A harmful gas monitoring and early warning method and system

Abstract

The present application relates to the field of gas monitoring and early warning technology, and specifically discloses a harmful gas monitoring and early warning method and system. The method mainly includes: receiving the first concentration data and the second concentration data of the specified gas transmitted by the first detector and the second detector respectively, and the detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than the second concentration data, and the first detector is configured to realize the real-time transmission of data, and the second detector is configured to realize the intermittent transmission of data. The present application adopts the real-time data and concentration growth rate prediction of the first detector, triggers the LSTM model to process the precise data of the second detector, predicts the future gas concentration, and judges in advance whether the concentration of harmful gases will reach the safety threshold and warns in advance, which can identify potential risks in advance and provide relevant departments with a longer response time, so as to take preventive measures more effectively to avoid or reduce environmental pollution and health hazards.

Description

Technical Field

[0001] The present application relates to the field of gas monitoring and early warning technology, and in particular to a harmful gas monitoring and early warning method and system. Background Art

[0002] With the acceleration of industrialization and the development of urbanization, the problem of harmful gas emissions has become increasingly serious, posing a huge threat to human health and the ecological environment. Harmful gases, such as sulfur dioxide and nitrogen oxides, not only damage the human respiratory system and nervous system, but also cause environmental problems such as acid rain and photochemical smog. By real-time monitoring of the concentration and changing trends of harmful gases, pollution sources can be discovered in a timely manner, the degree of pollution can be assessed, and decision-making support can be provided to environmental protection departments, thereby effectively controlling and reducing harmful gas emissions and protecting human health and the ecological environment.

[0003] Current harmful gas monitoring technology mainly relies on setting thresholds for early warning, that is, when the concentration of harmful gases reaches or exceeds the preset safety threshold, the system will trigger an alarm. However, this method has a significant limitation, and it is difficult to detect the growth or leakage of harmful gases at an earlier stage. This means that before the concentration of harmful gases reaches a dangerous level, it may have already caused potential threats to the environment and human health. Due to the lag in early warning time, the time for relevant departments to take treatment measures will also be shortened accordingly, thereby increasing the difficulty and risk of responding to emergencies. Summary of the invention

[0004] This application solves the problem of delayed warning time caused by the harmful gas monitoring technology in the prior art, which mainly relies on setting thresholds for warning, by providing a harmful gas monitoring and early warning method and system. It can identify potential risks in advance and provide relevant departments with a longer response time, so as to more effectively take preventive measures to avoid or reduce environmental pollution and health hazards.

[0005] In a first aspect, the present application provides a harmful gas monitoring and early warning method, which includes the following steps: receiving monitoring data, the monitoring data including: first concentration data and second concentration data of a specified gas transmitted by a first detector and a second detector respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data; calculating the concentration growth rate of the first concentration data, and determining whether the concentration growth rate reaches a preset predicted growth rate; if the concentration growth rate reaches the predicted growth rate, using an LSTM model to process the first concentration data or the second concentration data to obtain predicted concentration data of the specified gas in a specified time period in the future; determining whether the predicted concentration data reaches a safety threshold, and if so, issuing an early warning message.

[0006] The real-time data and concentration growth rate prediction of the first detector are used to trigger the LSTM model to process the precise data of the second detector. After obtaining the predicted concentration data, the data is immediately evaluated to determine whether it has reached the safety threshold. If the safety threshold is reached, the early warning mechanism will be immediately triggered to issue early warning information, which can identify potential risks in advance and provide valuable response time for relevant departments. Through early warning, relevant departments can take preventive measures more effectively, such as strengthening monitoring, adjusting emissions, evacuating people, etc., so as to avoid or reduce environmental pollution and health hazards that may be caused by excessive concentrations of harmful gases, and can play an important role in environmental protection and public safety.

[0007] Furthermore, the concentration growth rate includes a short-term maximum concentration growth rate Gs and a long-term concentration growth rate Gl.

[0008] Further, calculating the short-term concentration maximum growth rate Gs includes: obtaining all first concentration data Cs within the most recent specified time period, and the detection time point t corresponding to the first concentration data Cs; determining the first concentration data with the maximum value among all the first concentration data Cs; and the first concentration data with the minimum value , and determine the first concentration data and the first concentration data The corresponding detection time points and detection time point ; Calculate the maximum short-term concentration growth rate Gs, Gs is: .

[0009] Further, calculating the long-term concentration growth rate Gl includes: obtaining all first concentration data Cl and the detection time point t corresponding to the first concentration data Cl; determining the most recent specified number of first concentration data Cl among all first concentration data Cl; , and the first concentration data Corresponding detection time point , determine the earliest specified number of first concentration data , and the first concentration data Corresponding detection time point ; Calculate all first concentration data separately The average value of Car, all detection time points The average value tar, all first concentration data The average Cap of all detection time points The average value of tap; calculate the long-term concentration growth rate Gl, Gl is: .

[0010] Furthermore, judging whether the concentration growth rate reaches the preset predicted required growth rate includes: judging whether the short-term maximum concentration growth rate Gs reaches the preset value k1, and if so, judging that the concentration growth rate reaches the preset predicted required growth rate; judging whether the long-term concentration growth rate Gl reaches the preset value k2, and if so, judging that the concentration growth rate reaches the preset predicted required growth rate.

