, that is, the fire hydrant is opened for water flow test. During the test, the monitor collects the fire hydrant water test data and the fire hydrant pipe vibration data in real time. The fire hydrant water test data includes parameters such as water flow rate, flow rate and water pressure; the fire hydrant pipe vibration data records the vibration frequency and intensity generated by the pipe during the water flow process, and the generated fire hydrant water test operation monitoring data will be used for subsequent analysis. The fire hydrant pipe vibration data is input into the data analysis tool, and data preprocessing is performed, such as denoising and normalization. Select a suitable data clustering algorithm, such as K-means clustering (K-means), DBSCAN (density clustering) or hierarchical clustering (Hierarchical Clustering), and cluster analysis is performed on the vibration data. The clustering process divides the data into different categories through the algorithm and identifies the similarity of the vibration pattern. The resulting clustered tremor data for fire hydrant pipes will describe the categories and characteristics of different tremor patterns, providing basic data for subsequent analysis. By clustering the tremor data for fire hydrant pipes and applying statistical analysis and pattern recognition techniques to analyze different patterns in the tremor data, the characteristic frequency and amplitude of the tremor data are extracted through spectrum analysis or wavelet transform. By comparing different tremor patterns, their changing patterns and characteristics are determined, and the tremor pattern data for fire hydrant pipes is generated, including detailed descriptions of various tremor patterns, providing a reference for subsequent abnormal pattern recognition. Abnormal patterns that are significantly different from normal tremor patterns are identified from the tremor pattern data for fire hydrant pipes. Pipeline tremor anomaly detection algorithms, such as Isolation Forest or One-Class SVM, are applied, and the model is trained through the tremor patterns of various types of pipes in historical fire hydrants, so that the tremor data can be detected as outliers through the pipeline tremor anomaly detection algorithm. During the analysis process, it is necessary to set appropriate thresholds to distinguish normal and abnormal tremor patterns. The abnormal tremor pattern data of fire hydrant pipes generated, including the natural tremor data of the abnormal tremor pattern of fire hydrant pipes identified, are used to facilitate further fault analysis and maintenance decisions. The flow and water pressure data in the fire hydrant water test data are used to analyze the water flow-pressure distribution of the fire hydrant system. The flow data and valve pressure data of the fire hydrant water test data are used to simulate and analyze the water flow and pressure distribution. The data are input into the simulation software to generate distribution maps of the pressure field and flow velocity field. The final generated water flow-pressure distribution data will include the distribution of water flow and pressure in the fire hydrant pipe, providing a basis for subsequent pressure anomaly analysis.Compare and analyze the water flow-pressure distribution data with the fluid dynamics model to identify abnormal areas in the pressure distribution. Analyze the expected water pressure strength under each pipeline through the fluid dynamics model and pipeline structure. Apply pressure anomaly detection algorithms, such as statistical control charts or threshold detection, to analyze the fluctuation of pressure data, calculate the deviation between the actual pressure data and the expected pressure value, and identify the area and degree of pressure anomaly. The final generated fire hydrant water test pressure anomaly data will include the specific location and degree of abnormal pressure points, providing a basis for subsequent equipment maintenance and adjustment. Combine the fire hydrant pipeline abnormal vibration pattern data and the fire hydrant water test pressure anomaly data, identify the fire hydrant nodes corresponding to the abnormal vibration pattern data of each fire hydrant pipeline and the fire hydrant water test pressure anomaly data, and conduct a comprehensive analysis. Apply multivariate analysis techniques, such as principal component analysis (PCA) or canonical correlation analysis (CCA), to identify the association between abnormal patterns and pressure anomalies. By comparing the characteristics of different abnormal data and identifying potential causes of failures, the generated fire hydrant abnormal characteristic data will include the abnormal vibration characteristics of fire hydrant pipes, the abnormal characteristics of fire hydrant water test pressure, and the combined characteristics of abnormal vibration of fire hydrant pipes and abnormal fire hydrant water test pressure, so as to facilitate the formulation of maintenance and improvement measures.

