Intelligent remote fire-fighting monitoring system based on big data of Internet of Things

Abstract

The invention discloses an intelligent remote fire-fighting monitoring system based on Internet of Things big data, which comprises a data acquisition layer and a platform layer, and is characterized in that the data acquisition layer acquires data through a sensor and then transmits the acquired data to the platform layer; in an early warning temperature threshold value dynamic generation module, through analyzing temperature data over the years in an area where a building is located, a low temperature range, a medium temperature range and a high temperature range are divided, and aging factors of a temperature sensor are combined; the initial temperature threshold value for finding potential fire risks and triggering early warning and the initial temperature change rate threshold value for triggering early warning through temperature change are dynamically adjusted according to the temperature, humidity and temperature difference conditions, an original static threshold value is replaced, dynamic adjustment is achieved, and therefore the probability of triggering false alarm is reduced.

Description

Technical Field

[0001] The present invention relates to the field of fire monitoring technology, and in particular to an intelligent remote fire monitoring system based on Internet of Things big data. Background Art

[0002] Intelligent fire safety monitoring system is a system that uses advanced technology and equipment to monitor, prevent and deal with fire and other safety risks. These systems usually integrate various sensors, monitoring equipment, data analysis algorithms and communication technologies to improve the efficiency and accuracy of fire detection, early warning and response, aiming to provide faster and more reliable response to protect people's lives and property.

[0003] However, traditional fire safety monitoring systems are mostly static detection systems, relying on the monitoring sampling mode and alarm mode preset by the manufacturer to detect user data. When abnormal data is detected, the alarm is triggered. However, in daily life, due to the different temperature and humidity differences in the monitored buildings at different times, and the temperature measurement deviation of the temperature sensor increases with the increase in usage time, this monitoring system that relies on static data can easily trigger false alarms. Summary of the invention

[0004] In order to solve the technical problems existing in the background technology, the present invention proposes an intelligent remote fire monitoring system based on Internet of Things big data.

[0005] The present invention proposes a smart remote fire monitoring system based on Internet of Things big data, comprising: Data collection layer and platform layer, including: Data collection layer: collects data through sensors and then transmits the collected data to the platform layer; The platform layer includes: Data processing module, for data cleaning and data conversion; A data storage module stores the data after data cleaning and data conversion; Early warning temperature threshold dynamic generation module: dynamic threshold is used for the temperature threshold that detects potential fire risk and triggers early warning, and early warning temperature dynamic threshold is generated; Early warning and alarm generation module: generates early warning levels and alarm levels through data analysis based on the early warning temperature dynamic threshold generated by the early warning temperature threshold dynamic generation module; Fire prediction model generation module: Based on the warning temperature dynamic threshold generated by the warning temperature threshold dynamic generation module and combined with historical data analysis, a fire prediction model is constructed.

[0006] Preferably, in the data acquisition layer, the sensors include: smoke sensors, temperature sensors, flame detectors, combustible gas detectors, and fire water system pressure sensors.

[0007] Preferably, in the early warning temperature threshold dynamic generation module, the temperature threshold for detecting potential fire risks and triggering early warning is generated using a dynamic threshold method, as follows: When a building is remotely monitored for fire, the temperature data of the area in the building is obtained. The temperature data of the past years is set as the temperature data of n years. The average temperature of each day is calculated based on the temperature data of n years. The average temperature data of the same day in the past years are added and then divided by n to obtain the historical average temperature of each day. Arrange these historical average temperature data in date order to form a complete annual average temperature data set; Set two temperature values ​​as M1 and M2, and M1<M2; In the annual mean temperature dataset, the historical mean temperatures for each day are examined sequentially; When the historical average temperature is lower than M1, record this date as the start date of the low temperature range, and continue to check subsequent dates. Once a date with a historical average temperature greater than M1 is encountered, the previous date is taken as the end date of the low temperature range; When encountering a date with a historical average temperature greater than M1, record it as the start date of the medium temperature range, and continue to check subsequent dates. Once encountering a date with a historical average temperature less than M1 or greater than M2, determine the previous date as the end date of the medium temperature range; When encountering a date with a historical average temperature higher than M2, record it as the start date of the high temperature range, and continue to check subsequent dates. When encountering a date with an average temperature lower than M2, take the previous date as the end date of the high temperature range; The third initial temperature threshold value of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained as K1, and the initial temperature change rate threshold value for triggering an early warning by temperature change is set as V1; When the current date is within the low temperature range, the initial temperature threshold for detecting potential fire risks and triggering an early warning is reduced to Y times of K1, and the initial temperature change rate threshold for triggering an early warning is reduced to Y times of V1; When the current date is within the medium temperature range, the values ​​of K1 and V1 remain unchanged; When the current date is within the high temperature range, the initial temperature threshold for detecting potential fire risks and triggering an early warning rises to X times of K1, and the initial temperature change rate threshold for triggering an early warning due to temperature changes rises to X times of V1.

