Early fire spreading rule and fire early warning and alarming method based on Internet platform

Abstract

The invention provides an early fire spreading rule and fire early warning alarm method based on an Internet platform, and the method comprises the steps: carrying out the linear regression fitting of a temperature parameter and a time parameter through an early fire intelligent operation and maintenance Internet platform, obtaining a first feature parameter and a second feature parameter, and comparing the first feature parameter and the second feature parameter to judge the early fire stage of a fire scene; and generating a graded early warning signal according to a comparison result of the early fire stage and the duration, and pushing the graded early warning signal to execute graded early warning. An early fire database of cigarette end ignition, electric heating wire ignition, electric spark ignition and electric welding slag ignition is studied, temperature field changes of different temperature sensors in the same space are monitored in real time through uniform array arrangement of the temperature sensors in the space, early fire is judged according to a research model and the database, and early warning and alarming are carried out. Meanwhile, the intelligent automatic water-spraying fire extinguishing system is linked to start electric water spraying to extinguish fire early. The early fire spreading rule is summarized, and the timeliness of fire early warning is improved.

Description

Technical Field

[0001] This invention relates to the field of fire prevention technology, and in particular to a method for early fire spread patterns and fire early warning alarms based on an internet platform. Background Technology

[0002] The development of a fire generally goes through the following stages: Figure 1 The five stages shown are: initial stage, development stage, complete combustion stage, decay stage, and extinction stage. Figure 1 The fire developed without the intervention of fire-fighting facilities. In buildings with typical floor heights, when an automatic fire alarm system sounds and security personnel manually extinguish the fire, the fire size is relatively small. When an automatic sprinkler system is activated, the ceiling temperature is typically below 250°C with a standard response sprinkler and around 100°C with a fast response sprinkler. In traditional office spaces, this fire size is usually a small fire between 0.25 and 0.50 MW, depending on the ceiling height, and is easily extinguished. However, if a flashover occurs, all combustible materials in the room will ignite simultaneously. Therefore, research on early fire detection and extinguishing is essential. However, current traditional fire detection methods often only detect fires and activate sprinkler systems during the development stage, failing to detect fires in their initial or even earlier stages. They cannot identify fires in a timely manner or rely on intelligent methods such as big data and machine learning for timely fire identification. Instead, they can only detect fires based on experience or single data and methods such as temperature monitoring. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the present invention provides a method, device and system for early fire spread patterns and fire early warning alarm based on the Internet platform. The present invention summarizes the early fire spread patterns and improves the timeliness of fire early warning.

[0004] This invention provides a method for early fire spread patterns and fire early warning alarms based on an internet platform, comprising:

[0005] The front-end sensing devices acquire temperature and time parameters of the fire scene according to the data collection instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and upload them to the early-stage fire intelligent operation and maintenance internet platform.

[0006] The early-stage fire safety intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature and time parameters to obtain the first and second characteristic parameters; and,

[0007] Match the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate reference parameter and / or the time reference parameter, and determine the early stage of the fire based on the matching results.

[0008] The second feature parameter is compared with the duration parameter to obtain the duration comparison result;

[0009] Based on the comparison results of the early stage of the fire and its duration, a graded early warning signal is generated and pushed out to implement the graded early warning.

[0010] Furthermore, temperature and time parameters of the fire scene were obtained, and a large database of early-stage fire parameters was constructed.

[0011] The data from the early-onset fire parameter database is input into the early-onset fire detection machine learning model to obtain the first feature parameter and the second feature parameter. The early-onset fire detection machine learning model is obtained by linear regression fitting of the data from the early-onset fire parameter database.

[0012] Match the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate reference parameter and / or the time reference parameter, and determine the early stage of the fire based on the matching results.

[0013] The second feature parameter is compared with the duration parameter to obtain the duration comparison result;

[0014] Based on the comparison results of the early stage of fire and its duration, a graded early warning signal is issued through the early fire intelligent operation and maintenance Internet platform.

[0015] Implement tiered early warnings based on the tiered warning signals.

[0016] Furthermore, the sliding window method is used to obtain the temperature and time parameters of the fire scene. The first characteristic parameter is calculated by linear regression fitting of the temperature and time parameter data within the window. The continuous cumulative time corresponding to the first characteristic parameter is the second characteristic parameter.

[0017] Furthermore, the temperature rise rate comparison parameters include a first temperature rise rate, a second temperature rise rate, and a third temperature rise rate, and the time comparison parameters include a first time comparison parameter, a second time comparison parameter, and a third time comparison parameter.

[0018] Furthermore, the early stages of a fire include: the initial or smoldering stage, the slow growth stage, and the rapid growth stage.

[0019] Furthermore, when the first characteristic parameter and / or the second characteristic parameter meet the first temperature rise rate and / or the first time reference parameter, the fire is in the initial or smoldering stage; when the first characteristic parameter and / or the second characteristic parameter meet the second temperature rise rate and / or the second time reference parameter, the fire is in the slow growth stage; when the first characteristic parameter and / or the second characteristic parameter meet the third temperature rise rate and / or the third time reference parameter, the fire is in the rapid growth stage.

[0020] Furthermore, historical fire data is input into the fire spread law function to obtain temperature rise rate comparison parameters and time comparison parameters.

[0021] Furthermore, the fire spread law function was obtained by fitting a linear regression model:

[0022] y=kx+b

[0023] Where: y is the temperature parameter, x is the time parameter, and k is the first characteristic parameter.

[0024] Another aspect of the present invention provides a device for early fire spread patterns and fire early warning alarms, comprising:

[0025] The parameter acquisition module is used to acquire temperature and time parameters at the fire scene.

[0026] The linear regression fitting module is used to perform linear regression fitting on temperature and time parameters to obtain the first and second feature parameters.

[0027] The matching module matches the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate reference parameter and / or the time reference parameter and obtains the matching result;

[0028] The comparison module compares the second feature parameter with the duration parameter to obtain the duration comparison result.

[0029] The early warning module determines the early stage of the fire based on the matching results and the duration comparison results, and issues graded early warning signals.

[0030] The early warning execution module executes tiered early warnings based on the tiered early warning signals.

[0031] Furthermore, it also includes:

[0032] The comparison parameter fitting module inputs historical fire data into the fire spread law function to obtain the temperature rise rate comparison parameter and the time comparison parameter.

[0033] In another aspect, this invention provides a system for detecting the early spread of fires and for early warning and alarm systems, employing the aforementioned method for detecting the early spread of fires and for early warning and alarm systems based on an internet platform, comprising:

[0034] Temperature acquisition mechanism, used to acquire temperature parameters;

[0035] Time acquisition mechanism; the time acquisition mechanism is used to acquire time parameters.