[0101] Preferably, step S4 comprises the following steps:

[0102] Step S41: Based on the rule engine technology, the circuit abnormality feature data, the combustible gas leakage feature data and the fire hydrant abnormality feature data are processed in a data linkage manner to generate abnormal fire protection linkage data;

[0103] Step S42: performing time series feature analysis on the abnormal fire linkage data based on a preset long short-term memory neural network algorithm to generate abnormal fire linkage time series feature data;

[0104] Step S43: Designing a fire safety prediction strategy gradient model based on the abnormal fire linkage time series feature data to construct a fire safety prediction strategy gradient model;

[0105] Step S44: using the fire safety prediction strategy gradient model to evaluate the fire safety prediction demand of a single abnormal feature on the abnormal fire linkage data to generate fire safety prediction demand data;

[0106] Step S45: Feedback the fire safety prediction demand data to the terminal to execute the fire safety prediction intelligent repair operation.

[0107] The present invention performs data linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data based on rule engine technology, conducts comprehensive analysis on abnormal data from different detection systems, identifies potential comprehensive fire risks through data linkage processing, and provides a comprehensive overview of abnormal situations, which helps to improve the accuracy and response speed of overall fire safety management. Based on the preset long short-term memory neural network (LSTM) algorithm, the abnormal fire linkage data is analyzed for time series features, and the powerful time series data processing capability of the LSTM network is utilized to capture and analyze the time change patterns in the abnormal data, identify potential time trend behaviors, and enhance the ability to predict the development of abnormal events. A strategy gradient model for fire safety prediction is designed based on abnormal fire linkage time series feature data to construct a fire safety prediction strategy gradient model, convert the time series feature data into a fire safety prediction strategy, and provide a scientific basis for formulating effective prevention and response measures, thereby optimizing the fire safety management strategy. Use the fire safety prediction policy gradient model to evaluate the fire safety prediction needs of a single abnormal feature of abnormal fire linkage data. Based on the policy gradient model, separate evaluations are performed on different types of abnormalities to accurately identify and quantify the safety needs of various abnormal features, support the formulation of personalized safety precautions, and improve the accuracy and practicality of prediction. Feedback the fire safety prediction demand data to the terminal to perform fire safety prediction intelligent repair operations, convert the prediction results into actual operation instructions, and use the intelligent system to automatically execute repairs or response measures to ensure rapid response and handling of safety hazards, and improve the automation level and effectiveness of the fire safety system.

[0108] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation step flow chart of step S4 in the embodiment, step S4 in this embodiment includes:

[0109] Step S41: Based on the rule engine technology, the circuit abnormality feature data, the combustible gas leakage feature data and the fire hydrant abnormality feature data are processed in a data linkage manner to generate abnormal fire protection linkage data;

[0110] In an embodiment of the present invention, rule engine technology is used to perform linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data, and the rules of the rule engine (such as IF-THEN rules) are set, including the trigger conditions of circuit abnormality, combustible gas leakage and fire hydrant abnormality and their associated logic, and define how to combine data from different sources together, such as identifying abnormal nodes of circuit leakage and abnormal temperature conditions, whether there will be an association with combustible gas leakage, and whether the fire hydrant can perform fire fighting. The rule engine performs data matching and logical judgment according to the set rules, identifies the correlation between abnormal events, and applies the rule engine to fuse different data to generate comprehensive abnormal fire fighting linkage data, which comprehensively reflects the linkage abnormal status of circuits, combustible gas and fire hydrant equipment.

