Fire safety management method and system

Abstract

The invention discloses a fire safety management method, and relates to the technical field of fire safety, and the method comprises the steps: obtaining the threat level of each region according to the multi-source heterogeneous data of each region of a to-be-managed building in a plurality of historical periods, and setting an Internet of Things node; according to the multi-source heterogeneous data of the region of which the threat level is greater than the threat level threshold, in combination with the threat level, obtaining a data monitoring type of the region, the number of Internet of Things monitoring nodes and point distribution; the method comprises the following steps: constructing a multi-source data prediction model, carrying out real-time monitoring on each region according to a multi-source heterogeneous data time sequence collected by Internet of Things nodes arranged in each region in real time, and timely deploying fire-fighting resources, enhancing patrol frequency or taking other prevention measures according to a monitoring result so as to reduce the possibility of fire occurrence and guarantee the safety of buildings and personnel. And the response speed and the management effect of fire safety management are effectively improved.

Description

Technical Field

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

[0002] The prior art CN117408446A "An intelligent fire safety management method and system" establishes a BIM model, and then uses the BIM model in combination with the fire emergency evacuation plan to perform fire simulation and personnel evacuation simulation, improves the fire emergency evacuation plan through the simulation results, and provides information support for fire command and rescue work, so as to carry out evacuation in an orderly manner; The prior art CN117495311A "A regional fire safety physical examination method" collects fire safety management data in the target area, and obtains a fire spread model and fire data based on the fire safety management data in the target area. The fire data includes building data, personnel evacuation data and fire data. Based on the building data and personnel evacuation data, the emergency time safety factor when a fire occurs in the target area is obtained. Based on the fire data, the fire safety factor when a fire occurs in the target area is obtained, thereby obtaining a comprehensive fire safety evaluation value of the target area, thereby judging whether the fire safety physical examination of the target area is qualified.

[0003] At present, some solutions attempt to improve the intelligence level of fire management by integrating video surveillance and simple data analysis functions. These systems can achieve real-time monitoring and basic risk warning to a certain extent, but they usually lack the ability to deeply analyze multiple risk factors in complex environments. In addition, these systems often rely on a single type of data source, such as video or temperature data, ignoring the importance of multimodal data in fire management, resulting in incomplete risk assessment and inaccurate response strategies. A single data source and simple data analysis methods cannot fully understand and predict fire risks in complex environments, resulting in inaccurate warning information and untimely response measures. In addition, existing technologies lack optimization considerations for fire resource allocation and cannot achieve efficient use of resources, thus affecting the efficiency and effectiveness of overall fire management. Summary of the invention

[0004] In order to solve the above technical problems, the object of the present invention is to provide a fire safety management method, comprising the following steps: Step s1: Based on the multi-source heterogeneous data of various areas of the building to be managed in several historical periods, the threat level of each area is obtained and the IoT node is set; Step s2: obtaining the data monitoring type, the number of IoT monitoring nodes, and the point distribution of the area according to the multi-source heterogeneous data of the area whose threat level is greater than the threat level threshold and the threat level; Step s3: Build a multi-source data prediction model to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area.

[0005] Furthermore, based on the multi-source heterogeneous data of various areas of the building to be managed in several historical periods, the threat level of each area is obtained, and the process of setting up the IoT node includes: Acquire multi-source heterogeneous data in several historical periods of each area of ​​the building to be managed, wherein the multi-source heterogeneous data includes building structure data, environmental data, personnel dynamic data, fire equipment status data, alarm and response data, and historical fire data; use the multi-source heterogeneous data in several historical periods of each area as evaluation indicators, set the indicator weight matrix and threat level of the evaluation indicators, and judge the membership matrix of each area to the threat level through fuzzy comprehensive evaluation; The threat level of each area is obtained according to the membership matrix and the indicator weight matrix, and the threat level threshold is preset. The Internet of Things monitoring nodes are set in the areas where the threat level is greater than the threat level threshold.