[0036] The controller acquires temperature and time parameters of the fire scene; performs linear regression fitting on the temperature and time parameters to obtain a first characteristic parameter and a second characteristic parameter; matches the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate reference parameter and / or the time reference parameter, determines the early stage of the fire scene based on the matching result, compares the second characteristic parameter with the duration parameter to obtain the duration comparison result; and issues a graded early warning signal based on the early stage of the fire scene and the duration comparison result.

[0037] The nozzle assembly executes graded warnings based on graded warning signals.

[0038] In another aspect, the present invention provides an intelligent operation and maintenance system for early-stage fires, which is used to detect early-stage fires and perform intelligent operation and maintenance. The intelligent operation and maintenance system for early-stage fires is equipped with the aforementioned early-stage fire spread pattern and fire early warning alarm device.

[0039] In another aspect, the present invention provides an intelligent operation and maintenance internet platform for early-stage fires, which is used to detect early-stage fires and perform intelligent operation and maintenance. The intelligent operation and maintenance internet platform for early-stage fires adopts the above-mentioned internet platform-based method for early-stage fire spread patterns and fire early warning alarms.

[0040] As can be seen from the above solutions, the advantages of the present invention are:

[0041] In one embodiment, the front-end sensing device acquires temperature and time parameters of the fire scene according to the data acquisition instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and uploads them to the platform. The platform performs linear regression fitting on the received temperature and time parameters to obtain a first feature parameter and a second feature parameter, thus obtaining feature parameters that represent the real-time status of the fire scene. The first and / or second feature parameters are matched with the temperature rise rate comparison parameter and / or time comparison parameter. Based on the matching result, the early-stage fire scene is determined. The second feature parameter is then compared with the duration parameter to obtain the duration comparison result. Based on the early-stage fire scene and the duration comparison result, a graded early warning signal is generated and pushed to execute the graded early warning, so that the graded early warning can be executed in a timely manner, allowing for timely detection and extinguishing of the fire in the early-stage fire scene. The research has established a database of early-onset fires caused by cigarette butt ignition, heating wire ignition, electric spark ignition, and welding slag ignition. By uniformly arranging temperature sensors in a spatial array, the temperature field changes of different temperature sensors in the same space are monitored in real time. Based on the research model and database, early-onset fires can be identified and early warning alarms can be issued. At the same time, the intelligent automatic sprinkler system can be linked to activate electric sprinklers to extinguish early-onset fires. Attached Figure Description

[0042] Figure 1This is a schematic diagram of the fire development process in existing technology;

[0043] Figure 2 A flowchart illustrating the early fire spread pattern and fire early warning alarm method based on an internet platform, as provided in an embodiment of the present invention;

[0044] Figure 3 The graph shows the fitting results of temperature and time parameters when the burning material is a stack of wood and the burning material is 0m away from the fire source.

[0045] Figure 4 The graph shows the fitting results of temperature and time parameters when the burning material is a stack of wood, 1.5m away from the fire source.

[0046] Figure 5 The graph shows the fitting results of temperature and time parameters when the burning material is a stack of wood, at a distance of 2.12m from the fire source.

[0047] Figure 6 The graph shows the fitting results of temperature and time parameters when the burning material is a stack of wood, 3m away from the fire source.

[0048] Figure 7 The fitting result of temperature and time parameters when the burning material is a standard burning material and the burning material is 0m away from the fire source.

[0049] Figure 8 The figure shows the fitting results of temperature and time parameters when the burning material is a standard burning material and is 1.5m away from the fire source.

[0050] Figure 9 The figure shows the fitting results of temperature and time parameters when the burning material is a standard burning material and the burning material is 2.12m away from the fire source.

[0051] Figure 10 The graph shows the fitting results of temperature and time parameters when the burning material is a standard burning material, and the burning material is 3m away from the fire source.

[0052] Figure 11 A diagram illustrating the fire's development during its early stages;

[0053] Figure 12 A schematic diagram illustrating the results of detecting and responding to early-stage fires;

[0054] Figure 13 This is a diagram illustrating the early stages of a fire's spread.

[0055] Figure 14 The test results for test number 1 are as follows: when a combustible material is placed directly below a nozzle.

[0056] Figure 15The test results for test number 2 are as follows: when a combustible material is placed directly below a nozzle.

[0057] Figure 16 The result of test number 3 is when a combustible material is placed directly below a nozzle;

[0058] Figure 17 The result of test number 1 is when the combustible material is placed directly below the center of the four nozzles;

[0059] Figure 18 The result of test number 2 is when the combustible material is placed directly below the center of the four nozzles;

[0060] Figure 19 The result of test number 3 is when the combustible material is placed directly below the center of the four nozzles;

[0061] Figure 20 A schematic diagram illustrating the early-stage fire spread pattern and fire early warning alarm devices;

[0062] Figure 21 Schematic diagram of the nozzle assembly and temperature acquisition mechanism.

[0063] In the attached figures, the following labels are used:

[0064] 10. Early-morning fire spread patterns and fire early warning alarm devices;

[0065] 11-Parameter Acquisition Module;

[0066] 12-Linear Regression Fitting Module;

[0067] 13-Matching module;

[0068] 14-Comparison Module;

[0069] 15 - Early Warning Issuance Module;

[0070] 16-Early Warning Execution Module;

[0071] 20 - Temperature acquisition mechanism;

[0072] 30 - Nozzle assembly;

[0073] 40 - Anemometer. Detailed Implementation

[0074] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments to further understand the purpose, solution and effect of the present invention, but it is not intended to limit the scope of protection of the appended claims.

[0075] References to "embodiment," "another embodiment," "this embodiment," etc., in the specification refer to embodiments that may include specific features, structures, or characteristics, but not every embodiment must include these specific features, structures, or characteristics. Furthermore, such expressions do not refer to the same embodiment. Moreover, when describing specific features, structures, or characteristics in conjunction with embodiments, whether or not explicitly described, it is indicated that incorporating such features, structures, or characteristics into other embodiments is within the knowledge of those skilled in the art.

[0076] The specification and subsequent claims use certain terms to refer to specific components or parts. Those skilled in the art will understand that users or manufacturers may use different names or terms to refer to the same component or part. This specification and claims do not distinguish components or parts by differences in name, but rather by differences in function. The terms "comprising" and "including" used throughout the specification and claims are open-ended and should be interpreted as "including but not limited to". Furthermore, the term "connection" here includes any direct and indirect electrical connection means. Indirect electrical connection means include connections via other means.