[0111] Step S42: performing time series feature analysis on the abnormal fire linkage data based on a preset long short-term memory neural network algorithm to generate abnormal fire linkage time series feature data;

[0112] In the embodiment of the present invention, the abnormal fire linkage data is organized into a format suitable for time series analysis to ensure that the timestamp and sequence order of the data are accurate. The LSTM algorithm is used for time series modeling. LSTM can effectively capture the time dependency in the data. When training the LSTM model, appropriate hyperparameters such as learning rate, batch size and sequence length are selected, and the time series results of historical fire linkage data are used for model training. Through model prediction and feature extraction, abnormal fire linkage time series feature data is generated, which contains the patterns and characteristics of abnormal events changing over time, providing a basis for subsequent strategy design.

[0113] Step S43: Designing a fire safety prediction strategy gradient model based on the abnormal fire linkage time series feature data to construct a fire safety prediction strategy gradient model;

[0114] In an embodiment of the present invention, the objectives and strategies of fire safety prediction are defined, including the response and handling strategies for abnormal events. A suitable policy gradient algorithm, such as proximal policy optimization (PPO) or deep deterministic policy gradient (DDPG), is selected to design a policy gradient model. The abnormal fire linkage time series feature data is input into the model for training and optimization. The policy gradient model will generate a fire safety prediction strategy by optimizing the policy function. The strategy can dynamically adjust the response measures according to the time series characteristics of the abnormal data. The fire safety prediction policy gradient model finally constructed will be used to guide the actual safety prediction and response work.

[0115] Step S44: using the fire safety prediction strategy gradient model to evaluate the fire safety prediction demand of a single abnormal feature on the abnormal fire linkage data to generate fire safety prediction demand data;

[0116] In the embodiment of the present invention, a single abnormal feature in the abnormal fire linkage data is extracted as the input data for evaluation. The constructed fire safety prediction strategy gradient model is applied to analyze and evaluate the single abnormal feature. The model will generate corresponding prediction requirements based on the feature data, such as the required warning level, response measures and resource scheduling suggestions. The generated fire safety prediction demand data will include safety requirements and response measures for specific abnormalities, providing decision support for actual operations.

[0117] Step S45: Feedback the fire safety prediction demand data to the terminal to execute the fire safety prediction intelligent repair operation.

[0118] In an embodiment of the present invention, the generated fire safety prediction demand data is transmitted to the terminal control system through a data interface or a communication protocol. After receiving the data, the terminal control system parses the safety prediction demand information therein and starts the corresponding intelligent repair operation. According to the prediction demand data, the system will automatically perform necessary safety operations, such as dispatching fire resources, activating the alarm system, or performing on-site intervention. The terminal control system ensures the effective execution and tracking of the repair operation through real-time monitoring and feedback functions.

[0119] Preferably, step S43 includes the following steps:

[0120] Step S431: performing abnormal fire linkage data trend prediction based on abnormal fire linkage time series feature data to generate abnormal fire linkage prediction data;

[0121] Step S432: Based on the preset fire risk assessment decision, the abnormal fire linkage prediction data is marked with fire risk characteristics to generate abnormal fire linkage mark prediction data;

[0122] Step S433: Design a fire safety prediction strategy gradient model based on the abnormal fire linkage identification prediction data to construct a fire safety prediction strategy gradient model.

[0123] The present invention predicts the trend of abnormal fire linkage data based on the abnormal fire linkage time series feature data. By analyzing the time trend and change pattern in the historical data, it can predict the potential abnormal trends and risks in the future, identify the potential fire risks in advance, provide data support for the formulation of preventive measures, and reduce the possibility of sudden fire incidents. Based on the preset fire risk assessment decision, the abnormal fire linkage prediction data is marked with fire risk characteristics, and the predicted data is compared with the fire risk standard to identify the risk characteristics, so as to determine the potential fire risk level, which helps to clarify the severity of various risks, support the formulation of targeted safety measures and resource allocation, and improve the accuracy of fire prevention and control. The strategy gradient model design for fire safety prediction based on the abnormal fire linkage identification prediction data, by designing the strategy gradient model, the identification prediction data is converted into a specific safety prediction strategy, and the decision-making process is optimized. The model can formulate effective safety strategies based on the risk identification data, and enhance the response capability and management efficiency to different types of fire risks.

