Fire data processing method, system and storage medium based on temperature and smoke sensing

Abstract

The present invention provides a method, system and storage medium for processing fire data based on temperature and smoke sensing, which relates to the field of data processing technology. The method includes collecting temperature data and smoke concentration data of each monitoring point to construct corresponding temperature and smoke time series data; calibrating the monitoring point that triggers the smoke sensor as a suspected fire trigger point, identifying the concentration fragment that rises rapidly within a set time window, synchronously extracting the temperature time series fragment of the corresponding time period, and generating a temperature and smoke response difference mapping pair for each suspected trigger point; extracting the smoke concentration rising rate and the corresponding temperature delay response parameter, and marking them as asynchronous early trigger points; traversing all asynchronous early trigger points and clustering them according to their spatial topological positions to construct a suspected fire gathering area, constructing a regional response parameter set, and executing multi-conditional reasoning based on combined threshold judgment and logical rule chain to determine whether to trigger a regional fire warning instruction.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, system and storage medium for processing fire data based on temperature and smoke sensing. Background Art

[0002] Existing fire monitoring systems mostly rely on single sensor threshold triggering or simple time-series correlation judgment, which makes it difficult to effectively distinguish between real fires and local interference. They also lack the ability to analyze the asynchronous smoke and temperature responses of early fires, and usually ignore the coupling characteristics of the spatial topological relationship of sensors and physical responses.

[0003] In the prior art, publication number CN118467897A discloses a data processing method, namely, collecting smoke sensor data and temperature sensor data through equipment, performing fuzzy operations on them, generating smoke sensor factors and temperature sensor factors, performing clarification operations, determining multiple smoke sensor membership parameters and multiple temperature sensor membership parameters, and generating target detection results through the smoke sensor membership parameters and temperature sensor membership parameters; however, this method does not take into account the spatial topological position of the data acquisition equipment and the relationship mapping relationship between temperature and smoke sensor when the fire occurs, and lacks further consideration of the suspected fire cluster area, resulting in false alarms or missed alarms due to individual differences of sensors, environmental noise or response delays in complex building environments. In particular, there is a lack of effective detection methods for scenarios where smoke concentration suddenly changes due to insufficient combustion in the early stage of fire spread but the temperature response lags. Therefore, it is necessary to simultaneously consider the relationship mapping relationship between temperature and smoke sensor and the spatial topological position of the sensor to issue an alarm warning for the clustered area where the fire occurs.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a method, system and storage medium for processing fire data based on temperature and smoke sensing to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The method for processing fire data based on temperature and smoke sensors includes the following specific steps:

[0008] S1: Based on the installation parameters of the temperature sensors and smoke sensors in the building, the spatial topological location of each monitoring point is determined, the temperature data and smoke concentration data of each monitoring point are collected, and the corresponding temperature time series data and smoke time series data are constructed through a unified time index;

[0009] S2: The monitoring point that triggers the smoke sensor is calibrated as the suspected fire trigger point. Based on the trigger time, the concentration change rate sequence is constructed based on the smoke time series data. The concentration segments that rise rapidly within the set time window are identified. The temperature time series segments of the corresponding time period are simultaneously extracted to generate the temperature-smoke response difference mapping pair for each suspected trigger point.

[0010] S3: Based on the temperature-smoke response difference mapping pair, the smoke concentration rising rate and the corresponding temperature delay response parameter are extracted. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point;

[0011] S4: Traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed;

[0012] S5: Based on the response data of all asynchronous early trigger points in the suspected fire cluster area, a regional response parameter set is constructed. The parameter set includes the mean smoke concentration rising rate, the mean temperature delay response, the spatial density of the trigger points and the response time distribution. Multi-conditional reasoning based on combined threshold judgment and logical rule chain is performed to determine whether to trigger the regional fire warning instruction.

[0013] Furthermore, constructing corresponding temperature time series data and smoke time series data through a unified time index includes the following steps:

[0014] When constructing temperature time series data and smoke time series data, first use the initial time As the starting point, every fixed sampling interval Record data once, The time of each sampling point is ,in The starting index of the time window.