[0077] Figure 2 The method for early fire spread patterns and fire early warning alarm based on an internet platform provided by this invention includes the following steps:

[0078] S1: The front-end sensing device obtains the temperature and time parameters of the fire scene according to the data collection instructions issued by the early fire intelligent operation and maintenance Internet platform, and uploads them to the early fire intelligent operation and maintenance Internet platform.

[0079] In one embodiment, the front-end sensing device acquires the temperature and time parameters of the fire scene according to the data acquisition instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and uploads them to the early-stage fire intelligent operation and maintenance internet platform for real-time monitoring of the fire scene.

[0080] S2: The early-stage fire intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature and time parameters to obtain the first and second characteristic parameters;

[0081] In one embodiment, the early-stage fire intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature and time parameters to obtain the first feature parameter and the second feature parameter.

[0082] S3: Match the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate comparison parameter and / or the time comparison parameter, and determine the early stage of the fire based on the matching result;

[0083] In one embodiment, the first characteristic parameter and / or the second characteristic parameter are matched with the temperature rise rate comparison parameter and / or the time comparison parameter to obtain the matching result of whether the first characteristic parameter and the second characteristic parameter match the temperature rise rate comparison parameter and the time comparison parameter. Based on the matching result, the early stage of fire at the fire site can be determined.

[0084] In one embodiment, the first characteristic parameter is matched with the temperature rise rate comparison parameter to obtain a matching result of whether the first characteristic parameter and the temperature rise rate comparison parameter match. Based on the matching result, the early stage of the fire can be determined.

[0085] In one embodiment, the second feature parameter is matched with the time comparison parameter to obtain a matching result of whether the second feature parameter and the time comparison parameter match. The fire stage of the fire scene can be determined based on the matching result.

[0086] S4: Compare the second feature parameter with the duration parameter to obtain the duration comparison result;

[0087] S5: Based on the comparison results of the early stage of the fire and its duration, generate and push graded early warning signals to implement graded early warning.

[0088] In one embodiment, a graded early warning is executed based on the issued graded early warning signal to control the fire scene.

[0089] In one embodiment, the front-end sensing device acquires temperature and time parameters of the fire scene according to the data acquisition instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and uploads them to the platform. The platform performs linear regression fitting on the received temperature and time parameters to obtain a first feature parameter and a second feature parameter, thus obtaining feature parameters that represent the real-time status of the fire scene. The first and / or second feature parameters are matched with the temperature rise rate comparison parameter and / or time comparison parameter. Based on the matching result, the early-stage fire scene is determined. The second feature parameter is then compared with the duration parameter to obtain the duration comparison result. Based on the early-stage fire scene and the duration comparison result, a graded early warning signal is generated and pushed to execute the graded early warning, so that the graded early warning can be executed in a timely manner, allowing for timely detection and extinguishing of the fire in the early-stage fire scene.

[0090] In one embodiment, temperature and time parameters of the fire scene are acquired, and a large database of early-stage fire parameters is constructed. Data from this database is input into an early-stage fire detection machine learning model to obtain a first feature parameter and a second feature parameter. The early-stage fire detection machine learning model is obtained through linear regression fitting of the data from the database. The first and / or second feature parameters are matched with a temperature rise rate reference parameter and / or a time reference parameter, and the early-stage fire phase at the fire scene is determined based on the matching results. The second feature parameter is compared with a duration parameter to obtain a duration comparison result. Based on the early-stage fire phase and the duration comparison result, a graded early warning signal is issued through the early-stage fire intelligent operation and maintenance internet platform. Graded early warnings are executed based on the graded early warning signal.

[0091] In one embodiment, the sliding window method is used to obtain the fire temperature parameters and time parameters, and the first feature parameter is calculated by linear regression fitting of the data within the window; the continuous cumulative time corresponding to the first feature parameter is the second feature parameter.

[0092] In one embodiment, a sliding window method is used to obtain the temperature and time parameters of the fire scene. Specifically, a 10-second sliding window is used to process the collected temperature parameters. The window is updated by sliding to the right every 2 seconds to ensure that adjacent windows overlap by 50%, thereby ensuring the continuity of the temperature change trend. The first characteristic parameter of the time period is calculated by linear regression fitting of the data within the window. At the same time, the second characteristic parameter of the temperature change trend corresponding to the current first characteristic parameter is simultaneously calculated.

[0093] In one embodiment, the average temperature rise rate for that period is calculated using linear regression fitting of the data within the window, which is the first characteristic parameter; simultaneously, the continuous cumulative time of the temperature change trend corresponding to the current first characteristic parameter is calculated, which is the second characteristic parameter. The second characteristic parameter is the key duration condition for issuing graded early warning signals.

[0094] In one embodiment, a graded early warning signal is issued based on the comparison results of the early stage of the fire and its duration; this provides characteristic parameters for accurate and real-time identification of early-stage fires.

[0095] In one embodiment, the temperature rise rate reference parameters include a first temperature rise rate, a second temperature rise rate, and a third temperature rise rate, and the time reference parameters include a first time reference parameter, a second time reference parameter, and a third time reference parameter. This allows the first and second characteristic parameters to be matched with more precise temperature rise rate reference parameters, thereby accurately matching the early stage of a fire and improving the accuracy of early fire warnings.

[0096] In one embodiment, the early fire stage includes: an initiation or smoldering stage, a slow growth stage, and a rapid growth stage. More precisely dividing the early fire stage into more detailed early fire stages allows for more accurate identification of the early fire stage at the fire location. This enables earlier fire identification and allows for the implementation of more appropriate firefighting measures based on the specific early fire stage, resulting in better firefighting effectiveness.

[0097] In one embodiment, when the first characteristic parameter and the second characteristic parameter meet the first temperature rise rate and the first time reference parameter, the fire is in the initial or smoldering stage; when the first characteristic parameter and the second characteristic parameter meet the second temperature rise rate and the second time reference parameter, the fire is in the slow growth stage; when the first characteristic parameter and the second characteristic parameter meet the third temperature rise rate and the third time reference parameter, the fire is in the rapid growth stage.

[0098] In one embodiment, when the first characteristic parameter meets the first temperature rise rate, the fire is in the initial or smoldering stage; when the first characteristic parameter meets the second temperature rise rate, the fire is in the slow growth stage; and when the first characteristic parameter meets the third temperature rise rate, the fire is in the rapid growth stage.

[0099] In one embodiment, when the second characteristic parameter meets the first time reference parameter, the fire is in the initial or smoldering stage; when the second characteristic parameter meets the second time reference parameter, the fire is in the slow growth stage; and when the second characteristic parameter meets the third time reference parameter, the fire is in the rapid growth stage.

[0100] In one embodiment, the temperature rise rate reference parameter and the time reference parameter are range values.

[0101] In one embodiment, the temperature rise rate reference parameter and the time reference parameter are both numerical values.