[0006] Further, the process of obtaining the data monitoring type, the number of IoT monitoring nodes, and the point distribution of the area according to the multi-source heterogeneous data of the area with a threat level greater than the threat level threshold and the threat level includes: According to the historical fire data of the area, several fire types and several fire type location features of sudden fires in the area are obtained, and at the same time, coupling coefficients between several fire types and several data types are obtained, and the several data types include environmental data, personnel dynamic data, fire equipment status data, alarm and response data. A coupling coefficient threshold is preset, and data types with coupling coefficients greater than the coupling coefficient threshold with several fire types are marked as associated data types. According to the associated data types of several fire types, data monitoring types of location features of several fire types are obtained, and the corresponding Internet of Things monitoring node types are determined according to the data monitoring types, and the number of each type of Internet of Things monitoring nodes in the area is determined according to the threat level; According to the maximum coverage range of each type of IoT monitoring node and the preset range of fire type location characteristics, the distribution coverage rate of each type of IoT monitoring node is obtained, and the fitness function of the particle swarm algorithm is obtained according to the distribution coverage rate of each type of IoT monitoring node. At the same time, the adaptive dynamic weight that changes with the number of iterations is introduced to improve the particle swarm algorithm. The optimal layout position of each type of IoT monitoring node is obtained through the improved particle swarm algorithm.

[0007] Furthermore, the process of obtaining coupling coefficients between the plurality of fire types and the plurality of data types includes: Obtain an average standard deviation of a certain key data type of a certain fire type, a Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and a Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type; The average standard deviation of a certain key data type of the certain fire type, the difference in Pearson correlation coefficient between the certain key data type and other key data types of the certain fire type, and the difference in Pearson correlation coefficient between the certain key data type and other non-key data types of the certain fire type are used as evaluation indicators, and the indicator weight matrix and coupling coefficient interval of the evaluation indicator are set. The membership matrix between the certain key data type and the certain fire type for different coupling coefficient intervals is judged through fuzzy comprehensive evaluation, and the coupling coefficient between the certain key data type and the certain fire type is obtained according to the membership matrix and the indicator weight matrix. Similarly, the coupling coefficients between several fire types and several data types are obtained.

[0008] Furthermore, the process of obtaining the average standard deviation of a certain key data type of a certain fire type, the Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and the Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type includes: Obtain the multi-source heterogeneous data time series sequence when several fire types did not occur in several historical periods in the region and the corresponding multi-source heterogeneous data time series sequence when several fire types occurred, obtain the value range of several data types extracted from the multi-source heterogeneous data time series sequence when several fire types did not occur, and extract the value range of several data types from the multi-source heterogeneous data time series sequence when several fire types occurred; Compare the value ranges of several data types when several fire types occur with the value ranges of several data types when several fire types do not occur; if the value range of a certain data type when a certain fire type occurs is inconsistent with the value range when the certain fire type does not occur, mark the data type as a key data type of the certain data type; if the value range of a certain data type when a certain fire type occurs is consistent with the value range when the certain fire type does not occur, mark the data type as a non-key data type of the certain data type; and so on, obtain several key data types and several non-key data types of the several fire types; Obtain the average standard deviation difference between the average standard deviation of a key data type of a certain fire type when the certain fire type occurs and the average standard deviation when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other non-key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur.

[0009] Furthermore, in the process of obtaining the threat level of each region, the process of setting the indicator weight matrix of multi-source heterogeneous data in several historical periods of each region includes: A normal threshold interval corresponding to each data type in the multi-source heterogeneous data of each region is preset, and the frequency of each data type value in each region not being located in the corresponding normal threshold interval and the average fluctuation amplitude of each data type value are obtained according to the multi-source heterogeneous data time series in several historical periods of each region. The stability coefficient of each data type in each region is obtained according to the frequency and the average fluctuation amplitude, and the number of occurrences of each fire type in each region is obtained at the same time. The heat coefficient of each data type in each region is obtained according to the coupling coefficient between each data type in each region and each fire type and the number of occurrences of each fire type. The weight coefficient of each data type in each region is obtained according to the stability coefficient and the heat coefficient, and an indicator weight matrix is ​​constructed according to the weight coefficient of each data type in each region.