[0015] Temperature time series data Depend on The temperature matrix is ​​composed of the measurement values ​​of the temperature sensors, and each row in the temperature matrix corresponds to a time point , each column corresponds to a temperature sensor, element Indicates the Sensor No. The temperature of the sampling points, the overall dimension is ;

[0016] Similarly, smoke time series data Depend on The smoke concentration matrix is ​​composed of the measured values ​​of smoke sensors, and each row corresponds to a time point. , each column corresponds to a smoke sensor, the matrix elements Indicates the Sensor No. The smoke concentration at each sampling point is ; The two types of data have the same time series .

[0017] Furthermore, the temperature matrix and smoke concentration matrix are:

[0018] ;

[0019] in,

[0020] ;

[0021] Indicates the number of temperature sensors; Indicates the Sampling point time; represents the temperature matrix; Indicates the Sensor No. The temperature of each sampling point; represents the sampling interval; Indicates the starting time index of the time window; Indicates the initial time;

[0022] The smoke time series data is:

[0023] ;

[0024] represents the smoke concentration matrix; Indicates the Sensor No. Smoke concentration at each sampling point; Indicates the number of smoke sensors.

[0025] Furthermore, a concentration change rate sequence is constructed based on the smoke time series data. The specific steps are as follows:

[0026] For each smoke sensor , calculate the time The rate of change of:

[0027] ;

[0028] in,

[0029] ;

[0030] ;

[0031] Indicates the number index of smoke sensors; Indicates the The sensors in The concentration change rate at each moment; Indicates the total number of sampling points;

[0032] Construct the concentration change rate matrix:

[0033] ;

[0034] in, represents the concentration change rate matrix;

[0035] Furthermore, a temperature-smoke response difference mapping pair is generated for each suspected trigger point. The specific steps are as follows:

[0036] Set the time window length to ,in, Indicates the number of sampling intervals, for each time window , calculate the smoke rate change:

[0037] ;

[0038] Indicates a smoke sensor in the window The smoke change rate;

[0039] Conditions for determining rapid rise in smoke concentration:

[0040] ;

[0041] Indicates the rate threshold; Indicates the continuous growth coefficient of the control rate; represents a time index variable;

[0042] Screening and smoke sensors Spatial distance is less than Temperature sensor:

[0043] ;

[0044] in, Indicates the number index of temperature sensors; Indicates temperature sensor The spatial coordinates of Indicates smoke sensor The spatial coordinates of represents the maximum association distance; Indicates smoke sensor Spatial distance is less than A collection of temperature sensors;

[0045] For each associated temperature sensor , extract the temperature series within the window:

[0046] ;

[0047] Indicates the In the time window, dimensional data sequence of a temperature sensor;

[0048] Construct a temperature-smoke response difference mapping pair for each suspected trigger point:

[0049] ;

[0050] in,

[0051] ;

[0052] represents a pair of temperature-smoke response difference maps; Indicates the delay time of temperature relative to smoke; Indicates the maximum allowed delay time; Indicates the Smoke sensors at time The rate of change of concentration; Indicates the In the time window, Temperature sensors at relative time index The temperature value at Indicates temperature sensor Response relative to smoke sensor Delay time; Represents the local time index within the current window.

[0053] Furthermore, the specific steps for marking the monitoring point as an asynchronous early trigger point are as follows:

[0054] In the temperature and smoke response difference mapping analysis, the asynchronous early trigger point is marked by setting a complex condition: when a smoke sensor Real-time rate of change Exceeds the preset smoke change rate threshold When , and this sensor is associated with the temperature sensor Response time difference Greater than the temperature delay threshold When , it is determined that there is an early trigger caused by the significant asynchronous response of smoke and temperature at this location, and it is marked as an asynchronous early trigger point.

[0055] Furthermore, constructing the suspected fire gathering area includes the following steps:

[0056] Based on temperature-smoke response difference mapping , define the trigger point set by the following conditions:

[0057] ;

[0058] Represents the set of all valid fire trigger points; express The rate of change of radiation intensity at a point; Indicates the trigger points; Indicates smoke sensor The smoke signal rise rate;

[0059] The monitoring area is divided into two parts with side lengths of Orthogonal network:

[0060] ;

[0061] Indicates the Rank The grid cell of the column; Represents the grid side length; Represents the row index of the grid cell; Represents the column index of the grid cell;

[0062] Count the number of trigger points in each grid:

[0063] ;

[0064] Indicates the number of trigger points; Represents the indicator function, when the element belong The value is 1 when it is set, otherwise it is 0;

[0065] Construct suspected fire gathering areas:

[0066] ;

[0067] in, Indicates the trigger point threshold; Indicates suspected fire cluster areas.