[0008] Preferably, the third initial temperature threshold value K1 for the temperature sensor to detect potential fire risk and trigger an early warning is obtained in the following manner: The second initial temperature threshold value K0 of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained; Condition 1: When the humidity value measured by the humidity sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; Condition 2: On the same day, when the temperature difference measured by the temperature sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; When condition 1 or condition 2 is met, K1=ZK0; When conditions 1 and 2 are met at the same time, K1=Z(ZK0); When condition 1 or condition 2 is not met, K1=K0.

[0009] Preferably, the second initial temperature threshold K0 of the temperature sensor for detecting a potential fire risk and triggering an early warning is obtained in the following manner: Assume that the first initial temperature threshold of the temperature sensor for detecting potential fire risk and triggering an early warning is K; Start timing from when the temperature sensor is put into use, and record its use time t as the aging time; By obtaining the technical specifications of the temperature sensor, the measurement deviation of the temperature sensor per unit time during the initial use phase is obtained. , represents the temperature measurement deviation caused by aging per unit time; Calculation of measurement deviations due to aging of temperature sensors , the formula is: ; Where k is the aging coefficient. By taking temperature, humidity, usage time and usage frequency as aging factors, a theoretical analysis method is used to establish a theoretical relationship model between the aging coefficient and each aging factor. The actual environmental parameters are substituted into the theoretical model to calculate the aging coefficient k. Then K0= K+ .

[0010] Preferably, in the fire prediction model generation module, the fire prediction model is constructed as follows: Obtain historical fire accident data, historical daily monitoring data, and warning temperature dynamic thresholds, and use the random forest classification algorithm to build a fire prediction model to predict the probability of fire. When the predicted probability of fire is greater than the set threshold, an alarm is issued.

[0011] Preferably, in the early warning and alarm generation module, the early warning level and the alarm level are generated as follows: When the temperature reaches the temperature threshold that detects potential fire risks and triggers an early warning, a first-level fire warning is issued; When the temperature reaches the temperature threshold for detecting potential fire risks and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold, a secondary fire early warning is issued; When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration continues to rise within Q minutes and exceeds Wppm, a level 3 fire warning will be issued; If the concentration of combustible gas reaches 10%-15% of the lower explosion limit, a level 3 fire warning will be issued; When the temperature reaches the temperature threshold for detecting potential fire risk and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold: If the smoke concentration rise rate exceeds 0.25Wppm per minute and lasts for more than 3 minutes, a first-level fire alarm will be issued; If the smoke concentration reaches Wppm or above within 1 minute, a first-level fire alarm will be issued; If the concentration of combustible gas reaches 20%-30% of the lower explosion limit, a first-level fire alarm will be issued; When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration exceeds 2Wppm, a second-level fire alarm will be issued; If the flame detector detects a flame signal, a secondary fire alarm will be issued; If the concentration of combustible gas reaches more than 50% of the lower explosion limit, a second-level fire alarm will be issued.