[0010] Furthermore, a multi-source data prediction model is constructed, and the process of real-time monitoring of each area based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area includes: Construct a multi-source data prediction model based on deep learning, use multi-source heterogeneous data in several historical periods of each region as a test set and a training set, input the training set into the multi-source data prediction model for training until the loss function training is stable, save the model parameters, test the multi-source data prediction model through the test set until it meets the preset requirements, and output the multi-source data prediction model; The multi-source heterogeneous data time series of each area are collected in real time through the Internet of Things nodes deployed in each area, the multi-source heterogeneous data time series are input into the multi-source data prediction model, and the estimated multi-source heterogeneous data time series in the current period are output according to the multi-source data prediction model; The multi-source heterogeneous data time series sequences of each area collected in real time and the estimated multi-source heterogeneous data time series sequences are compared with the normal threshold intervals corresponding to each data type in the multi-source heterogeneous data. If there is a data type whose corresponding value is not within the corresponding normal threshold interval, the alarm information of the data type is generated.

[0011] A fire safety management system comprises a monitoring center, wherein the monitoring center is communicatively connected with a data analysis module, a data processing module and a data monitoring module; The data analysis module is used to obtain the threat level of each area and set the Internet of Things node based on the multi-source heterogeneous data of each area of ​​the building to be managed in several historical periods; The data processing module is used to obtain the data monitoring type, the number of IoT monitoring nodes and the point distribution of the area according to the multi-source heterogeneous data of the area with a threat level greater than the threat level threshold in combination with the threat level; The data monitoring module is used to build a multi-source data prediction model, and to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the Internet of Things nodes deployed in each area.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: according to the multi-source heterogeneous data of each area of ​​the building to be managed in several historical periods, the threat level of each area is obtained and the Internet of Things node is set; according to the multi-source heterogeneous data of the area whose threat level is greater than the threat level threshold, the data monitoring type of the area, the number of Internet of Things monitoring nodes and the point distribution are obtained in combination with the threat level; a multi-source data prediction model is constructed, and each area is monitored in real time according to the multi-source heterogeneous data time series sequence collected in real time by the Internet of Things nodes deployed in each area; according to the monitoring results, fire-fighting resources are deployed in time, the inspection frequency is increased or other preventive measures are taken to reduce the possibility of fire, ensure the safety of buildings and personnel, effectively improve the response speed and management effect of fire safety management, minimize the risk of fire and protect the safety of buildings and users. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a schematic diagram of a fire safety management method according to an embodiment of the present application.

[0014] Figure 2 This is a schematic diagram of a fire safety management system according to an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0016] like Figure 1 As shown, a fire safety management method comprises the following steps: Step s1: Based on the multi-source heterogeneous data of various areas of the building to be managed in several historical periods, the threat level of each area is obtained and the IoT node is set; Step s2: obtaining the data monitoring type, the number of IoT monitoring nodes, and the point distribution of the area according to the multi-source heterogeneous data of the area whose threat level is greater than the threat level threshold and the threat level; Step s3: Build a multi-source data prediction model to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area.

[0017] It should be further explained that, in the specific implementation process, according to the multi-source heterogeneous data of each area of ​​the building to be managed in several historical periods, the threat level of each area is obtained, and the process of setting up the IoT node includes: Acquire multi-source heterogeneous data in several historical periods of each area of ​​the building to be managed, wherein the multi-source heterogeneous data includes building structure data, environmental data, personnel dynamic data, fire equipment status data, alarm and response data, and historical fire data; use the multi-source heterogeneous data in several historical periods of each area as evaluation indicators, set the indicator weight matrix and threat level of the evaluation indicators, and judge the membership matrix of each area to the threat level through fuzzy comprehensive evaluation; The threat level of each area is obtained according to the membership matrix and the indicator weight matrix, and the threat level threshold is preset. The Internet of Things monitoring nodes are set in the areas where the threat level is greater than the threat level threshold.