[0068] Furthermore, the specific steps for constructing the regional response parameter set are as follows:

[0069] Based on suspected fire cluster areas , Indicates the For each fire area, the mean smoke concentration rising rate, the mean temperature delay response, the spatial density of trigger points and the distribution of response time were constructed;

[0070] The constructed smoke concentration rising rate mean is:

[0071] ;

[0072] in, Indicates area The number of grids included in ; The transmission is the average rate of increase of smoke concentration;

[0073] The mean of the temperature delay response is constructed as:

[0074] ;

[0075] represents the mean value of temperature delay response;

[0076] The spatial density of the trigger points is constructed as:

[0077] ;

[0078] represents the spatial density of trigger points;

[0079] The standard deviation of the response time distribution is constructed as:

[0080] ;

[0081] Indicates the Standard deviation of the response time distribution in each fire area; Indicates the The observed response time of each data point; No. The observed mean response time of each fire area;

[0082] Adopt hierarchical decision logic and define rule chains:

[0083] ;

[0084] in, Indicates the smoke rising rate threshold; Indicates the temperature delay threshold; represents the spatial density threshold; Indicates the time distribution threshold.

[0085] The present invention further provides a fire data processing system based on temperature and smoke sensing, wherein the processing system is used to execute the above-mentioned processing method, including:

[0086] The data acquisition module is used to determine the spatial topological location of each monitoring point based on the installation parameters of the temperature sensors and smoke sensors in the building, collect the temperature data and smoke concentration data of each monitoring point, and construct the corresponding temperature time series data and smoke time series data through a unified time index;

[0087] The trigger identification module is used to calibrate the monitoring point that triggers the smoke sensor as a suspected fire trigger point. Based on the trigger time, it constructs a concentration change rate sequence based on the smoke time series data, identifies the concentration segments that rise rapidly within the set time window, and simultaneously extracts the temperature time series segments of the corresponding time period to generate a temperature-smoke response difference mapping pair for each suspected trigger point.

[0088] The feature recognition module is used to extract the smoke concentration rising rate and the corresponding temperature delay response parameter based on the temperature-smoke response difference mapping pair. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point;

[0089] The fire clustering module is used to traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed;

[0090] The fire reasoning module is used to construct a regional response parameter set based on the response data of all asynchronous early trigger points in the suspected fire cluster area. The parameter set includes the mean smoke concentration rise rate, the mean temperature delay response, the spatial density of trigger points and the response time distribution. It performs multi-conditional reasoning based on combined threshold judgment and logical rule chain to determine whether to trigger the regional fire warning instruction.

[0091] The present invention further provides a storage medium, characterized in that the storage medium stores a computer program that can be run on a processor, and when the computer program is executed by the processor, the steps of the above-mentioned method for processing fire data based on temperature and smoke sensing are implemented.

[0092] Compared with the prior art, the present invention has the following beneficial effects:

[0093] The present invention constructs a concentration change rate sequence based on smoke time series data, generates a temperature-smoke response difference mapping pair for each suspected trigger point, and marks the monitoring point as an asynchronous early trigger point, which can accurately distinguish between real fire events and interference events; based on the temperature-smoke response difference mapping pair and the trigger point spatial density analysis, it can identify early fire conditions in advance, and realize the command operation of triggering fire warning in the combustion stage through the mean value of the smoke concentration rising rate, the mean value of the temperature delay response, the spatial density of the trigger point and the response time distribution, further improving the accuracy of fire prediction, and can also determine the fire spread path in advance through the trigger point clustering algorithm, so that it has stronger robustness in complex areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 A flow chart of the overall method of fire data processing method, system and storage medium based on temperature and smoke sensing;

[0095] Figure 2 Schematic diagram of asynchronous early trigger point data in a fire data processing method, system, and storage medium based on temperature and smoke sensing;

[0096] Figure 3 The figure is a block diagram of the overall system of the temperature and smoke sensing fire data processing method, system and storage medium. DETAILED DESCRIPTION