[0012] A smart remote fire monitoring method based on Internet of Things big data, characterized in that it includes the following steps: S1, collect environmental data in the building through sensors in the data collection layer, and then transmit the collected data to the platform layer; S2, in the platform layer, the data processing module receives the data from the data acquisition layer, performs data cleaning, removes erroneous values, abnormal values ​​and duplicate values, and then performs data conversion. After that, the data storage module stores the cleaned and converted data; S3, the warning temperature threshold dynamic generation module calculates the daily average temperature based on the temperature data of the area where the building is located for n years and sorts it into a data set, divides the temperature zone, and dynamically adjusts the warning threshold K1 and temperature change rate threshold V1 of the temperature sensor to detect potential fire risks in combination with humidity, temperature difference, and sensor aging factors; S4, the early warning and alarm generation module generates corresponding first-level, second-level, and third-level fire early warnings and first-level and second-level fire alarms based on the early warning temperature dynamic threshold, through data analysis, combined with smoke density, combustible gas concentration, and flame signal factors; S5. The fire prediction model generation module uses the dynamic threshold of the warning temperature, integrates historical fire accident data and daily monitoring data, and uses the random forest classification algorithm to build a model, and issues an alarm when the predicted probability of fire exceeds the set threshold.

[0013] In the present invention, the proposed intelligent remote fire monitoring system based on IoT big data has the following beneficial technical effects: 1. In the early warning temperature threshold dynamic generation module, the temperature data of the area where the building is located in previous years is analyzed to divide the low temperature, medium temperature and high temperature ranges. The initial temperature threshold for detecting potential fire risks and triggering early warnings, as well as the initial temperature change rate threshold for temperature changes triggering early warnings are dynamically adjusted in combination with the aging factors of the temperature sensors, humidity and temperature difference conditions. The original static threshold is replaced to achieve dynamic adjustment, thereby reducing the probability of triggering false alarms.

[0014] 2. Regarding the aging problem of the temperature sensor itself, this application introduces the aging coefficient k, constructs a theoretical relationship model, and calculates the measurement deviation caused by aging to ensure that during the entire life cycle of the temperature sensor, there will be no missed or false fire alarms due to sensor aging. At the same time, it also takes into account the impact of humidity and temperature difference environmental factors on temperature measurement to form a dynamic threshold for early warning temperature. Moreover, within the same day, when the temperature difference measured by the temperature sensor is greater than the set threshold, the threshold of the early warning temperature can be flexibly adjusted according to the temperature difference change, thereby increasing the probability of early detection of fire and improving system stability and reliability.

[0015] 3. The early warning and alarm generation module generates corresponding first-level, second-level, and third-level fire warnings and first-level and second-level fire alarms based on the dynamic threshold of the early warning temperature, combined with the smoke concentration, combustible gas concentration, and flame signal factors. It comprehensively considers the smoke concentration, combustible gas concentration, flame signal, and the dynamic threshold of the early warning temperature, and can capture fire signs more comprehensively and verify the possibility of fire from multiple angles, greatly enhancing the reliability of early warnings and alarms and reducing the risk of misjudgment.

[0016] 4. The fire prediction model generation module uses the dynamic threshold of early warning temperature, integrates historical fire accident data and daily monitoring data, and uses the random forest classification algorithm to build a model. It issues an alarm when the predicted probability of fire exceeds the set threshold. The fire prediction model is built with the help of the dynamic threshold of early warning temperature. Since the dynamic threshold of early warning temperature can be adjusted according to different environmental conditions, the model can more accurately capture fire risks in various environments, thereby improving the accuracy of prediction.