[0018] It should be further explained that, in the specific implementation process, the process of obtaining the threat level of each area according to the membership matrix and the indicator weight matrix includes: The indicator weight matrix and the membership matrix of the evaluation indicators are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicators, the membership of each area for different threat levels is obtained according to the fuzzy comprehensive evaluation matrix, the threat level with the highest membership corresponding to each area is screened out, and the threat level with the highest membership corresponding to each area is used as the threat level of each area; Wherein, the formula is: in, is the fuzzy comprehensive evaluation matrix of the evaluation index, is the indicator weight matrix of the evaluation indicator, is the membership matrix, The weight matrix representing the evaluation index is multiplied by the elements at the corresponding positions of the membership matrix. and It is a weighted parameter for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0019] It should be further explained that, in the specific implementation process, the types of data included in environmental data are: Temperature and humidity: Ambient temperature and humidity have a direct impact on the occurrence and spread of fire, so they need to be monitored and recorded in real time; Gas concentration: Monitor key gases (such as CO, , smoke, etc.) for early fire identification; Meteorological conditions: including wind speed, wind direction and other information, which affect the spread of fire and the safety of firefighting operations; The types of data included in building structure data are: Building floor plan: including floor layout, passage location and dimensions, etc., for planning evacuation routes and fire escapes; Building materials and flame retardant grades: the combustion properties of different materials directly affect the speed and difficulty of fire spread; The data types of fire equipment status data include: Location and quantity of fire-fighting equipment: including the location distribution and quantity of fire extinguishers, fire hydrants, automatic sprinkler systems, etc.; Equipment operating status: real-time monitoring of the working status of fire-fighting equipment, such as the water supply pressure of the sprinkler system, the filling status of the fire extinguisher, etc.; The data types of personnel dynamic data include: Density and distribution of people: monitor the density and distribution of people in the building for evacuation safety and rescue dispatch; Dynamic camera monitoring: use surveillance cameras to obtain real-time scene information to help firefighters understand the fire situation and personnel status in real time; Alarm and response data include the following data types: Fire alarm status: real-time monitoring of fire alarm status and trigger information; Fire brigade dispatch data: fire brigade dispatch status and response time, used to evaluate fire response efficiency; The types of fire history data include: Historical fire records: Analyze past fire cases to understand the main causes and spread of fires, and guide the formulation of prevention and response strategies.

[0020] It should be further explained that, in the specific implementation process, the process of obtaining the data monitoring type, the number of IoT monitoring nodes and the point distribution of the area according to the multi-source heterogeneous data of the area with a threat level greater than the threat level threshold and the threat level includes: According to the historical fire data of the area, several fire types and several fire type location features of sudden fires in the area are obtained, and the coupling coefficients between several fire types and several data types are obtained at the same time. The several data types include environmental data, personnel dynamic data, fire equipment status data, alarm and response data, and a coupling coefficient threshold is preset. The data types with coupling coefficients greater than the coupling coefficient threshold are marked as associated data types. According to the associated data types of several fire types, the data monitoring types of the location features of several fire types are obtained, and the corresponding Internet of Things monitoring node types are determined according to the data monitoring types. The Internet of Things monitoring nodes include the following specific devices and functions: intelligent smoke detectors, temperature and humidity sensors, electrical fire early warning devices, water pressure sensors: used to monitor the water pressure of fire hydrants or automatic sprinkler systems to ensure that sufficient water source pressure can be provided during fire extinguishing, gas leak detectors, video surveillance cameras, fire equipment status monitoring devices: monitor the working status and use of equipment such as fire extinguishers, fire hydrants, and automatic sprinkler systems to ensure that they can be used normally when needed, intelligent access control systems: management systems that control the entry and exit of personnel in buildings, which can help ensure the safe evacuation of personnel during fires, and determine the number of various types of Internet of Things monitoring nodes in the area according to the threat level; According to the maximum coverage range of each type of IoT monitoring node and the preset range of fire type location characteristics, the distribution coverage rate of each type of IoT monitoring node is obtained, and the fitness function of the particle swarm algorithm is obtained according to the distribution coverage rate of each type of IoT monitoring node. At the same time, the adaptive dynamic weight that changes with the number of iterations is introduced to improve the particle swarm algorithm. The optimal layout position of each type of IoT monitoring node is obtained through the improved particle swarm algorithm.