[0097] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0098] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0099] Example:

[0100] See also Figure 1 and Figure 2 , the present invention provides a technical solution:

[0101] The method for processing fire data based on temperature and smoke sensors includes the following specific steps:

[0102] S1: Based on the installation parameters of the temperature sensors and smoke sensors in the building, the spatial topological location of each monitoring point is determined, the temperature data and smoke concentration data of each monitoring point are collected, and the corresponding temperature time series data and smoke time series data are constructed through a unified time index;

[0103] The construction of corresponding temperature time series data and smoke time series data by using a unified time index includes the following steps:

[0104] When constructing temperature time series data and smoke time series data, first use the initial time As the starting point, every fixed sampling interval Record data once, The time of each sampling point is ,in The starting index of the time window.

[0105] Temperature time series data Depend on The temperature matrix is ​​composed of the measurement values ​​of the temperature sensors, and each row in the temperature matrix corresponds to a time point , each column corresponds to a temperature sensor, element Indicates the Sensor No. The temperature of the sampling points, the overall dimension is ;

[0106] Similarly, smoke time series data Depend on The smoke concentration matrix is ​​composed of the measured values ​​of smoke sensors, and each row corresponds to a time point. , each column corresponds to a smoke sensor, the matrix elements Indicates the Sensor No. The smoke concentration at each sampling point is ; The two types of data have the same time series .

[0107] The temperature matrix and smoke concentration matrix are:

[0108] ;

[0109] in,

[0110] ;

[0111] Indicates the number of temperature sensors; Indicates the Sampling point time; represents the temperature matrix; Indicates the Sensor No. The temperature of each sampling point; represents the sampling interval; Indicates the starting time index of the time window; Indicates the initial time;

[0112] The smoke time series data is:

[0113] ;

[0114] represents the smoke concentration matrix; Indicates the Sensor No. Smoke concentration at each sampling point; Indicates the number of smoke sensors.

[0115] Determining the spatial topological position of each monitoring point specifically includes: allocating corresponding temperature sensors and smoke concentration sensors according to the building use, such as warehouse, corridor, office, etc., recording the spatial three-dimensional coordinates of each sensor through laser positioning equipment, and using the reference points of the building structure, such as the building origin and the floor center point, as the reference coordinate system, uniformly converting all sensor positions into relative coordinates to generate a distance list between each sensor.

[0116] S2: The monitoring point that triggers the smoke sensor is calibrated as the suspected fire trigger point. Based on the trigger time, the concentration change rate sequence is constructed based on the smoke time series data. The concentration segments that rise rapidly within the set time window are identified. The temperature time series segments of the corresponding time period are simultaneously extracted to generate the temperature-smoke response difference mapping pair for each suspected trigger point.

[0117] The specific steps of constructing a concentration change rate sequence based on smoke time series data are as follows:

[0118] For each smoke sensor , calculate the time The rate of change of:

[0119] ;

[0120] in,

[0121] ;

[0122] ;

[0123] Indicates the number index of smoke sensors; Indicates the The sensors in The concentration change rate at each moment; Indicates the total number of sampling points;

[0124] Dependent variable For smoke sensors exist The concentration change rate at the moment, by calculating the concentration difference change rate at adjacent moments through central difference, can capture the nonlinear diffusion characteristics of the early stage of fire and suppress the noise of the sensor; the independent variable Reflects the physical change of smoke concentration, and its difference directly drives The amplitude, and the time interval As a system parameter, the sensitivity is balanced by adjusting the denominator; the concentration rises and Positive correlation, concentration decreases and There is a negative correlation, and When it decreases, the sampling rate increases. This formula quantifies the physical correlation between the dynamic characteristics of the fire and the sensor response, providing a core judgment basis for fire monitoring.

[0125] Construct the concentration change rate matrix:

[0126] ;

[0127] in, represents the concentration change rate matrix.