[0017] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a principle block diagram of the system of the present invention; Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0019] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0020] like Figure 1 The smart remote fire monitoring system based on IoT big data shown in the figure includes: The data collection layer collects data through sensors and then transmits the collected data to the platform layer; In the data acquisition layer, sensors include: smoke sensors, temperature sensors, flame detectors, combustible gas detectors, and fire water system pressure sensors; Platform layer, the platform layer includes: Data processing module, for data cleaning and data conversion; A data storage module stores the data after data cleaning and data conversion; Early warning temperature threshold dynamic generation module: dynamic threshold is used for the temperature threshold that detects potential fire risk and triggers early warning, and early warning temperature dynamic threshold is generated; Early warning and alarm generation module: generates early warning levels and alarm levels through data analysis based on the early warning temperature dynamic threshold generated by the early warning temperature threshold dynamic generation module; Fire prediction model generation module: Based on the early warning temperature dynamic threshold generated by the early warning temperature threshold dynamic generation module and combined with historical data analysis, a fire prediction model is constructed; In the early warning temperature threshold dynamic generation module, the temperature threshold for detecting potential fire risks and triggering early warnings is generated using a dynamic threshold method, as follows: When conducting remote fire monitoring of a building, Get the temperature data of the area in the building over the years, set the temperature data of the years to the temperature data of n years, calculate the average temperature of each day based on the temperature data of n years, add the average temperature data of the same day in the past years, and then divide by n to get the historical average temperature of each day; Arrange these historical average temperature data in date order to form a complete annual average temperature data set; Set two temperature values ​​as M1 and M2, and M1<M2; In an optional embodiment, the values ​​of M1 and M2 are obtained based on the annual average temperature data set and a cluster analysis method; In the annual mean temperature dataset, the historical mean temperatures for each day are examined sequentially; When the historical average temperature is lower than M1, record this date as the start date of the low temperature range, and continue to check subsequent dates. Once a date with a historical average temperature greater than M1 is encountered, the previous date is taken as the end date of the low temperature range; When encountering a date with a historical average temperature greater than M1, record it as the start date of the medium temperature range, and continue to check subsequent dates. Once encountering a date with a historical average temperature less than M1 or greater than M2, determine the previous date as the end date of the medium temperature range; When encountering a date with a historical average temperature higher than M2, record it as the start date of the high temperature range, and continue to check subsequent dates. When encountering a date with an average temperature lower than M2, take the previous date as the end date of the high temperature range; The third initial temperature threshold value of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained as K1, and the initial temperature change rate threshold value for triggering an early warning by temperature change is set as V1; When the current date is within the low temperature range, the initial temperature threshold for detecting potential fire risks and triggering an early warning is reduced to Y times of K1, and the initial temperature change rate threshold for triggering an early warning is reduced to Y times of V1; In an optional embodiment, 0.8≤Y<1; The initial temperature change rate threshold for triggering an early warning with temperature changes is reduced to Y times of V1. This is because the normal rise in indoor temperature of a building in a low-temperature environment is relatively stable, while the temperature change caused by a fire will be more abrupt. Reducing V1 helps to more sensitively capture this abnormal change and warn of the fire in advance. At the same time, the lower V1 value also takes into account the large temperature difference between indoor and outdoor buildings in a low-temperature environment, and the faster heat propagation when a fire occurs, ensuring that the abnormal temperature rise can be detected in the early stages of a fire.

[0021] When the current date is within the medium temperature range, the values ​​of K1 and V1 remain unchanged; When the current date is within the high temperature range, the initial temperature threshold for detecting potential fire risks and triggering warnings rises to X times of K1, and the initial temperature change rate threshold for temperature changes triggering warnings rises to X times of V1; In an optional embodiment, 1<X≤1.5; This is because in a high temperature environment, increasing V1 can reduce false alarms caused by normal ambient temperature changes. At the same time, once a fire occurs in a high temperature environment, the high temperature environment may accelerate the combustion speed, but the temperature change rate caused by the initial fire still needs to be distinguished from the normal slow change. By increasing V1, while ensuring sensitivity to the rapid temperature changes caused by the fire, it can better adapt to the characteristics of the high temperature environment and avoid too many false alarms.