[0102] In one embodiment, the graded early warning signal is divided into early warning, fire alarm, and activation of the fire extinguishing system. The initial or smoldering stage corresponds to the early warning, the slow growth stage corresponds to the fire alarm, and the rapid growth stage corresponds to the activation of the fire extinguishing system.

[0103] In one embodiment, the response method for the early warning is a local audible and visual warning, a light flashing frequency, and a "fire risk" prompt pushed to the monitoring platform. The response method for the fire alarm is an enhanced audible and visual warning, an increased light flashing frequency, and a push of a "fire alarm" alarm. The response method for activating the fire extinguishing system is a high-level audible and visual alarm, which triggers the activation of the automatic sprinkler system.

[0104] In one embodiment, the duration parameter can be a specific parameter; specifically, the duration parameter is selected as 20s.

[0105] In one embodiment, the duration parameter can be adjusted for different environments, combustibles, etc.

[0106] In one embodiment, a tiered early warning signal is issued based on a comparison of the early stage of the fire and its duration. Specifically, if the fire is determined to be in the initial or smoldering stage and the second characteristic parameter is ≥20s, a tiered early warning signal is issued; if the fire is determined to be in the slow growth stage and the second characteristic parameter is ≥20s, a tiered fire alarm signal is issued; and if the fire is determined to be in the rapid growth stage and the second characteristic parameter is ≥20s, a tiered early warning signal is issued to activate the fire suppression system. Of course, the duration parameter can be adjusted according to different needs, historical experience, or different environments.

[0107] In one embodiment, historical fire data is input into a fire spread law function to obtain temperature rise rate comparison parameters and time comparison parameters.

[0108] In one embodiment, the historical fire data can be either real historical fire data or historical data obtained from experiments. The historical fire data can be referred to as an early-stage fire database.

[0109] In one embodiment, the fire spread law function is obtained by linear regression fitting:

[0110] y=kx+b

[0111] Where: y is the temperature parameter, x is the time parameter, and k is the first characteristic parameter.

[0112] In one embodiment, k is the average temperature rise rate, which is the first characteristic parameter.

[0113] In one embodiment, a third characteristic parameter and a heat release rate comparison parameter are also included. The third characteristic parameter is obtained by linear regression fitting of the temperature parameter and the time parameter. The first characteristic parameter, the second characteristic parameter, and the third characteristic parameter are matched with the temperature rise rate comparison parameter, the time comparison parameter, and the heat release rate comparison parameter, and the early stage of the fire is determined based on the matching results. The third characteristic parameter is the heat release rate.

[0114] In one embodiment, extensive experimental studies have revealed that for early-stage fires, the relative position of the fire source and detection devices, such as sprinklers and thermocouples, directly affects heat transfer efficiency, leading to significant differences in temperature monitoring data and detection response characteristics: when the fire source is directly below or near a sprinkler, the heat transfer path is short and attenuation is minimal; when the fire source is between two sprinklers, the heat needs to diffuse outwards before being detected, resulting in a longer transfer path; and when the fire source is between four sprinklers, the heat diffusion range is wider and attenuation is more pronounced. Through extensive experimental research, a formula relating fire size to sprinkler distance has been obtained:

[0115]

[0116] In the formula: r is the distance from the nozzle to the ignition point, in meters;

[0117] Q represents the fire scale in the initial stage of ignition, in kW;

[0118] H is the distance between the fuel surface and the ceiling, in meters (m).

[0119] The radius coefficient;

[0120] T is the gas temperature when the nozzle is activated, in °C;

[0121] The ambient temperature is in °C.

[0122] This formula is mainly applicable to small-scale fires in the early stages of a fire. It utilizes parameters such as heat flow meter to monitor radiant heat, thermocouples to monitor temperature, stopwatch to monitor fire growth time, a meter stick to measure the horizontal radius r between the actual fire source and the sprinkler head, and cameras to monitor ignition time, fire start time, and sprinkler head activation time. The correlation coefficient value in the formula relating the fire source initiation point to the distance of the sprinkler head is then determined using these monitored parameters.

[0123] Heat flow meters were installed on both sides of the burning material and connected to a multi-channel touch-screen data logger to record changes in radiant heat and measure Q. Therefore, when acquiring historical data to fit temperature rise rate and time control parameters, and when experimentally verifying the early fire spread pattern and fire early warning alarm method based on the Internet platform, different distances between the burning material and the fire source were considered.

[0124] In one embodiment, the combustible material can be ignited by means of cigarette butt ignition, heating wire ignition, electric spark ignition, welding slag ignition, etc.

[0125] In one embodiment, after acquiring parameters such as temperature and time, preprocessing is required. Specifically, temperature-time parameter sequence data for early-stage fires are collected through numerous fire experiments and on-site monitoring equipment. A Kalman filter algorithm combined with linear interpolation is used to complete missing data, resulting in a standardized temperature parameter sequence. To address potential noise and missing values ​​in the collected data, a three-step process of "denoising-completion-labeling" is employed. The specific process is as follows: Denoising: A Kalman filter algorithm is used; Completion: For single-point missing data occurring during transmission, linear interpolation is used to complete the data.

[0126] In one embodiment, multiple experiments were conducted with different combustibles and varying distances between the combustibles and the fire source to obtain temperature and time parameters. Curves were then fitted to show the changes in temperature and the third characteristic parameter over time during the early stages of a fire to ensure the reliability of the fitting results. The temperature parameter curve over time consists of three straight line segments: the initial or smoldering stage, the slow growth stage, and the rapid growth stage—essentially the subdivisions of the early fire stage. The fitting results are shown in the figure. The curve representing the temperature parameter over time is named "temperature," represented by the red broken line in the figure. This red broken line is composed of three diagonal straight lines—light red, medium red, and dark red—connected to represent the initial or smoldering stage, the slow growth stage, and the rapid growth stage, respectively (y1, y2, and y3). As shown in the figure, the slope of the temperature parameter curve over time is the first characteristic parameter, which can be represented by k. The first characteristic parameter gradually increases as the fire progresses. The third characteristic parameter, the heat release rate (HRR), remains largely unchanged or shows a slight increase during the initial or smoldering stage, indicating a small fire. During the slow growth stage, the third characteristic parameter increases slightly at a relatively slow rate, and the fire size does not expand further. In the rapid growth stage, the third characteristic parameter increases significantly, and the fire size increases exponentially, along with the average temperature rise rate. These results illustrate the early-stage fire spread pattern. Experimental studies reveal that early-stage fires develop through three stages: the initial or smoldering stage, the slow growth stage, and the rapid growth stage. Understanding the early-stage fire spread pattern can help improve fire detection methods.