[0021] It should be further explained that, in the specific implementation process, the calculation formula for obtaining the distribution coverage of each type of IoT monitoring node according to the maximum coverage range of each type of IoT monitoring node and the preset range of the fire type location characteristics is: Among them, q represents the distribution coverage of IoT monitoring node a, Indicates the maximum range covered by IoT monitoring node a, The distance between IoT monitoring node a and fire type location feature b, Indicates the preset range of fire type location feature b, represents the conversion coefficient. When the range covered by the IoT monitoring node a is outside the preset range of the location feature b, the distribution coverage rate is 0. Aggregate the distribution coverage rates of each IoT monitoring node to obtain the fitness function. The particle swarm algorithm is improved by introducing an adaptive dynamic weight that changes with the number of iterations. The particle swarm algorithm expands the search space, allowing particles to jump out of the currently searched optimal position, thereby searching in a larger space, which is conducive to particle search for a better solution. The adaptive dynamic weight reflects the proportion of the particle inheriting the previous speed, changes with the number of particle iterations, optimizes the particle search, and performs weighted average processing on the adaptive dynamic weights of different iteration numbers. The adaptive dynamic weight calculation formula is:

[0022] Where N represents the total number of iterations, represents the adaptive dynamic weight after the nth iteration; Represents the weight coefficient corresponding to the nth iteration; Represents the bias coefficient.

[0023] It should be further explained that, in the specific implementation process, the process of obtaining the coupling coefficients between several fire types and several data types includes: Obtain an average standard deviation of a certain key data type of a certain fire type, a Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and a Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type; The average standard deviation of a certain key data type of the certain fire type, the difference in Pearson correlation coefficient between the certain key data type and other key data types of the certain fire type, and the difference in Pearson correlation coefficient between the certain key data type and other non-key data types of the certain fire type are used as evaluation indicators, and the indicator weight matrix and coupling coefficient interval of the evaluation indicator are set. The membership matrix between the certain key data type and the certain fire type for different coupling coefficient intervals is judged through fuzzy comprehensive evaluation, and the coupling coefficient between the certain key data type and the certain fire type is obtained according to the membership matrix and the indicator weight matrix. Similarly, the coupling coefficients between several fire types and several data types are obtained.

[0024] It should be further explained that in the specific implementation process, several fire types include: Electrical fire: Most equipment and facilities in a building rely on electricity, so electrical fires are a common type of fire. They may be caused by aging wires, short circuits, overloads, or electrical equipment failures. Kitchen fires: Fires caused by careless use of cooking grease or malfunctioning kitchen equipment, especially in catering establishments and homes; Chemical fires: Fires caused by leakage or improper operation of chemicals such as solvents and medicines stored and used in buildings such as hospitals, laboratories, and factories; Building materials and decoration fires: Fires caused by welding, open flame operation, etc. during construction at construction sites or in newly renovated buildings; Fires caused by heating equipment and electrical equipment: such as fires caused by prolonged use or equipment failure of heaters, electric kettles, electric blankets, etc. Fires caused by accumulation of paper materials and combustibles: such as large amounts of paper, documents or other combustible materials piled up in offices or warehouses; Elevator fires: Although less common, elevator fires can also occur due to malfunctioning elevator equipment or electrical problems; Man-made fires: such as careless smoking, playing with fire or intentional arson.

[0025] It should be further explained that, in a specific implementation process, the process of obtaining the average standard deviation of a certain key data type of a certain fire type, the Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and the Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type includes: Obtain the multi-source heterogeneous data time series sequence when several fire types did not occur in several historical periods in the region and the corresponding multi-source heterogeneous data time series sequence when several fire types occurred, obtain the value range of several data types extracted from the multi-source heterogeneous data time series sequence when several fire types did not occur, and extract the value range of several data types from the multi-source heterogeneous data time series sequence when several fire types occurred; Compare the value ranges of several data types when several fire types occur with the value ranges of several data types when several fire types do not occur; if the value range of a certain data type when a certain fire type occurs is inconsistent with the value range when the certain fire type does not occur, mark the data type as a key data type of the certain data type; if the value range of a certain data type when a certain fire type occurs is consistent with the value range when the certain fire type does not occur, mark the data type as a non-key data type of the certain data type; and so on, obtain several key data types and several non-key data types of the several fire types; Obtain the average standard deviation difference between the average standard deviation of a key data type of a certain fire type when the certain fire type occurs and the average standard deviation when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other non-key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur.