[0022] The third initial temperature threshold value K1 for the temperature sensor to detect potential fire risk and trigger an early warning is obtained as follows: The second initial temperature threshold value K0 of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained; Condition 1: When the humidity value measured by the humidity sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; Condition 2: On the same day, when the temperature difference measured by the temperature sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; In an optional embodiment, 1<Z≤1.1; The temperature difference refers to the difference between the lowest temperature measured by the temperature sensor and the current temperature; When condition 1 or condition 2 is met, K1=ZK0; When conditions 1 and 2 are met at the same time, K1=Z(ZK0); When condition 1 or condition 2 is not met, K1=K0; The second initial temperature threshold K0 of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained in the following manner: Assume that the first initial temperature threshold of the temperature sensor for detecting potential fire risk and triggering an early warning is K; Start timing from when the temperature sensor is put into use, and record its use time t as the aging time; By obtaining the technical specifications of the temperature sensor, the measurement deviation of the temperature sensor per unit time during the initial use phase is obtained. , represents the temperature measurement deviation caused by aging per unit time; Calculation of measurement deviations due to aging of temperature sensors , the formula is: ; Where k is the aging coefficient. By taking temperature, humidity, usage time and usage frequency as aging factors, a theoretical analysis method is used to establish a theoretical relationship model between the aging coefficient and each aging factor. The actual environmental parameters are substituted into the theoretical model to calculate the aging coefficient k. Then K0= K+ .

[0023] For the aging problem of the temperature sensor itself, this application introduces the aging coefficient k, constructs a theoretical relationship model, and calculates the measurement deviation caused by aging to ensure that during the entire life cycle of the temperature sensor, there will be no missed or false alarms of fire due to sensor aging. At the same time, it also takes into account the impact of humidity and temperature difference environmental factors on temperature measurement to form a dynamic threshold for early warning temperature. Moreover, within the same day, when the temperature difference measured by the temperature sensor is greater than the set threshold, the threshold of the early warning temperature can be flexibly adjusted according to the temperature difference change, thereby increasing the probability of early detection of fire and improving system stability and reliability.

[0024] In the dynamic generation module of the warning temperature threshold, the temperature data of the area where the building is located in previous years is analyzed to divide the low temperature, medium temperature and high temperature ranges. The initial temperature threshold for detecting potential fire risks and triggering warnings and the initial temperature change rate threshold for temperature changes triggering warnings are dynamically adjusted in combination with the aging factors of the temperature sensors, humidity and temperature difference conditions. The original static threshold is replaced to achieve dynamic adjustment, thereby reducing the probability of triggering false alarms.

[0025] In the early warning and alarm generation module, the early warning level and alarm level are generated as follows: When the temperature reaches the temperature threshold that detects potential fire risks and triggers an early warning, a first-level fire warning is issued; When the temperature reaches the temperature threshold for detecting potential fire risks and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold, a secondary fire early warning is issued; When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration continues to rise within Q minutes and exceeds Wppm, a Level 3 fire warning will be issued, indicating that smoldering may occur and further attention is required; In an optional embodiment, the value range of Q is [3,5]; In an optional embodiment, Q=5; If the concentration of combustible gas reaches 10%-15% of the lower explosion limit, a third-level fire warning will be issued; personnel will be arranged to investigate the source of gas leakage; When the temperature reaches the temperature threshold for detecting potential fire risk and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold: If the smoke concentration rise rate exceeds 0.25Wppm per minute and lasts for more than 3 minutes, a first-level fire alarm will be issued; If the smoke concentration reaches Wppm or above within 1 minute, a first-level fire alarm will be issued; At this time, the alarm system of the local area can be activated to notify relevant personnel to conduct on-site inspection.

[0026] If the concentration of combustible gas reaches 20%-30% of the lower explosion limit, a first-level fire alarm will be issued; at this time, you can prepare to start the ventilation emergency equipment.

[0027] When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration exceeds 2Wppm, a second-level fire alarm will be issued; If the flame detector detects a flame signal, a secondary fire alarm will be issued; At this time, a full alarm can be triggered, linking the fire water system to automatically spray water to extinguish the fire, and at the same time notify the fire department.