[0128] The specific steps of generating the temperature-smoke response difference mapping pair for each suspected trigger point are as follows:

[0129] Set the time window length to ,in, Indicates the number of sampling intervals, for each time window , calculate the smoke rate change:

[0130] ;

[0131] Indicates a smoke sensor in the window The smoke change rate;

[0132] Conditions for determining rapid rise in smoke concentration:

[0133] ;

[0134] Indicates the rate threshold; Indicates the continuous growth coefficient of the control rate; represents a time index variable;

[0135] Screening and smoke sensors Spatial distance is less than Temperature sensor:

[0136] ;

[0137] in, Indicates the number index of temperature sensors; Indicates temperature sensor The spatial coordinates of Indicates smoke sensor The spatial coordinates of represents the maximum association distance; Indicates smoke sensor Spatial distance is less than A collection of temperature sensors;

[0138] For each associated temperature sensor , extract the temperature series within the window:

[0139] ;

[0140] Indicates the In the time window, dimensional data sequence of a temperature sensor;

[0141] Construct a temperature-smoke response difference mapping pair for each suspected trigger point:

[0142] ;

[0143] in,

[0144] ;

[0145] represents a pair of temperature-smoke response difference maps; Indicates the delay time of temperature relative to smoke; Indicates the maximum allowed delay time; Indicates the Smoke sensors at time The rate of change of concentration; Indicates the In the time window, Temperature sensors at relative time index The temperature value at Indicates temperature sensor Response relative to smoke sensor Delay time; Represents the local time index within the current window.

[0146] S3: Based on the temperature-smoke response difference mapping pair, the smoke concentration rising rate and the corresponding temperature delay response parameter are extracted. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point;

[0147] The specific steps of marking the monitoring point as an asynchronous early trigger point are:

[0148] In the temperature and smoke response difference mapping analysis, the asynchronous early trigger point is marked by setting a complex condition: when a smoke sensor Real-time rate of change Exceeds the preset smoke change rate threshold When , and this sensor is associated with the temperature sensor Response time difference Greater than the temperature delay threshold When , it is determined that there is an early trigger caused by the significant asynchronous response of smoke and temperature at this location, and it is marked as an asynchronous early trigger point.

[0149] In this embodiment, five monitoring points are selected, and based on the smoke change rate threshold and the temperature delay threshold, the smoke change rate and the associated temperature response time difference data are established. The experimental marking data is shown in Table 1:

[0150] Table 1: Asynchronous early trigger point markers

[0151]

[0152] As can be seen from Table 1, the monitoring points are numbered 1 to 5, and the smoke change rate threshold is set at 2.5%. The smoke change rates of four monitoring points, 1, 2, 4, and 5, exceed the smoke change rate threshold. At the same time, the temperature delay threshold is set at 30 seconds. Based on the smoke change rate exceeding the threshold, it can be seen that the difference in the associated temperature response time of monitoring points 1 and 4 exceeds the temperature delay threshold, which marks them as asynchronous early trigger points.

[0153] S4: Traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed;

[0154] The construction of the suspected fire gathering area comprises the following steps:

[0155] Based on temperature-smoke response difference mapping , define the trigger point set by the following conditions:

[0156] ;

[0157] Represents the set of all valid fire trigger points; express The rate of change of radiation intensity at a point; Indicates the trigger points; Indicates smoke sensor The smoke signal rise rate;

[0158] The monitoring area is divided into two parts with side lengths of Orthogonal network:

[0159] ;

[0160] Indicates the Rank The grid cell of the column; Represents the grid side length; Represents the row index of the grid cell; Represents the column index of the grid cell;

[0161] Count the number of trigger points in each grid:

[0162] ;

[0163] Indicates the number of trigger points; Represents the indicator function, when the element belong The value is 1 when it is set, otherwise it is 0;

[0164] Indicates that in a specific area In the data set Data points that meet the filter conditions The cumulative number of The scope of relaxation, more will be counted, resulting in Increase; if The data points in are dense, then The value will also increase significantly.

[0165] Construct suspected fire gathering areas:

[0166] ;

[0167] in, Indicates the trigger point threshold; Indicates suspected fire cluster areas.

[0168] S5: Based on the response data of all asynchronous early trigger points in the suspected fire cluster area, a regional response parameter set is constructed. The parameter set includes the mean smoke concentration rising rate, the mean temperature delay response, the spatial density of the trigger points and the response time distribution. Multi-conditional reasoning based on combined threshold judgment and logical rule chain is performed to determine whether to trigger the regional fire warning instruction.