[0127] Specifically, such as Figures 3 to 6 The figure shows the curves of the early-stage temperature parameter and the third characteristic parameter of a fire over time when the combustible material is a stack of wood, made of white pine, classified as 1A, and weighing approximately 24 kg. Figure 3 The fitting results are for the burning material being 0m away from the fire source, where y1=0.16x+26.53, y2=0.55x+11.95, and y3=1.35x-36.25; Figure 4 The fitting results are for when the burning material is 1.5m away from the fire source, where y1=0.15x+26.54, y2=0.48x+13.49, and y3=1.06x-24.51; Figure 5 The fitting results are when the burning material is 2.12m away from the fire source, where y1=0.16x+26.53, y2=0.23x+23.60, and y3=1.19x-62.94; Figure 6 The values ​​are the fitting results when the burning material is 3m away from the fire source, where y1=0.11x+26.59, y2=0.30x+15.44, and y3=1.19x-74.45.

[0128] like Figures 7 to 10As shown, the combustible material is a standard combustible: it consists of a cardboard box and polystyrene plastic cups. The cardboard box measures 500 mm × 500 mm × 500 mm. Each cardboard box contains 125 plastic cups, arranged in 5 layers with 25 plastic cups per layer. Figure 7 The fitting results are for the burning material being 0m away from the fire source, where y1=0.14x+26.55, y2=0.27x+20.87, and y3=1.63x-102.29; Figure 8 The fitting results when the burning material is 1.5m away from the fire source are as follows: y1=0.13x+26.57, y2=0.28x+19.47, y3=1.61x-105.91; Figure 9 The fitting results are when the burning material is 2.12m away from the fire source, where y1=0.11x+26.58, y2=0.26x+19.49, and y3=1.26x-73.87; Figure 10 The values ​​are the fitting results when the burning material is 3m away from the fire source, where y1=0.09x+26.60, y2=0.19x+19.13, and y3=1.08x-95.76.

[0129] Based on the analysis of heat release rate variation, combustion reaction process, and average temperature rise rate (K) from a large amount of fire experimental data, and combined with the temperature parameter-time parameter sequence fitting results at different fire source distances, the characteristic parameters of different development stages of early-stage fires are now clarified as follows:

[0130] (1) Initial stage

[0131] Table 1. Characteristic parameters of the initial stage of timber stacking:

[0132] Feature parameters scope average value First characteristic parameter 0.11-0.16℃ / s 0.135℃ / s Third characteristic parameter ≤25kw 20kW Second characteristic parameter 30-60s 45s

[0133] Table 2. Characteristic parameters of standard combustors in the initial stage:

[0134] Feature parameters scope average value First characteristic parameter 0.11-0.14℃ / s 0.125℃ / s Third characteristic parameter ≤15kw 10kW Second characteristic parameter 30-50s 40s

[0135] Combustion characteristics:

[0136] The combustion material is partially ignited, with only a small area on the surface participating in combustion. There is no obvious flame, but smoke is continuously produced. The heat release rate increases slowly, and the overall state is in a low heat release state. The temperature rises slowly over time, with the temperature difference from the ambient temperature within 10℃. The further away from the fire source, the lower the temperature rise rate. The average temperature rise rate is within 0.2℃ / s, with average temperature rise rates of 0.135℃ / s and 0.125℃ / s, respectively.

[0137] (2) Slow growth stage

[0138] Table 3. Characteristic parameters of the slow growth stage of timber stacks:

[0139] Feature parameters scope average value First characteristic parameter 0.23-0.55℃ / s 0.39℃ / s Third characteristic parameter 20-40kw 30kW Second characteristic parameter 20-40s 30s

[0140] Table 4. Characteristic parameters of the slow growth phase of standard combustors:

[0141] Feature parameters scope average value First characteristic parameter 0.26-0.41℃ / s 0.385℃ / s Third characteristic parameter 10-30kw 20kW Second characteristic parameter 30-50s 40s

[0142] Combustion characteristics:

[0143] A large amount of smoke was generated, flames appeared, the heat release rate continued to increase, the heat release was significantly higher than in the initial stage, the temperature rose faster over time, the temperature difference with the ambient temperature was within 20℃, the average temperature rise rate was within 0.6℃ / s, and the average temperature rise rates were 0.39℃ / s and 0.385℃ / s, respectively.

[0144] (3) Rapid growth phase

[0145] Table 5. Characteristic parameters of the rapid growth stage of timber stacks:

[0146] Feature parameters scope average value First characteristic parameter 1.06-1.35℃ / s 1.205℃ / s Third characteristic parameter 15-130kw 75kw Second characteristic parameter 50-110s 80s

[0147] Table 6. Characteristic parameters of the rapid growth phase of standard combustors:

[0148] Feature parameters scope average value First characteristic parameter 1.06-1.63℃ / s 1.345℃ / s Third characteristic parameter 60-300kw 180kW Second characteristic parameter 40-80s 60s

[0149] Combustion characteristics:

[0150] The burning material burns over a large area, with a flame height of ≥0.5m, a heat release rate of 300kW, releasing a large amount of heat, and a rapid temperature rise, with a temperature difference of 80℃ from the ambient temperature. The average temperature rise rate is above 1℃ / s. The average temperature rise rates are 1.205℃ / s and 1.345℃ / s, respectively.

[0151] like Figure 11 The diagram shows the fire development process in the early stages of a fire. The points of origin of the early fire, the activation of the (early) intelligent automatic sprinkler system, and the activation of the traditional automatic sprinkler system can be roughly represented as follows: Figure 11 As shown in the diagram. Specifically, based on extensive experimental analysis and statistical data, small fires with a scale between 0.25 and 0.50 MW are defined as early-stage fires, and further divided into initiation or smoldering stages, slow-growth stages, and rapid-growth stages based on the rate of temperature rise and heat release. Extensive fire statistics indicate that the temperature at which the first conventional automatic sprinkler head is activated is between 100°C and 250°C. The fire temperature at the activation of an automatic sprinkler head is far below the flashover temperature of 600°C. Therefore, the early stage of a fire can be considered to be shortly after the initial initiation stage has entered the development stage, before reaching the flashover stage, or only in the room where the fire originated, but before the fire compartment has experienced flashover.

[0152] Figure 12This diagram illustrates the results of detecting and responding to early-stage fires. It shows the time it took for the smart sprinkler head to activate after sensing an early-stage fire, as obtained from the experiment. It can also be compared with... Figure 11 Based on the analysis, the timing of the smart sprinkler head's action is the same as the start-up time of the (early morning) smart automatic sprinkler system, while the timing of the sprinkler head's action is the same as the start-up time of the traditional automatic sprinkler system.