[0028] If the concentration of combustible gas reaches more than 50% of the lower explosion limit, a secondary fire alarm will be issued; At this time, regardless of the status of other sensors, it is directly judged as a high fire risk, and emergency measures are taken to cut off the gas supply and evacuate personnel.

[0029] The fire prediction model constructed through the above rule-based method can make full use of various sensor information, flexibly judge the fire risk according to the actual situation, and provide strong support for fire warning and emergency response.

[0030] The early warning and alarm generation module generates corresponding first-, second- and third-level fire warnings and first- and second-level fire alarms based on the dynamic threshold of the early warning temperature, combined with factors such as smoke density, combustible gas concentration and flame signal. It comprehensively considers smoke density, combustible gas concentration, flame signal and the dynamic threshold of the early warning temperature, can capture signs of fire more comprehensively, and verify the possibility of fire from multiple angles, greatly enhancing the reliability of early warnings and alarms and reducing the risk of misjudgment.

[0031] In the fire prediction model generation module, the fire prediction model is constructed as follows: Obtain historical fire accident data, historical daily monitoring data, and warning temperature dynamic thresholds, and use the random forest classification algorithm to build a fire prediction model to predict the probability of fire. When the predicted probability of fire is greater than the set threshold, an alarm is issued.

[0032] The fire prediction model generation module uses the dynamic warning temperature threshold, integrates historical fire accident data and daily monitoring data, and uses the random forest classification algorithm to build a model. It issues an alarm when the predicted probability of a fire exceeds the set threshold. The fire prediction model is built with the help of the dynamic warning temperature threshold. Since the dynamic warning temperature threshold can be adjusted according to different environmental conditions, the model can more accurately capture fire risks in various environments, thereby improving the accuracy of prediction.

[0033] like Figure 2 The method for intelligent remote fire monitoring based on IoT big data includes the following steps: S1. Collect environmental data in the building through sensors in the data collection layer. Smoke sensors are used to monitor smoke concentration in the air, temperature sensors are used to measure ambient temperature, flame detectors are used to capture flame signals, combustible gas detectors are used to detect combustible gas leaks, and fire water system pressure sensors are used to monitor fire water pressure. Then the collected data is transmitted to the platform layer. S2, in the platform layer, the data processing module receives the data from the data collection layer, performs data cleaning, removes erroneous values, abnormal values ​​and duplicate values, and then performs data conversion to standardize data in different formats and units. After that, the data storage module stores the cleaned and converted data; S3, the warning temperature threshold dynamic generation module calculates the daily average temperature based on the temperature data of the area where the building is located for n years and sorts it into a data set, divides the temperature zone, and dynamically adjusts the warning threshold K1 and temperature change rate threshold V1 of the temperature sensor to detect potential fire risks in combination with humidity, temperature difference, and sensor aging factors; S4, the early warning and alarm generation module generates corresponding first-level, second-level, and third-level fire early warnings and first-level and second-level fire alarms based on the early warning temperature dynamic threshold, through data analysis, combined with smoke density, combustible gas concentration, and flame signal factors; S5. The fire prediction model generation module uses the dynamic threshold of the warning temperature, integrates historical fire accident data and daily monitoring data, and uses the random forest classification algorithm to build a model, and issues an alarm when the predicted probability of fire exceeds the set threshold.

[0034] A smart remote fire monitoring method based on Internet of Things big data also includes the following steps: Use time series analysis to predict the changing trend of the operating status of fire protection facilities, plan maintenance and repair in advance, and reduce the failure rate of equipment; Use convolutional neural networks in deep learning to intelligently analyze surveillance video images, assist in flame and smoke recognition, improve fire alarm accuracy, and provide scientific and accurate decision-making references for fire management; When constructing a fire prediction model, the random forest classification algorithm performs stratified sampling on historical data, so that the proportion of different types of fire accident data and daily monitoring data in the sample is close to the actual situation, thereby improving the generalization ability of the model.

[0035] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0036] In the embodiments provided by the present invention, it should be understood that the disclosed system or method can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0037] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0038] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0039] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic features of the present invention.