[0124] In the embodiment of the present invention, the abnormal fire linkage data trend prediction is performed according to the abnormal fire linkage time series feature data, the prediction results of the abnormal fire linkage time series features in the above steps are extracted, and the data is trend analyzed, that is, the trend prediction results of the normal fire linkage data, and the generated abnormal fire linkage prediction data will contain the prediction results of the development of abnormal situations in the future period of time, providing a basis for subsequent risk assessment and prejudgment. Define the standards and characteristics of fire risk assessment decisions, including the probability of fire occurrence, the scope of influence and the warning level. Compare the abnormal fire linkage prediction data with the preset risk assessment standards, identify the fire risk characteristics in the data, and use a risk identification algorithm, such as a classifier model or a decision tree, to analyze and identify the prediction data. The generated abnormal fire linkage identification prediction data will contain the fire risk level and related characteristics of each prediction data point, providing information support for the formulation of emergency response strategies. Determine the goals and strategies of fire safety prejudgment, including how to respond to different levels of fire risks. Select a suitable policy gradient algorithm, such as proximal policy optimization (PPO) or deep deterministic policy gradient (DDPG), to build a policy gradient model. Use the abnormal fire linkage identification prediction data as model input to perform policy optimization training. The model will adjust the strategy according to the fire risk characteristics in the predicted data and generate prediction strategies for different fire risk levels. Through training and verification, a fire safety prediction strategy gradient model that can be dynamically adjusted according to the prediction data of abnormal fire linkage identification is finally constructed to analyze the urgency of fire safety prevention under various fire abnormal characteristics.

[0125] Preferably, step S433 includes the following steps:

[0126] Step S4331: Designing a strategy function of abnormal fire linkage data and fire safety prediction based on the abnormal fire linkage identification prediction data to generate a fire safety prediction strategy function;

[0127] Step S4332: performing a fire safety prediction strategy return analysis of a single abnormal feature on the fire risk assessment decision of the abnormal fire linkage identification prediction data according to the fire safety prediction strategy function, and generating fire safety prediction strategy return data;

[0128] Step S4333: Perform policy gradient parameter analysis on the fire safety prediction strategy function through the fire safety prediction strategy return data to construct a fire safety prediction strategy gradient model.

[0129] The present invention designs a strategy function for abnormal fire linkage data and fire safety prediction based on abnormal fire linkage identification prediction data. By establishing a strategy function, the prediction data is converted into specific prediction rules and decision criteria. This strategy function can clarify the relationship between abnormal data and fire safety prediction, making the strategy formulation more scientific and systematic, thereby improving the prediction and management capabilities of fire risks. According to the fire safety prediction strategy function, a single abnormal feature fire safety prediction strategy return analysis is performed on the fire risk assessment decision of the abnormal fire linkage identification prediction data. By analyzing the strategy return data, the effectiveness of different fire safety prediction strategies on fire risks is evaluated, and feedback information is provided to optimize the application of the strategy, ensure the actual effect and response capability of the strategy, and further improve the accuracy of the prediction. The fire safety prediction strategy function is analyzed by the fire safety prediction strategy return data to construct a fire safety prediction strategy gradient model, and the fire safety prediction strategy return data is used to optimize the strategy gradient parameters of the strategy function, so that the model can better adapt to the actual situation and improve the prediction strategy. By continuously adjusting the strategy parameters, the prediction accuracy and response capability of the model are improved, ensuring that the fire safety demand level corresponding to the fire risk can be effectively analyzed under different circumstances.