[0169] The specific steps of constructing the regional response parameter set are:

[0170] Based on suspected fire cluster areas , Indicates the For each fire area, the mean smoke concentration rising rate, the mean temperature delay response, the spatial density of trigger points and the distribution of response time were constructed;

[0171] The constructed smoke concentration rising rate mean is:

[0172] ;

[0173] in, Indicates area The number of grids included in ; The transmission is the average rate of increase of smoke concentration;

[0174] The mean of the temperature delay response is constructed as:

[0175] ;

[0176] represents the mean value of temperature delay response;

[0177] The spatial density of the trigger points is constructed as:

[0178] ;

[0179] represents the spatial density of trigger points;

[0180] The standard deviation of the response time distribution is constructed as:

[0181] ;

[0182] Indicates the Standard deviation of the response time distribution in each fire area; Indicates the The observed response time of each data point; No. The observed mean response time of each fire area;

[0183] Adopt hierarchical decision logic and define rule chains:

[0184] ;

[0185] in, Indicates the smoke rising rate threshold; Indicates the temperature delay threshold; represents the spatial density threshold; Indicates the time distribution threshold.

[0186] See also Figure 3 The present invention further provides a fire data processing system based on temperature and smoke sensing, wherein the processing system is used to execute the above-mentioned processing method, including:

[0187] The data acquisition module is used to determine the spatial topological location of each monitoring point based on the installation parameters of the temperature sensors and smoke sensors in the building, collect the temperature data and smoke concentration data of each monitoring point, and construct the corresponding temperature time series data and smoke time series data through a unified time index;

[0188] The trigger identification module is used to calibrate the monitoring point that triggers the smoke sensor as a suspected fire trigger point. Based on the trigger time, it constructs a concentration change rate sequence based on the smoke time series data, identifies the concentration segments that rise rapidly within the set time window, and simultaneously extracts the temperature time series segments of the corresponding time period to generate a temperature-smoke response difference mapping pair for each suspected trigger point.

[0189] The feature recognition module is used to extract the smoke concentration rising rate and the corresponding temperature delay response parameter based on the temperature-smoke response difference mapping pair. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point;

[0190] The fire clustering module is used to traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed;

[0191] The fire reasoning module is used to construct a regional response parameter set based on the response data of all asynchronous early trigger points in the suspected fire cluster area. The parameter set includes the mean smoke concentration rise rate, the mean temperature delay response, the spatial density of trigger points and the response time distribution. It performs multi-conditional reasoning based on combined threshold judgment and logical rule chain to determine whether to trigger the regional fire warning instruction.

[0192] The present invention further provides a storage medium, characterized in that the storage medium stores a computer program that can be run on a processor, and when the computer program is executed by the processor, the steps of the above-mentioned method for processing fire data based on temperature and smoke sensing are implemented.

[0193] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0194] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0195] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0196] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for processing fire data based on temperature and smoke sensing, characterized in that: include: S1: Based on the installation parameters of the temperature sensors and smoke sensors in the building, the spatial topological location of each monitoring point is determined, the temperature data and smoke concentration data of each monitoring point are collected, and the corresponding temperature time series data and smoke time series data are constructed through a unified time index; S2: The monitoring point that triggers the smoke sensor is calibrated as the suspected fire trigger point. Based on the trigger time, the concentration change rate sequence is constructed based on the smoke time series data. The concentration segments that rise rapidly within the set time window are identified. The temperature time series segments of the corresponding time period are simultaneously extracted to generate the temperature-smoke response difference mapping pair for each suspected trigger point. S3: Based on the temperature-smoke response difference mapping pair, the smoke concentration rising rate and the corresponding temperature delay response parameter are extracted. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point; S4: Traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed; S5: Based on the response data of all asynchronous early trigger points in the suspected fire cluster area, a regional response parameter set is constructed. The parameter set includes the mean smoke concentration rising rate, the mean temperature delay response, the spatial density of the trigger points and the response time distribution. Multi-conditional reasoning based on combined threshold judgment and logical rule chain is performed to determine whether to trigger the regional fire warning instruction.