[0040] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can replace or change the technical solution and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A smart remote fire monitoring system based on Internet of Things big data, characterized in that: include: Data collection layer and platform layer, including: Data collection layer: collects data through sensors and then transmits the collected data to the platform layer; The platform layer includes: Data processing module, for data cleaning and data conversion; A data storage module stores the data after data cleaning and data conversion; Warning temperature threshold dynamic generation module: dynamic threshold is used for the temperature threshold that detects potential fire risk and triggers warning, and the warning temperature dynamic threshold is generated; Early warning and alarm generation module: generates early warning levels and alarm levels through data analysis based on the early warning temperature dynamic threshold generated by the early warning temperature threshold dynamic generation module; Fire prediction model generation module: Based on the warning temperature dynamic threshold generated by the warning temperature threshold dynamic generation module and combined with historical data analysis, a fire prediction model is constructed.

2. According to the intelligent remote fire monitoring system based on Internet of Things big data according to claim 1, it is characterized in that: In the data acquisition layer, sensors include: smoke sensors, temperature sensors, flame detectors, combustible gas detectors, and fire water system pressure sensors.

3. The intelligent remote fire monitoring system based on Internet of Things big data according to claim 1 is characterized in that: In the early warning temperature threshold dynamic generation module, the temperature threshold for detecting potential fire risks and triggering early warnings is generated using a dynamic threshold method, as follows: When a building is remotely monitored for fire, the temperature data of the area in the building is obtained. The temperature data of the past years is set as the temperature data of n years. The average temperature of each day is calculated based on the temperature data of n years. The average temperature data of the same day in the past years are added and then divided by n to obtain the historical average temperature of each day. Arrange these historical average temperature data in date order to form a complete annual average temperature data set; Set two temperature values ​​as M1 and M2, and M1<M2; In the annual mean temperature dataset, the historical mean temperatures for each day are examined sequentially; When the historical average temperature is lower than M1, record this date as the start date of the low temperature range, and continue to check subsequent dates. Once a date with a historical average temperature greater than M1 is encountered, the previous date is taken as the end date of the low temperature range; When encountering a date with a historical average temperature greater than M1, record it as the start date of the medium temperature range, and continue to check subsequent dates. Once encountering a date with a historical average temperature less than M1 or greater than M2, determine the previous date as the end date of the medium temperature range; When encountering a date with a historical average temperature higher than M2, record it as the start date of the high temperature range, and continue to check subsequent dates. When encountering a date with an average temperature lower than M2, take the previous date as the end date of the high temperature range; The third initial temperature threshold value of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained as K1, and the initial temperature change rate threshold value for triggering an early warning by temperature change is set as V1; When the current date is within the low temperature range, the initial temperature threshold for detecting potential fire risks and triggering an early warning is reduced to Y times of K1, and the initial temperature change rate threshold for triggering an early warning is reduced to Y times of V1; When the current date is within the medium temperature range, the values ​​of K1 and V1 remain unchanged; When the current date is within the high temperature range, the initial temperature threshold for detecting potential fire risks and triggering an early warning rises to X times of K1, and the initial temperature change rate threshold for triggering an early warning due to temperature changes rises to X times of V1.

4. The intelligent remote fire monitoring system based on Internet of Things big data according to claim 3 is characterized in that: The third initial temperature threshold value K1 for the temperature sensor to detect potential fire risk and trigger an early warning is obtained as follows: The second initial temperature threshold value K0 of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained; Condition 1: When the humidity value measured by the humidity sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; Condition 2: On the same day, when the temperature difference measured by the temperature sensor is greater than the set threshold, the initial temperature threshold K0 for detecting potential fire risk and triggering an early warning rises to Z times K0, and Z < X; When condition 1 or condition 2 is met, K1=ZK0; When conditions 1 and 2 are met at the same time, K1=Z(ZK0); When condition 1 or condition 2 is not met, K1=K0.