[0153] Figure 13 The statistical results for the average temperature rise rate, or first characteristic parameter, of the three stages of the early-stage fire spread process are as follows: The average temperature rise rate in the initial stage is approximately 0.13℃ / s, with a heat release rate below 20kW and an average duration of approximately 40s; the average temperature rise rate in the slow growth stage is approximately 0.39℃ / s, with a heat release rate below 30kW and an average duration of approximately 40s; and the average temperature rise rate in the rapid growth stage is approximately 1.3℃ / s, with a heat release rate below 200kW and an average duration of approximately 70s.

[0154] In one embodiment, the determination logic of "k matching the early fire stage and t≥20s triggering response" provides accurate and real-time feature parameter support.

[0155] In one embodiment, such as Figure 20 As shown, a fire early warning alarm device 10 is provided, which includes: a parameter acquisition module 11, a linear regression fitting module 12, a matching module 13, a comparison module 14, an early warning issuance module 15, and an early warning execution module 16. The parameter acquisition module 11 is used to acquire the temperature parameters and time parameters of the fire scene; the linear regression fitting module 12 is used to perform linear regression fitting on the temperature parameters and time parameters to obtain a first characteristic parameter and a second characteristic parameter; the matching module 13 matches the first characteristic parameter and / or the second characteristic parameter with the temperature rise rate comparison parameter and / or the time comparison parameter and obtains the matching result; the comparison module 14 compares the second characteristic parameter with the duration parameter and obtains the duration comparison result; the early warning issuance module 15 determines the fire stage of the fire scene based on the matching result and the duration comparison result and issues a graded early warning signal; the early warning execution module 16 executes the graded early warning based on the graded early warning signal.

[0156] In one embodiment, the early fire spread pattern and fire early warning alarm device 10 further includes: a comparison parameter fitting module, which inputs historical fire data into the fire spread pattern function to obtain a temperature rise rate comparison parameter and a time comparison parameter.

[0157] This device embodiment can be implemented in conjunction with the implementation methods described above. The relevant technical details mentioned in the implementation methods of the above embodiments remain valid in the implementation methods of this method embodiment, and will not be repeated here to avoid repetition.

[0158] In one embodiment, a system for detecting the early spread of fires and providing early warning alarms is provided. This system employs the aforementioned method for detecting the early spread of fires and providing early warning alarms based on an internet platform. It includes: a temperature acquisition mechanism 20, a time acquisition mechanism, a controller, and a sprinkler assembly 30. The temperature acquisition mechanism 20 acquires temperature parameters; the time acquisition mechanism acquires time parameters; the controller acquires the temperature and time parameters of the fire scene; linear regression fitting is performed on the temperature and time parameters to obtain a first characteristic parameter and a second characteristic parameter; the first and / or second characteristic parameters are matched with a temperature rise rate comparison parameter and / or a time comparison parameter; based on the matching result, the early fire stage at the fire scene is determined; the second characteristic parameter is compared with a duration parameter to obtain a duration comparison result; and a graded early warning signal is issued based on the early fire stage and the duration comparison result. The sprinkler assembly 30 executes the graded early warning based on the graded early warning signal.

[0159] In one embodiment, the temperature acquisition mechanism 20 can be a thermocouple, a temperature sensor, etc., and the nozzle assembly 30 can be multiple water spray nozzles.

[0160] In one embodiment, the controller can be a smart automatic sprinkler system. Research has established a database of early-onset fires caused by cigarette butt ignition, heating wire ignition, electric spark ignition, and welding slag ignition. By uniformly arranging temperature sensors in a spatial array, the temperature field changes of different temperature sensors in the same space are monitored in real time. Based on the research model and database, early-onset fires are identified and warnings are issued. Simultaneously, the smart automatic sprinkler system is activated electrically to spray water and extinguish early-onset fires.

[0161] In one embodiment, the fire site can be a test space, which is a ventilated but enclosed space without forced exhaust, with a flat roof measuring 13,500 mm × 10,000 mm and a height of 3,000 mm; the air inlet measures 1,600 mm × 2,000 mm and has an area of ​​3.2 m², and can use natural ventilation to exhaust the smoke.

[0162] The arrangement of sprinkler heads and sensors is as follows Figure 21 As shown, 12 identical nozzle assemblies 30 are installed below the ceiling, arranged in a square pattern, using a balanced water supply network, with a nozzle spacing of 3200 mm. Five temperature acquisition mechanisms 20 (thermocouples) are arranged: four are located next to the heat-sensitive elements of nozzles 6, 10, 11, and 7, numbered TC1~TC4; the remaining one (not shown) is located in the center of the easternmost side of the test space, serving as the ambient temperature measurement point, numbered TC5; one anemometer 40 is also present. A camera and thermal imager are set up at the test site, horizontally positioned 5000 mm from the fire source and 900 mm above the ground. Figure 21As shown, X is the distance between the two nozzle assemblies in the vertical direction, and Y is the distance between the two nozzle assemblies in the horizontal direction.

[0163] like Figure 21 As shown, Figure 21 The solid black circle icon "●" indicates nozzle assembly 30; Figure 21 The "T" icon with a rectangle in the middle represents the temperature acquisition mechanism 20; Figure 21 The circled numbers represent the number of the temperature acquisition unit 20; the numbers with rectangles in the figure represent the location of the test fire source. Specifically, the number "1" with a rectangle represents the fire source directly below one nozzle, and the number "2" with a rectangle represents the fire source directly below the center of four nozzles.

[0164] In one embodiment, a database of early-onset fires based on cigarette butt ignition, heating wire ignition, electric spark ignition, and welding slag ignition was constructed. Temperature sensors were uniformly arrayed in space to monitor temperature field changes in the same space in real time. Based on the research model and database, early-onset fires were identified and early warning alarms were issued. Simultaneously, a smart automatic sprinkler system was activated to electrically spray water to extinguish early-onset fires. The model can refer to early-onset fire spread patterns and fire early warning alarm methods based on an internet platform, or a machine learning model for early-onset fire detection. The database can refer to an early-onset fire database, a large database of early-onset fire parameters, etc.

[0165] In one embodiment, taking a test case as an example, the test uses a thermoelectric dual-drive automatic sprinkler head. At the beginning of the test, the temperature at each sprinkler head is detected simultaneously. According to the present invention, when the judgment condition is met, the intelligent automatic sprinkler fire extinguishing system can be linked to control the heating module at the sprinkler head to start the sprinkler head, so as to realize the judgment and extinguishing of early fire.

[0166] The combustible material is a stack of white pine wood, with external dimensions of 500 mm × 500 mm × 480 mm. The stack is placed on a support, with the bottom of the stack 400 mm from the ground. 400 g of pleated A4 paper is placed under the stack as an accelerant. Six sheets of A4 paper are joined together to form a cuboid of 400 mm × 400 mm × 80 mm, and the remaining pleated A4 paper is placed inside. The fire source arrangement is shown in the figure.