2. The method for processing fire data based on temperature and smoke sensing according to claim 1, characterized in that: The construction of corresponding temperature time series data and smoke time series data by using a unified time index includes the following steps: When constructing temperature time series data and smoke time series data, we first start with the initial time t0 and record data every fixed sampling interval Δt. The time of the kth sampling point is t k =t0+kΔt, where k is the starting index of the time window; The temperature time series data T consists of the measurement values ​​of N temperature sensors, and each row in the temperature matrix corresponds to a time point t k , each column corresponds to a temperature sensor, element T N (t k ) represents the temperature of the kth sampling point of the Nth sensor, and the overall dimension is (k+1)×N; The smoke time series data S is composed of the measurement values ​​of M smoke sensors, and each row in the smoke concentration matrix corresponds to a time point t k , each column corresponds to a smoke sensor, the matrix element S M (t k ) represents the smoke concentration at the kth sampling point of the Mth sensor, with a dimension of (k+1)×M; the two types of data have the same time series t0, t1,…, t k .

3. The method for processing fire data based on temperature and smoke sensing according to claim 2, characterized in that: The temperature matrix and smoke concentration matrix are: in, t k =t0+kΔt N represents the number of temperature sensors; t k represents the time of the kth sampling point; T represents the temperature matrix; T N (t k ) represents the temperature of the kth sampling point of the Nth sensor; Δt represents the sampling interval; k represents the starting time index of the time window; t0 represents the initial time; The smoke time series data is: S represents the smoke density matrix; S M (t k ) represents the smoke concentration at the kth sampling point of the Mth sensor; M represents the number of smoke sensors.

4. The method for processing fire data based on temperature and smoke sensing according to claim 3 is characterized in that: The specific steps of constructing a concentration change rate sequence based on smoke time series data are as follows: For each smoke sensor j∈[1,M], calculate the time t k The rate of change of: in, j represents the number index of smoke sensors; r j (t k ) indicates that the jth sensor is at t k The concentration change rate at the moment; K represents the total number of sampling points; Construct the concentration change rate matrix: Where R represents the concentration change rate matrix.

5. The method for processing fire data based on temperature and smoke sensing according to claim 4, characterized in that: The specific steps of generating the temperature-smoke response difference mapping pair for each suspected trigger point are as follows: Set the time window length to W = wΔt, where w represents the number of sampling intervals. For each time window Calculate the smoke velocity change: S r (j) represents the smoke change rate of smoke sensor j in the window; Conditions for determining rapid rise in smoke concentration: θ represents the rate threshold; α represents the coefficient for controlling the continuous growth of the rate; t represents the time index variable; Filter the space distance between the smoke sensor and the smoke sensor j that is less than D max Temperature sensor: N j ={n∈[1,N]||x n -y j ||2≤D max } Where n represents the number index of temperature sensors; x n represents the spatial coordinate of temperature sensor n; y j represents the spatial coordinates of smoke sensor j; D max Indicates the maximum association distance; N j Indicates that the spatial distance from smoke sensor j is less than D max A collection of temperature sensors; For each associated temperature sensor Extract the temperature series within the window: T i,n =[T n (t k ),T n (t k+1 ),…,T n (t k+w-1 )] T i,n Represents the dimensional data sequence of the nth temperature sensor in the i-th time window; Construct a temperature-smoke response difference mapping pair for each suspected trigger point: in, represents the temperature-smoke response difference mapping pair; τ represents the delay time of temperature relative to smoke; δ represents the maximum allowed delay time; r j (t) represents the concentration change rate of the j-th smoke sensor at time t; T i,n (t+τ-k) represents the temperature value of the nth temperature sensor at the relative time index t+τ-k in the i-th time window; Δτ(j,n) represents the delay time of the response of temperature sensor n relative to smoke sensor j; t represents the local time index in the current window.

6. The method for processing fire data based on temperature and smoke sensing according to claim 5, characterized in that: The specific steps of marking the monitoring point as an asynchronous early trigger point are: In the temperature and smoke response difference mapping analysis, the asynchronous early trigger point is marked by setting a composite condition: when the real-time change rate S of a certain smoke sensor j r (j) Exceeding the preset smoke change rate threshold S threshold When S r (j)>S threshold , and the response time difference Δτ(j,n) between the sensor and the associated temperature sensor n is greater than the temperature delay threshold τ th When Δτ(j,n)>τ th , it is determined that there is an early trigger caused by the significant asynchronous response of smoke and temperature at this location, and it is marked as an asynchronous early trigger point.