5. The intelligent remote fire monitoring system based on Internet of Things big data according to claim 4 is characterized in that: The second initial temperature threshold K0 of the temperature sensor for detecting potential fire risk and triggering an early warning is obtained in the following manner: Assume that the first initial temperature threshold of the temperature sensor for detecting potential fire risk and triggering an early warning is K; Start timing from when the temperature sensor is put into use, and record its use time t as the aging time; By obtaining the technical specifications of the temperature sensor, the measurement deviation of the temperature sensor per unit time during the initial use phase is obtained. , represents the temperature measurement deviation caused by aging per unit time; Calculation of measurement deviations due to aging of temperature sensors , the formula is: ; Where k is the aging coefficient. By taking temperature, humidity, usage time and usage frequency as aging factors, a theoretical analysis method is used to establish a theoretical relationship model between the aging coefficient and each aging factor. The actual environmental parameters are substituted into the theoretical model to calculate the aging coefficient k. Then K0= K+ .

6. The intelligent remote fire monitoring system based on Internet of Things big data according to claim 1 is characterized in that: In the fire prediction model generation module, the fire prediction model is constructed as follows: Obtain historical fire accident data, historical daily monitoring data, and warning temperature dynamic thresholds, and use the random forest classification algorithm to build a fire prediction model to predict the probability of fire. When the predicted probability of fire is greater than the set threshold, an alarm is issued.

7. The intelligent remote fire monitoring system based on Internet of Things big data according to claim 1 or 6 is characterized in that: In the early warning and alarm generation module, the early warning level and alarm level are generated as follows: When the temperature reaches the temperature threshold that detects potential fire risks and triggers an early warning, a first-level fire warning is issued; When the temperature reaches the temperature threshold for detecting potential fire risks and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold, a secondary fire early warning is issued; When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration continues to rise within Q minutes and exceeds Wppm, a level 3 fire warning will be issued; If the concentration of combustible gas reaches 10%-15% of the lower explosion limit, a level 3 fire warning will be issued; When the temperature reaches the temperature threshold for detecting potential fire risk and triggering an early warning, and the temperature change rate reaches the temperature change rate threshold: If the smoke concentration rise rate exceeds 0.25Wppm per minute and lasts for more than 3 minutes, a first-level fire alarm will be issued; If the smoke concentration reaches Wppm or above within 1 minute, a first-level fire alarm will be issued; If the concentration of combustible gas reaches 20%-30% of the lower explosion limit, a first-level fire alarm will be issued; When the temperature reaches the temperature threshold that detects potential fire risk and triggers an early warning: If the smoke concentration exceeds 2Wppm, a second-level fire alarm will be issued; If the flame detector detects a flame signal, a secondary fire alarm will be issued; If the concentration of combustible gas reaches more than 50% of the lower explosion limit, a second-level fire alarm will be issued.

8. The intelligent remote fire monitoring method based on Internet of Things big data according to any one of claims 1 to 7 is characterized in that: The following steps are involved: S1, collect environmental data in the building through sensors in the data collection layer, and then transmit the collected data to the platform layer; S2, in the platform layer, the data processing module receives the data from the data acquisition layer, performs data cleaning, removes erroneous values, abnormal values ​​and duplicate values, and then performs data conversion. After that, the data storage module stores the cleaned and converted data; S3, the warning temperature threshold dynamic generation module calculates the daily average temperature based on the temperature data of the area where the building is located for n years and sorts it into a data set, divides the temperature zone, and dynamically adjusts the warning threshold K1 and temperature change rate threshold V1 of the temperature sensor to detect potential fire risks in combination with humidity, temperature difference, and sensor aging factors; S4, the early warning and alarm generation module generates corresponding first-level, second-level, and third-level fire early warnings and first-level and second-level fire alarms based on the early warning temperature dynamic threshold, through data analysis, combined with smoke density, combustible gas concentration, and flame signal factors; S5. The fire prediction model generation module uses the dynamic threshold of the warning temperature, integrates historical fire accident data and daily monitoring data, and uses the random forest classification algorithm to build a model, and issues an alarm when the predicted probability of fire exceeds the set threshold.

Images