[0130] In the embodiment of the present invention, the abnormal fire linkage identification prediction data is organized into a format suitable for the input and output of the strategy function, ensuring that the data contains all key fire risk features, and defining the objectives of the strategy function, including how to adjust fire safety measures according to different fire risk features. For example, a strategy function is defined to optimize resource allocation, early warning strategy, or emergency response measures, and a machine learning or optimization algorithm, such as linear regression, support vector machine (SVM) or neural network, is used to construct a strategy function for fire safety prediction. The input of the strategy function is the abnormal fire linkage identification prediction data, and the output is the corresponding fire safety prediction priority requirement. The strategy function is trained using the historical abnormal fire linkage identification prediction data so that it can accurately map the mapping relationship between the fire risk characteristics and the fire safety prediction priority requirements to generate a fire safety prediction strategy function. The designed fire safety prediction strategy function is applied to the abnormal fire linkage identification prediction data and the corresponding fire risk assessment decision, and the reward (reward parameter) analysis of the fire safety prediction strategy function is performed according to the fire risk assessment decision. The reward analysis includes evaluating the effect of the prediction strategy, such as the fire probability and fire severity caused by the abnormal fire linkage identification prediction data. Generate fire safety prediction strategy return data, which contains the gap between the actual performance of the strategy and the expected goal, and helps identify the advantages and disadvantages of the strategy. Take the fire safety prediction strategy return data as input, perform policy gradient analysis, and select a suitable policy gradient algorithm, such as stochastic gradient descent (SGD) or its variants, to optimize the parameters of the policy function. Calculate the gradient of the policy function through the return data, and update the parameters of the policy function according to the gradient, gradually optimize the performance of the policy function, and after multiple rounds of training and verification, construct the final fire safety prediction policy gradient model. This model can dynamically adjust the prediction strategy based on the abnormal fire linkage identification prediction data, and is used to analyze the urgency of fire safety prevention under various fire abnormal characteristics.

[0131] This specification provides a smart fire safety prediction linkage management platform for executing the smart fire safety prediction linkage management method as described above. The smart fire safety prediction linkage management platform includes:

[0132] A circuit abnormality feature analysis module is used to collect circuit data of circuit equipment using an electrical appliance monitoring detector to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data, and circuit temperature data; perform circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; perform local temperature abnormality analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature abnormality data; perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data;

[0133] The combustible gas leakage characteristic analysis module is used to use infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; perform combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; perform combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data;

[0134] The fire hydrant abnormal feature module is used to use the fire hydrant monitor to monitor and process the fire hydrant water test operation of the fire hydrant equipment, and generate the fire hydrant water test operation monitoring data; to analyze the abnormal vibration mode of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generate the fire hydrant pipeline abnormal vibration mode data; to analyze the abnormal pressure of the fire hydrant water test according to the fire hydrant water test operation monitoring data, and generate the fire hydrant water test pressure abnormal data; to analyze the abnormal characteristics of the fire hydrant based on the abnormal vibration mode data of the fire hydrant pipeline and the abnormal data of the fire hydrant water test pressure, and generate the fire hydrant abnormal feature data;

[0135] The fire linkage safety intelligent processing module is used to perform data linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data based on rule engine technology to generate abnormal fire linkage data; perform fire safety prediction demand assessment on the abnormal fire linkage data with a single abnormal feature to generate fire safety prediction demand data; and feed back the fire safety prediction demand data to the terminal to execute fire safety prediction intelligent repair operations.