7. The method for processing fire data based on temperature and smoke sensing according to claim 6, characterized in that: The construction of the suspected fire gathering area comprises the following steps: Based on the temperature-smoke response difference mapping (S r (j),Δτ(j,n)), the trigger point set is defined by the following conditions: P={p h =x n ∣n∈N j ,S r (j)≥S threshold ∧Δτ(j,n)>τ th } P represents the set of all effective fire trigger points; p h Indicates the hth trigger point; S r (j) represents the rising rate of the smoke signal of smoke sensor j; Divide the monitoring area into an orthogonal network with a side length of l: G m,n =[ml,(m+1)l]×[nl,(n+1)l] G m,n Represents the grid cell at the mth row and nth column; l represents the grid side length; m represents the row index of the grid cell; n represents the column index of the grid cell; Count the number of trigger points in each grid: C m,n Indicates the number of trigger points; Represents the indicator function, when the element p h Belong to G m,n The value is 1 when it is set, otherwise it is 0; Construct suspected fire gathering areas: Among them, τ c Indicates the trigger point threshold; Indicates suspected fire cluster areas.

8. The method for processing fire data based on temperature and smoke sensing according to claim 7, characterized in that: The specific steps of constructing the regional response parameter set are: Based on suspected fire cluster areas R l Representing the lth fire area, construct the mean value of smoke concentration rising rate, mean value of temperature delay response, spatial density of trigger points and distribution of response time; The constructed smoke concentration rising rate mean is: Among them, |R l | represents region R l The number of grids included in ; The transmission is the average rate of increase of smoke concentration; The mean of the temperature delay response is constructed as: represents the mean value of temperature delay response; The spatial density of the trigger points is constructed as: D(l) represents the spatial density of trigger points; The standard deviation of the response time distribution is constructed as: σ T (l) represents the standard deviation of the response time distribution of the lth fire area; t h Represents the response duration observation value of the hth data point; The observed mean response time of the lth fire area; Adopt hierarchical decision logic and define rule chains: Among them, τ S represents the smoke rising rate threshold; τ Δ represents the temperature delay threshold; τ D represents the spatial density threshold; τ σ Indicates the time distribution threshold.

9. Fire data processing system based on temperature and smoke sensing, characterized by: The system is used to execute the method for processing fire data based on temperature and smoke sensing according to any one of claims 1 to 8, comprising: The data acquisition module is used to determine the spatial topological location of each monitoring point based on the installation parameters of the temperature sensors and smoke sensors in the building, collect the temperature data and smoke concentration data of each monitoring point, and construct the corresponding temperature time series data and smoke time series data through a unified time index; The trigger identification module is used to calibrate the monitoring point that triggers the smoke sensor as a suspected fire trigger point. Based on the trigger time, it constructs a concentration change rate sequence based on the smoke time series data, identifies the concentration segments that rise rapidly within the set time window, and simultaneously extracts the temperature time series segments of the corresponding time period to generate a temperature-smoke response difference mapping pair for each suspected trigger point. The feature recognition module is used to extract the smoke concentration rising rate and the corresponding temperature delay response parameter based on the temperature-smoke response difference mapping pair. If the combination of the two meets the preset asynchronous feature recognition standard, the monitoring point is marked as an asynchronous early trigger point; The fire clustering module is used to traverse all asynchronous early trigger points and cluster them according to their spatial topological positions. If the clustering results show that the number of trigger points exceeds the set threshold and the spatial distribution is continuous, a suspected fire cluster area is constructed; The fire reasoning module is used to construct a regional response parameter set based on the response data of all asynchronous early trigger points in the suspected fire cluster area. The parameter set includes the mean smoke concentration rise rate, the mean temperature delay response, the spatial density of trigger points and the response time distribution. It performs multi-conditional reasoning based on combined threshold judgment and logical rule chain to determine whether to trigger the regional fire warning instruction.

10. A storage medium, characterized in that: The storage medium stores a computer program that can be run on a processor, and when the computer program is executed by the processor, the steps of the method for processing fire data based on temperature and smoke sensing are implemented as described in any one of claims 1 to 8.

Images