[0026] It should be further explained that, in the specific implementation process, the calculation formula for obtaining the Pearson correlation coefficient between data types is: in, Represents the Pearson correlation coefficient between data type i and data type j. Indicates the value of data type i at time t, Indicates the average value size of data type i, Indicates the value of data type j at time t, It represents the average value of data type j, and k represents the total number of moments. It should be noted that the significance of obtaining the average standard deviation of a key data type of a certain fire type when the fire type occurs and the average standard deviation difference between the average standard deviation when the fire type does not occur, and obtaining the Pearson correlation coefficient between a key data type and other key data types of the fire type when the fire type occurs is that the data type when the numerical range changes before and after the fire type occurs is not necessarily the key data type for the fire type. For example, the smoke concentration changes before and after the fire occurs, but since the smoke concentration must change before and after the fire occurs, it is impossible to analyze whether the data type is one of the causes of the fire by the numerical change of the smoke concentration. Moreover, the numerical change is a static analysis method, which fails to reflect the advantages of the online fire safety management system. Therefore, the present invention determines the connection relationship between the fire type and the data type by analyzing the nonlinear change trend of each data type before and after the fire type occurs. In the present invention, the nonlinear change trend of each data type before and after the fire type occurs is determined and analyzed by obtaining the average standard deviation of the key data type, the difference of the Pearson correlation coefficient between the key data type and other key data types, and the difference of the Pearson correlation coefficient between the key data type and other non-key data types, so as to ensure the accuracy of the coupling coefficients between several fire types and several data types.

[0027] It should be further explained that, in the specific implementation process, in the process of obtaining the threat level of each area, the process of setting the indicator weight matrix of multi-source heterogeneous data in several historical periods of each area includes: A normal threshold interval corresponding to each data type in the multi-source heterogeneous data of each region is preset, and the frequency of each data type value in each region not being located in the corresponding normal threshold interval and the average fluctuation amplitude of each data type value are obtained according to the multi-source heterogeneous data time series in several historical periods of each region. The stability coefficient of each data type in each region is obtained according to the frequency and the average fluctuation amplitude, and the number of occurrences of each fire type in each region is obtained at the same time. The heat coefficient of each data type in each region is obtained according to the coupling coefficient between each data type in each region and each fire type and the number of occurrences of each fire type. The weight coefficient of each data type in each region is obtained according to the stability coefficient and the heat coefficient, and an indicator weight matrix is ​​constructed according to the weight coefficient of each data type in each region.

[0028] It should be further explained that, in the specific implementation process, the calculation formula for obtaining the stability coefficient based on the frequency and the average fluctuation amplitude is: in, represents the stability coefficient, rf represents the frequency, represents the average fluctuation range, Represents the conversion coefficient; the calculation formula for obtaining the heat coefficient of the data type based on the coupling coefficient between the data type and the fire type and the number of occurrences of the fire type is: in, Indicates the heat coefficient of data type y, Represents the coupling coefficient between data type y and fire type z, represents the number of occurrences of fire type z, Indicates the total number of fire types; The calculation formula for obtaining the weight coefficient of the data type according to the stability coefficient and the heat coefficient is: in, Represents the weight coefficient of data type y, Indicates the stability coefficient of data type y, It should be further explained that, in the specific implementation process, the process of constructing a multi-source data prediction model and monitoring each area in real time based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area includes: Construct a multi-source data prediction model based on deep learning, use multi-source heterogeneous data in several historical periods of each region as a test set and a training set, input the training set into the multi-source data prediction model for training until the loss function training is stable, save the model parameters, test the multi-source data prediction model through the test set until it meets the preset requirements, and output the multi-source data prediction model; The multi-source heterogeneous data time series of each area are collected in real time through the Internet of Things nodes deployed in each area, the multi-source heterogeneous data time series are input into the multi-source data prediction model, and the estimated multi-source heterogeneous data time series in the current period are output according to the multi-source data prediction model; The multi-source heterogeneous data time series sequences of each area collected in real time and the estimated multi-source heterogeneous data time series sequences are compared with the normal threshold intervals corresponding to each data type in the multi-source heterogeneous data. If there is a data type whose corresponding value is not within the corresponding normal threshold interval, the alarm information of the data type is generated.