[0136] The beneficial effect of the present application is that the present invention uses electrical monitoring detectors to collect detailed data on circuit equipment, including circuit distribution, current, temperature and other information, so that it can deeply understand the operating status of the circuit system and discover potential circuit problems and safety hazards early; analyze the characteristics of combustible gas leakage through infrared spectroscopy monitoring equipment, timely discover possible gas leakage sources, and reduce fire risks; monitor the water test operation of fire hydrants, analyze pipeline vibration and water pressure anomalies, and ensure the normal function of fire hydrants. Through detailed analysis of circuit current data, including residual current characteristic analysis, current waveform monitoring, high-order harmonic analysis, etc., it is possible to accurately identify potential anomalies in the circuit and prevent the occurrence of electrical fires; through infrared spectroscopy and chemical group analysis, combustible gas leaks in outdoor environments can be accurately identified and quantified to improve the accuracy of fire prevention; through vibration pattern analysis and pressure anomaly detection, it is ensured that fire hydrants can work normally in emergency situations and enhance fire extinguishing capabilities. The rule engine technology is used to link the abnormal circuit, combustible gas leakage and fire hydrant abnormal data to achieve a comprehensive evaluation of multiple abnormal data; the long short-term memory neural network algorithm is used to analyze the time series characteristics of abnormal fire linkage data to improve the ability to predict fire risks; a fire safety prediction strategy gradient model is constructed, and the prediction strategy is optimized through strategy functions and reward analysis to improve the accuracy and efficiency of fire safety warnings. The fire safety prediction demand data generated based on the intelligent model can evaluate and respond to the safety needs of a single abnormal feature in real time, feed back the prediction demand data to the terminal, perform intelligent repair operations, and take timely measures to repair, ensuring the safety and reliability of the fire protection system.

[0137] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0138] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A smart fire safety prediction linkage management method, characterized in that: The following steps are involved: Step S1: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data and circuit temperature data; performing circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; performing local temperature anomaly analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature anomaly data; performing circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature anomaly data to generate circuit abnormality feature data; Step S2: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; Perform combustible gas spectrum characteristic absorption peak analysis on environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; perform combustible gas leakage characteristic analysis in outdoor monitoring areas based on combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data; Step S3: using the fire hydrant monitor to monitor the fire hydrant water test operation of the fire hydrant equipment, and generating fire hydrant water test operation monitoring data; performing abnormal vibration mode analysis of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generating fire hydrant pipeline abnormal vibration mode data; performing abnormal analysis of the fire hydrant water test pressure according to the fire hydrant water test operation monitoring data, and generating fire hydrant water test pressure abnormality data; performing fire hydrant abnormal feature analysis based on the fire hydrant pipeline abnormal vibration mode data and the fire hydrant water test pressure abnormality data, and generating fire hydrant abnormal feature data; Step S4: Based on the rule engine technology, the circuit abnormal feature data, the combustible gas leakage feature data and the fire hydrant abnormal feature data are processed in a data linkage manner to generate abnormal fire linkage data; a fire safety prediction demand assessment of a single abnormal feature is performed on the abnormal fire linkage data to generate fire safety prediction demand data; the fire safety prediction demand data is fed back to the terminal to execute the fire safety prediction intelligent repair operation.

2. The intelligent fire safety prediction linkage management method according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: using an electrical appliance monitoring detector to collect circuit data of a circuit device to generate circuit data; Step S12: performing residual current characteristic analysis according to the circuit current data to generate residual current characteristic data; Step S13: performing circuit residual current distribution identification according to the circuit distribution data and the residual current characteristic data to generate circuit residual current distribution data; Step S14: extracting target circuit temperature data of the circuit residual current distribution from the circuit temperature data according to the circuit residual current distribution data to obtain target circuit temperature data; Step S15: performing spatial temperature distribution analysis on the target circuit temperature data to generate the target circuit spatial temperature distribution data; Step S16: performing temperature gradient analysis according to the target circuit spatial temperature distribution data to generate target circuit temperature gradient data; Step S17: performing local temperature anomaly analysis according to the target circuit temperature gradient data to generate local circuit temperature anomaly data; Step S18: Perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data.

3. The intelligent fire safety prediction linkage management method according to claim 2 is characterized in that: Step S12 includes the following steps: Step S121: Analyze the current difference between input and output according to the circuit current data to generate current difference data; Step S122: monitoring current waveform data according to circuit current data to generate current waveform data; Step S123: performing current fundamental frequency signal analysis according to the current waveform data to generate current fundamental frequency data; Step S124: performing current high-order harmonic characteristic analysis on the current waveform data according to the current fundamental frequency data to generate current high-order harmonic characteristic data; Step S125: performing intermittent residual current data analysis according to the current high-order harmonic characteristic data to generate intermittent residual current data; Step S126: performing residual current characteristic analysis according to the current difference data and the intermittent residual current data to generate residual current characteristic data.