[0167] (1) Test with the burning material placed directly below a nozzle.

[0168] To ensure the reliability of the experiment, three sets of parallel experiments were conducted, and the experimental data are as follows:

[0169] Table 7 Results of tests with burning material placed directly below a nozzle:

[0170] Test number Start-up nozzle serial number nozzle start-up time Average temperature rise rate (°C / s) Duration TC1 maximum temperature / °C Timber stack mass loss 1 No. 6 48s 1.4 15s 70.4 3.0% 2 No. 6 49s 1.3 15s 66.1 4.1% 3 No. 6 44s 1.2 15s 83.3 1.2%

[0171] like Figures 14 to 16 As shown, the temperature changes over time during experiments 1, 2, and 3 are curves.

[0172] (2) Fire extinguishing test with the burning material placed directly below the center of the four nozzles.

[0173] Table 8 Results of fire extinguishing test with burning material placed directly below the center of four nozzles:

[0174] Test number Start-up nozzle serial number nozzle start-up time Average temperature rise rate (°C / s) Duration TC maximum temperature / °C Timber loss 1 No. 10 No. 7 No. 6 2min 10s 2min 13s 2min 18s 1.1 15s 80.4 (TC2) 10.3% 2 No. 11 No. 6 No. 7 No. 10 1min49s 1min53s 1min55s 1min58s 1.3 15s 71.3 (TC3) 9.9% 3 No. 6, No. 11, No. 7, No. 10 2min03s 2min06s 2min08s 2min11s 1.2 15s 73.1 (TC1) 9.6%

[0175] like Figures 17 to 19 The figure shows the temperature change curves over time during experiments 1, 2, and 3.

[0176] This system embodiment can be implemented in conjunction with the implementation methods described above. The relevant technical details mentioned in the implementation methods of the above embodiments remain valid in the implementation methods of this method embodiment, and will not be repeated here to avoid repetition.

[0177] In one embodiment, an early-onset fire intelligent operation and maintenance system is also provided to detect early-onset fires and perform intelligent operation and maintenance. The early-onset fire intelligent operation and maintenance system is equipped with the aforementioned early-onset fire spread pattern and fire early warning alarm device.

[0178] In one embodiment, an intelligent operation and maintenance internet platform for early-stage fires is also provided to detect early-stage fires and perform intelligent operation and maintenance. The intelligent operation and maintenance internet platform for early-stage fires adopts the above-mentioned internet platform-based method for early-stage fire spread patterns and fire early warning alarms.

[0179] In one embodiment, the early-stage fire intelligent operation and maintenance internet platform sends data acquisition instructions to the front-end sensing devices to obtain temperature and time parameters of the fire scene, and uploads them to the early-stage fire intelligent operation and maintenance internet platform; the early-stage fire intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature and time parameters to obtain a first feature parameter and a second feature parameter; and matches the first feature parameter and / or the second feature parameter with the temperature rise rate comparison parameter and / or the time comparison parameter, and determines the early-stage fire scene based on the matching result; compares the second feature parameter with the duration parameter to obtain the duration comparison result, and uploads it to the early-stage fire intelligent operation and maintenance internet platform; and generates a graded early warning signal based on the early-stage fire scene and the duration comparison result, and pushes the graded early warning signal to execute the graded early warning.

[0180] In one embodiment, one or more of the following parameters are uploaded to the early-stage fire intelligent operation and maintenance internet platform to construct an early-stage fire parameter database: temperature parameter, time parameter, first characteristic parameter, second characteristic parameter, matching result, temperature rise rate comparison parameter, or time comparison parameter, and graded early warning signal.

[0181] In one embodiment, the early-stage fire intelligent operation and maintenance internet platform includes: a front-end sensing device, which acquires temperature and time parameters of the fire scene according to the data acquisition instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and uploads them to the early-stage fire intelligent operation and maintenance internet platform.

[0182] In one embodiment, the early-stage fire intelligent operation and maintenance internet platform further includes: a back-end processing device, which performs linear regression fitting on the received temperature parameters and time parameters to obtain a first feature parameter and a second feature parameter; matches the first feature parameter and / or the second feature parameter with a temperature rise rate comparison parameter and / or a time comparison parameter, and determines the early-stage fire location based on the matching result; compares the second feature parameter with a duration parameter to obtain a duration comparison result, and then uploads it to the early-stage fire intelligent operation and maintenance internet platform; and generates a graded early warning signal based on the early-stage fire location and the duration comparison result, and pushes the graded early warning signal to execute the graded early warning.

[0183] In one embodiment, the front-end sensing device includes a temperature acquisition mechanism 20 and a time acquisition mechanism, wherein the temperature acquisition mechanism 20 is used to acquire temperature parameters and the time acquisition mechanism is used to acquire time parameters.

[0184] In one embodiment, the front-end sensing device is electrically connected to the temperature acquisition mechanism 20 and the time acquisition mechanism. The front-end sensing device controls the temperature acquisition mechanism 20 to acquire temperature parameters and controls the time acquisition mechanism to acquire time parameters.

[0185] In one embodiment, one or more early-stage fire spread pattern and fire early warning alarm devices 10 can establish a wired or wireless communication connection with a host computer, such as an early-stage fire intelligent operation and maintenance internet platform. Optionally, the wired communication method can be, but is not limited to, at least one of the following: power line communication, optical fiber power line communication, the Internet, coaxial cable, and telephone line, etc.; optionally, the wireless communication method can be, but is not limited to, at least one of the following: infrared, Bluetooth, Z-wave, NFC, ZigBee, and WiFi, etc.

[0186] In one embodiment, the temperature acquisition mechanism 20 and the time acquisition mechanism acquire temperature and time parameters of the fire scene, transmit them to the controller via ZigBee / ZWave wireless protocol, and upload them to the early-stage fire intelligent operation and maintenance internet platform via WiFi / wireless communication; the early-stage fire intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature and time parameters to obtain a first feature parameter and a second feature parameter; and matches the first feature parameter and / or the second feature parameter with the temperature rise rate comparison parameter and / or the time comparison parameter, and determines the early-stage fire location based on the matching result; compares the second feature parameter with the duration parameter to obtain the duration comparison result, and uploads it to the early-stage fire intelligent operation and maintenance internet platform; and generates a graded early warning signal based on the early-stage fire location and the duration comparison result, and pushes the graded early warning signal to execute the graded early warning.