[0029] like Figure 2 As shown, a fire safety management system includes a monitoring center, wherein the monitoring center is communicatively connected with a data analysis module, a data processing module and a data monitoring module; The data analysis module is used to obtain the threat level of each area and set the Internet of Things node based on the multi-source heterogeneous data of each area of ​​the building to be managed in several historical periods; The data processing module is used to obtain the data monitoring type, the number of IoT monitoring nodes and the point distribution of the area according to the multi-source heterogeneous data of the area with a threat level greater than the threat level threshold in combination with the threat level; The data monitoring module is used to build a multi-source data prediction model, and to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the Internet of Things nodes deployed in each area.

[0030] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A fire safety management method, characterized in that: The following steps are involved: Step s1: Based on the multi-source heterogeneous data of various areas of the building to be managed in several historical periods, the threat level of each area is obtained and the IoT node is set; Step s2: obtaining the data monitoring type, the number of IoT monitoring nodes, and the point distribution of the area according to the multi-source heterogeneous data of the area whose threat level is greater than the threat level threshold and the threat level; Step s3: Build a multi-source data prediction model to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area.

2. A fire safety management method according to claim 1, characterized in that: Based on the multi-source heterogeneous data of various areas of the building to be managed in several historical periods, the threat level of each area is obtained. The process of setting up the IoT node includes: Acquire multi-source heterogeneous data in several historical periods of each area of ​​the building to be managed, wherein the multi-source heterogeneous data includes building structure data, environmental data, personnel dynamic data, fire equipment status data, alarm and response data, and historical fire data; use the multi-source heterogeneous data in several historical periods of each area as evaluation indicators, set the indicator weight matrix and threat level of the evaluation indicators, and judge the membership matrix of each area to the threat level through fuzzy comprehensive evaluation; The threat level of each area is obtained according to the membership matrix and the indicator weight matrix, and the threat level threshold is preset. The Internet of Things monitoring nodes are set in the areas where the threat level is greater than the threat level threshold.

3. A fire safety management method according to claim 2, characterized in that: The process of obtaining the data monitoring type, the number of IoT monitoring nodes, and the point distribution of the area according to the multi-source heterogeneous data of the area whose threat level is greater than the threat level threshold and the threat level is combined with the threat level includes: According to the historical fire data of the area, several fire types and several fire type location features of sudden fires in the area are obtained, and at the same time, coupling coefficients between several fire types and several data types are obtained, and the several data types include environmental data, personnel dynamic data, fire equipment status data, alarm and response data. A coupling coefficient threshold is preset, and data types with coupling coefficients with several fire types greater than the coupling coefficient threshold are marked as associated data types. According to the associated data types of several fire types, data monitoring types of location features of several fire types are obtained, and the corresponding Internet of Things monitoring node types are determined according to the data monitoring types, and the number of each type of Internet of Things monitoring nodes in the area is determined according to the threat level; According to the maximum coverage range of each type of IoT monitoring node and the preset range of fire type location characteristics, the distribution coverage rate of each type of IoT monitoring node is obtained, and the fitness function of the particle swarm algorithm is obtained according to the distribution coverage rate of each type of IoT monitoring node. At the same time, the adaptive dynamic weight that changes with the number of iterations is introduced to improve the particle swarm algorithm. The optimal layout position of each type of IoT monitoring node is obtained through the improved particle swarm algorithm.

4. A fire safety management method according to claim 3, characterized in that: The process of obtaining the coupling coefficients between several fire types and several data types includes: Obtain an average standard deviation of a certain key data type of a certain fire type, a Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and a Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type; The average standard deviation of a certain key data type of the certain fire type, the difference in Pearson correlation coefficient between the certain key data type and other key data types of the certain fire type, and the difference in Pearson correlation coefficient between the certain key data type and other non-key data types of the certain fire type are used as evaluation indicators, and the indicator weight matrix and coupling coefficient interval of the evaluation indicator are set. The membership matrix between the certain key data type and the certain fire type for different coupling coefficient intervals is judged through fuzzy comprehensive evaluation, and the coupling coefficient between the certain key data type and the certain fire type is obtained according to the membership matrix and the indicator weight matrix. Similarly, the coupling coefficients between several fire types and several data types are obtained.