4. The intelligent fire safety prediction linkage management method according to claim 3 is characterized in that: Step S125 includes the following steps: Step S1251: performing current harmonic amplitude characteristic analysis on current high-order harmonic characteristic data to generate current harmonic amplitude characteristic data; Step S1252: performing current harmonic phase characteristic analysis on the current high-order harmonic characteristic data to generate current harmonic phase characteristic data; Step S1253: performing current harmonic distortion characteristic analysis based on the current harmonic amplitude characteristic data and the current harmonic phase characteristic data to generate current harmonic distortion characteristic data; Step S1254: performing intermittent residual current data analysis based on the current harmonic distortion characteristic data to generate intermittent residual current data.

5. The intelligent fire safety prediction linkage management method according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: using infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; Step S22: Analyze the characteristic absorption peaks of the combustible gas spectrum on the environmental infrared spectrum monitoring data to generate the combustible gas spectrum absorption peak data; Step S23: performing a combustible gas chemical group analysis based on the combustible gas spectrum absorption peak data to generate combustible gas chemical group data; Step S24: performing a combustible gas copolymer analysis in the outdoor monitoring area according to the combustible gas chemical group data to generate combustible gas copolymer data; Step S25: performing absorbance analysis on the spectral absorption peak data of the combustible gas to generate spectral absorbance data of the combustible gas; Step S26: performing a characteristic analysis of the combustible gas concentration in the outdoor monitoring area according to the combustible gas spectral absorbance data to generate combustible gas concentration data; Step S27: Based on the combustible gas concentration data and the combustible gas copolymer data, a combustible gas leakage characteristic analysis is performed in the outdoor monitoring area to generate combustible gas leakage characteristic data.

6. The intelligent fire safety prediction linkage management method according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: using a fire hydrant monitor to perform fire hydrant water test operation monitoring processing on the fire hydrant equipment to generate fire hydrant water test operation monitoring data, wherein the fire hydrant water test operation monitoring data includes fire hydrant water test data and fire hydrant pipeline vibration data; Step S32: performing cluster analysis on the fire hydrant pipeline vibration data to generate fire hydrant pipeline cluster vibration data; Step S33: performing fire hydrant pipeline vibration pattern analysis according to the fire hydrant pipeline cluster vibration data to generate fire hydrant pipeline vibration pattern data; Step S34: extracting abnormal vibration patterns of fire hydrant pipelines according to the vibration pattern data of fire hydrant pipelines, and generating abnormal vibration pattern data of fire hydrant pipelines; Step S35: Perform water flow-pressure distribution analysis based on the fire hydrant water test data to generate water flow-pressure distribution data; Step S36: performing fire hydrant water test pressure anomaly analysis on the water flow-pressure distribution data based on the fluid dynamics model to generate fire hydrant water test pressure anomaly data; Step S37: Perform fire hydrant abnormality feature analysis based on the fire hydrant pipeline abnormal vibration pattern data and the fire hydrant water test pressure abnormality data to generate fire hydrant abnormality feature data.

7. The intelligent fire safety prediction linkage management method according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Based on the rule engine technology, the circuit abnormality feature data, the combustible gas leakage feature data and the fire hydrant abnormality feature data are processed in a data linkage manner to generate abnormal fire protection linkage data; Step S42: performing time series feature analysis on the abnormal fire linkage data based on a preset long short-term memory neural network algorithm to generate abnormal fire linkage time series feature data; Step S43: Designing a fire safety prediction strategy gradient model based on the abnormal fire linkage time series feature data to construct a fire safety prediction strategy gradient model; Step S44: using the fire safety prediction strategy gradient model to evaluate the fire safety prediction demand of a single abnormal feature on the abnormal fire linkage data to generate fire safety prediction demand data; Step S45: Feedback the fire safety prediction demand data to the terminal to execute the fire safety prediction intelligent repair operation.