[0187] In one embodiment, the temperature acquisition mechanism 20 and the time acquisition mechanism acquire temperature and time parameters of the fire scene and transmit them to the controller via the ZigBee / ZWave wireless protocol. The controller performs linear regression fitting on the received temperature and time parameters to obtain a first feature parameter and a second feature parameter; and matches the first feature parameter and / or the second feature parameter with the temperature rise rate comparison parameter and / or the time comparison parameter, and determines the early stage of the fire scene based on the matching result; compares the second feature parameter with the duration parameter to obtain the duration comparison result, and generates a graded early warning signal based on the early stage of the fire scene and the duration comparison result, and pushes the graded early warning signal to execute the graded early warning.

[0188] Furthermore, one embodiment of the present invention provides a computer storage medium, and another embodiment of the present invention provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the steps of the above-mentioned method for early fire spread patterns and fire early warning alarms based on an Internet platform, and achieve the same technical effect.

[0189] This invention also provides a computer program product that stores a program or instructions. When the program or instructions are executed by a processor, they implement the steps of the above-mentioned method for early fire spread patterns and fire early warning alarms based on an Internet platform, and achieve the same technical effect.

[0190] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A method for early fire spread patterns and fire early warning alarms based on an internet platform, characterized in that, Include: The front-end sensing device acquires the temperature and time parameters of the fire scene according to the data collection instructions issued by the early-stage fire intelligent operation and maintenance internet platform, and uploads them to the early-stage fire intelligent operation and maintenance internet platform. The early-stage fire intelligent operation and maintenance internet platform performs linear regression fitting on the received temperature parameters and time parameters to obtain the first feature parameter and the second feature parameter; as well as, The first characteristic parameter and / or the second characteristic parameter are matched with the temperature rise rate comparison parameter and / or the time comparison parameter, and the early stage of the fire is determined based on the matching result. The second feature parameter is compared with the duration parameter to obtain the duration comparison result; Based on the comparison results of the early stage of the fire and its duration, a graded early warning signal is generated and pushed out to implement the graded early warning.

2. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 1, characterized in that, The temperature and time parameters of the fire scene are obtained, and a large database of early-stage fire parameters is constructed. The data in the early fire parameter database is input into the early fire detection machine learning model to obtain the first feature parameter and the second feature parameter. The early fire detection machine learning model is obtained by linear regression fitting of the data in the early fire parameter database. The first characteristic parameter and / or the second characteristic parameter are matched with the temperature rise rate comparison parameter and / or the time comparison parameter, and the early stage of the fire is determined based on the matching result. The second feature parameter is compared with the duration parameter to obtain the duration comparison result; Based on the comparison results of the early stage of fire and its duration, the aforementioned graded early warning signal is issued through the early fire intelligent operation and maintenance internet platform. A tiered warning system will be implemented based on the aforementioned tiered warning signals.

3. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 1 or 2, characterized in that, The temperature and time parameters of the fire scene are obtained using the sliding window method. The first feature parameter is calculated by linear regression fitting of the temperature and time parameter data within the window. The second feature parameter is obtained by counting the continuous cumulative time corresponding to the first feature parameter.

4. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 1 or 2, characterized in that, The temperature rise rate comparison parameters include a first temperature rise rate, a second temperature rise rate, and a third temperature rise rate, and the time comparison parameters include a first time comparison parameter, a second time comparison parameter, and a third time comparison parameter.

5. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 4, characterized in that, The early fire phase includes: the initial or smoldering phase, the slow growth phase, and the rapid growth phase.

6. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 5, characterized in that, When the first characteristic parameter and / or the second characteristic parameter meet the first temperature rise rate and / or the first time reference parameter, the fire is in the initial or smoldering stage; when the first characteristic parameter and / or the second characteristic parameter meet the second temperature rise rate and / or the second time reference parameter, the fire is in the slow growth stage; when the first characteristic parameter and / or the second characteristic parameter meet the third temperature rise rate and / or the third time reference parameter, the fire is in the rapid growth stage.

7. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 1, characterized in that, By inputting historical fire data into the fire spread law function, the temperature rise rate comparison parameter and the time comparison parameter are obtained.

8. The method for early fire spread patterns and fire early warning alarm based on an internet platform according to claim 7, characterized in that, The fire spread law function is obtained by fitting the linear regression: y=kx+b Where: y is the temperature parameter, x is the time parameter, and k is the first characteristic function.

9. A fire early warning and alarm device for early fire spread patterns, characterized in that, Include: A parameter acquisition module, which is used to acquire temperature and time parameters of the fire scene; A linear regression fitting module is used to perform linear regression fitting on the temperature parameter and the time parameter to obtain a first feature parameter and a second feature parameter. The matching module matches the first feature parameter and / or the second feature parameter with the temperature rise rate reference parameter and / or the time reference parameter and obtains the matching result; The comparison module compares the second feature parameter with the duration parameter to obtain a duration comparison result; The early warning issuing module determines the early stage of the fire based on the matching results and the duration comparison results, and issues graded early warning signals. The early warning execution module executes tiered early warnings based on the tiered early warning signals.

10. The fire early warning and alarm device according to claim 9, characterized in that, Also includes: The comparison parameter fitting module inputs historical fire data into the fire spread law function to obtain the temperature rise rate comparison parameter and the time comparison parameter.

11. A system for detecting the early spread of fires and providing early warning alarms, employing the method for detecting the early spread of fires and providing early warning alarms based on an internet platform as described in any one of claims 1 to 8, characterized in that... Include: A temperature acquisition mechanism, wherein the temperature acquisition mechanism is used to acquire temperature parameters; A time acquisition mechanism, wherein the time acquisition mechanism is used to acquire time parameters; The controller acquires the temperature and time parameters of the fire scene; performs linear regression fitting on the temperature and time parameters to obtain a first feature parameter and a second feature parameter; matches the first feature parameter and / or the second feature parameter with a temperature rise rate reference parameter and / or a time reference parameter, determines the early stage of the fire scene based on the matching result, compares the second feature parameter with a duration parameter to obtain a duration comparison result, and issues a graded early warning signal based on the early stage of the fire scene and the duration comparison result. The nozzle assembly executes a graded warning based on the graded warning signal.

12. An intelligent operation and maintenance system for early-stage fires, characterized in that, The early-onset fire intelligent operation and maintenance system is used to detect early-onset fires and perform intelligent operation and maintenance. It is equipped with the early-onset fire spread pattern and fire early warning alarm device according to claim 9.

13. An intelligent operation and maintenance internet platform for early-stage fires, characterized in that, The intelligent operation and maintenance internet platform for early-stage fires, used to detect early-stage fires and perform intelligent operation and maintenance, adopts the early-stage fire spread pattern and fire early warning alarm method based on the internet platform as described in any one of claims 1 to 8.

Images