5. A fire safety management method according to claim 4, characterized in that: The process of obtaining the average standard deviation of a certain key data type of a certain fire type, the Pearson correlation coefficient difference between the certain key data type and other key data types of the certain fire type, and the Pearson correlation coefficient difference between the certain key data type and other non-key data types of the certain fire type includes: Obtain the multi-source heterogeneous data time series sequence when several fire types did not occur in several historical periods in the region and the corresponding multi-source heterogeneous data time series sequence when several fire types occurred, obtain the value range of several data types extracted from the multi-source heterogeneous data time series sequence when several fire types did not occur, and extract the value range of several data types from the multi-source heterogeneous data time series sequence when several fire types occurred; Compare the value ranges of several data types when several fire types occur with the value ranges of several data types when several fire types do not occur; if the value range of a certain data type when a certain fire type occurs is inconsistent with the value range when the certain fire type does not occur, mark the data type as a key data type of the certain data type; if the value range of a certain data type when a certain fire type occurs is consistent with the value range when the certain fire type does not occur, mark the data type as a non-key data type of the certain data type; and so on, obtain several key data types and several non-key data types of the several fire types; Obtain the average standard deviation difference between the average standard deviation of a key data type of a certain fire type when the certain fire type occurs and the average standard deviation when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur. At the same time, obtain the difference in Pearson correlation coefficients between a key data type and other non-key data types of the certain fire type when the certain fire type occurs and the Pearson correlation coefficients when the certain fire type does not occur.

6. A fire safety management method according to claim 5, characterized in that: In the process of obtaining the threat level of each area, the process of setting the indicator weight matrix of multi-source heterogeneous data in several historical periods of each area includes: A normal threshold interval corresponding to each data type in the multi-source heterogeneous data of each region is preset, and the frequency of each data type value in each region not being located in the corresponding normal threshold interval and the average fluctuation amplitude of each data type value are obtained according to the multi-source heterogeneous data time series in several historical periods of each region. The stability coefficient of each data type in each region is obtained according to the frequency and the average fluctuation amplitude, and the number of occurrences of each fire type in each region is obtained at the same time. The heat coefficient of each data type in each region is obtained according to the coupling coefficient between each data type in each region and each fire type and the number of occurrences of each fire type. The weight coefficient of each data type in each region is obtained according to the stability coefficient and the heat coefficient, and an indicator weight matrix is ​​constructed according to the weight coefficient of each data type in each region.

7. A fire safety management method according to claim 6, characterized in that: The process of building a multi-source data prediction model and monitoring each area in real time based on the multi-source heterogeneous data time series collected in real time by the IoT nodes deployed in each area includes: Construct a multi-source data prediction model based on deep learning, use multi-source heterogeneous data in several historical periods of each region as a test set and a training set, input the training set into the multi-source data prediction model for training until the loss function training is stable, save the model parameters, test the multi-source data prediction model through the test set until it meets the preset requirements, and output the multi-source data prediction model; The multi-source heterogeneous data time series of each area are collected in real time through the Internet of Things nodes deployed in each area, the multi-source heterogeneous data time series are input into the multi-source data prediction model, and the estimated multi-source heterogeneous data time series in the current period are output according to the multi-source data prediction model; The multi-source heterogeneous data time series sequences of each region collected in real time and the estimated multi-source heterogeneous data time series sequences are compared with the normal threshold intervals corresponding to each data type in the multi-source heterogeneous data. If there is a data type whose corresponding value is not within the corresponding normal threshold interval, the alarm information of the data type is generated.

8. A fire safety management system, specifically applied to a fire safety management method according to any one of claims 1 to 7, characterized in that: It includes a monitoring center, wherein the monitoring center is communicatively connected with a data analysis module, a data processing module and a data monitoring module; The data analysis module is used to obtain the threat level of each area and set the Internet of Things node based on the multi-source heterogeneous data of each area of ​​the building to be managed in several historical periods; The data processing module is used to obtain the data monitoring type, the number of IoT monitoring nodes and the point distribution of the area according to the multi-source heterogeneous data of the area with a threat level greater than the threat level threshold in combination with the threat level; The data monitoring module is used to build a multi-source data prediction model, and to monitor each area in real time based on the multi-source heterogeneous data time series collected in real time by the Internet of Things nodes deployed in each area.

Images