8. The intelligent fire safety prediction linkage management method according to claim 7 is characterized in that: Step S43 includes the following steps: Step S431: performing abnormal fire linkage data trend prediction based on abnormal fire linkage time series feature data to generate abnormal fire linkage prediction data; Step S432: Based on the preset fire risk assessment decision, the abnormal fire linkage prediction data is marked with fire risk characteristics to generate abnormal fire linkage mark prediction data; Step S433: Design a fire safety prediction strategy gradient model based on the abnormal fire linkage identification prediction data to construct a fire safety prediction strategy gradient model.

9. The intelligent fire safety prediction linkage management method according to claim 8 is characterized in that: Step S433 includes the following steps: Step S4331: Designing a strategy function of abnormal fire linkage data and fire safety prediction based on the abnormal fire linkage identification prediction data to generate a fire safety prediction strategy function; Step S4332: performing a fire safety prediction strategy return analysis of a single abnormal feature on the fire risk assessment decision of the abnormal fire linkage identification prediction data according to the fire safety prediction strategy function, and generating fire safety prediction strategy return data; Step S4333: Perform policy gradient parameter analysis on the fire safety prediction strategy function through the fire safety prediction strategy return data to construct a fire safety prediction strategy gradient model.

10. A smart fire safety prediction linkage management platform, characterized in that: Used to execute the intelligent fire safety prediction linkage management method as claimed in claim 1, the intelligent fire safety prediction linkage management platform comprises: A circuit abnormality feature analysis module is used to collect circuit data of circuit equipment using an electrical appliance monitoring detector to generate circuit data, wherein the circuit data includes circuit distribution data, circuit current data, and circuit temperature data; perform circuit residual current distribution analysis based on the circuit distribution data and the circuit current data to generate circuit residual current distribution data; perform local temperature abnormality analysis of the circuit residual current distribution on the circuit temperature data based on the circuit residual current distribution data to generate local circuit temperature abnormality data; perform circuit abnormality feature analysis based on the circuit residual current distribution data and the local circuit temperature abnormality data to generate circuit abnormality feature data; The combustible gas leakage characteristic analysis module is used to use infrared spectrum monitoring equipment to perform environmental infrared spectrum monitoring processing on a preset outdoor monitoring area to generate environmental infrared spectrum monitoring data; perform combustible gas spectrum characteristic absorption peak analysis on the environmental infrared spectrum monitoring data to generate combustible gas spectrum absorption peak data; perform combustible gas leakage characteristic analysis on the outdoor monitoring area based on the combustible gas spectrum absorption peak data to generate combustible gas leakage characteristic data; The fire hydrant abnormal feature module is used to use the fire hydrant monitor to monitor and process the fire hydrant water test operation of the fire hydrant equipment, and generate the fire hydrant water test operation monitoring data; to analyze the abnormal vibration mode of the fire hydrant pipeline according to the fire hydrant water test operation monitoring data, and generate the fire hydrant pipeline abnormal vibration mode data; to analyze the abnormal pressure of the fire hydrant water test according to the fire hydrant water test operation monitoring data, and generate the fire hydrant water test pressure abnormal data; to analyze the abnormal characteristics of the fire hydrant based on the abnormal vibration mode data of the fire hydrant pipeline and the abnormal data of the fire hydrant water test pressure, and generate the fire hydrant abnormal feature data; The fire linkage safety intelligent processing module is used to perform data linkage processing on circuit abnormality feature data, combustible gas leakage feature data and fire hydrant abnormality feature data based on rule engine technology to generate abnormal fire linkage data; perform fire safety prediction demand assessment on the abnormal fire linkage data with a single abnormal feature to generate fire safety prediction demand data; and feed back the fire safety prediction demand data to the terminal to execute fire safety prediction intelligent repair operations.

Images