[0011] Furthermore, if the maximum short-term concentration growth rate Gs reaches the preset value k1, the LSTM model processes the detection time point and detection time point if the long-term concentration growth rate Gl reaches the preset value k2, the LSTM model processes the first concentration data corresponding to the detection time points corresponding to all the second concentration data.

[0012] Furthermore, when the second concentration data is received, a reference interval is constructed based on the second concentration data, and it is determined whether the first concentration data received at the same time point is within the reference interval. If not, an abnormal detector is determined between the first detector and the second detector.

[0013] Furthermore, determining the abnormal detector includes: sending an open command to the third detector, and receiving third concentration data of the specified gas transmitted by the third detector; calculating the degree of deviation of the average values ​​of the first concentration data, the second concentration data and the third concentration data at the same time point; and determining the first detector or the second detector with the largest degree of deviation of the average value as the abnormal detector.

[0014] Furthermore, the harmful gas monitoring and early warning method also includes: after determining the abnormal detector, using a third detector to replace the abnormal detector.

[0015] In a second aspect, the present application provides a harmful gas monitoring and early warning system, which adopts the harmful gas monitoring and early warning method as described in the first aspect, and includes: a data receiving module, a growth rate calculation module, a data prediction module and an early warning module.

[0016] The data receiving module is used to receive monitoring data, wherein the monitoring data includes: first concentration data and second concentration data of a specified gas transmitted by the first detector and the second detector, respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data;

[0017] The growth rate calculation module is used to calculate the concentration growth rate of the first concentration data, and determine whether the concentration growth rate reaches a preset predicted growth rate;

[0018] The data prediction module is used to process the first concentration data or the second concentration data using the LSTM model if the concentration growth rate reaches the growth rate required for prediction, so as to obtain the predicted concentration data of the specified gas in the specified time period in the future;

[0019] The early warning module is used to determine whether the predicted concentration data reaches a safety threshold, and if so, issue an early warning message.

[0020] In a third aspect, the present application provides a harmful gas monitoring and early warning device, which includes: a memory and a processor.

[0021] The memory is used to store computer programs; the processor is used to implement the steps of the harmful gas monitoring and early warning method as described in the first aspect when executing the computer program.

[0022] The technical solution provided by this application has at least the following technical effects or advantages:

[0023] The real-time data and concentration growth rate prediction of the first detector are used to trigger the LSTM model to process the precise data of the second detector, predict the future gas concentration, and judge in advance whether the concentration of harmful gases will reach the safety threshold and issue a warning. This effectively solves the problem of delayed warning time caused by the harmful gas monitoring technology in the existing technology that mainly relies on setting thresholds for warning. It can identify potential risks in advance and provide relevant departments with a longer response time, so as to take preventive measures more effectively and avoid or reduce environmental pollution and health hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a flow chart of the harmful gas monitoring and early warning method in Example 1 of the present application;

[0025] Figure 2 This is a schematic diagram of a process for determining an anomaly detector in Embodiment 2 of the present application;

[0026] Figure 3 This is a module diagram of the harmful gas monitoring and early warning system in Example 3 of the present application. DETAILED DESCRIPTION

[0027] In order to solve the problem of delayed warning time caused by the harmful gas monitoring technology mainly relying on setting thresholds for early warning as proposed in the background technology, the present application utilizes the real-time data and concentration growth rate prediction of the first detector to trigger the LSTM model to process the precise data of the second detector, predict future gas concentrations, and determine in advance whether the concentration of harmful gases will reach the safety threshold and issue an early warning. This can identify potential risks in advance and provide relevant departments with a longer response time, thereby more effectively taking preventive measures to avoid or reduce environmental pollution and health hazards.

[0028] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0029] Embodiment 1:

[0030] like Figure 1 As shown, this embodiment provides a harmful gas monitoring and early warning method, which includes the following steps:

[0031] S100. Receive monitoring data, wherein the monitoring data includes: first concentration data and second concentration data of a specified gas transmitted by the first detector and the second detector respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data.

[0032] In S100, the first detector has a relatively low detection accuracy, and the data collected is relatively rough in numerical value, which means that the amount of data required to be transmitted is also correspondingly small during data transmission, making the first detector very suitable for being configured as an instant transmission mode. It can quickly send data to the processing center without sacrificing too much accuracy, thereby ensuring data immediacy while also ensuring transmission efficiency.

[0033] In contrast, the second detector has higher detection accuracy and can capture more subtle changes in gas concentration. However, high accuracy also means that the amount of data collected is larger and more detailed. In order to balance the needs of data transmission and bandwidth limitations, the second detector is configured for intermittent transmission. This can effectively reduce the amount of data transmitted while ensuring data accuracy, achieving a good balance between data transmission efficiency and accuracy.

[0034] For example, both the first detector and the second detector can adopt Honeywell RAEGuard 3 series fixed toxic and hazardous gas detectors. The first detector adopts low-precision mode and transmits data in real time, and the second detector adopts high-precision mode and transmits data intermittently. The intermittent time can be adaptively set according to specific needs.

[0035] S200. Calculate the concentration growth rate of the first concentration data, and determine whether the concentration growth rate reaches a preset predicted required growth rate.

[0036] In S200, the concentration growth rate is the concentration change per unit time. The concentration growth rate can be calculated by the difference between different first concentration data and the difference between corresponding detection time points, so as to quickly capture the changing trend of gas concentration. The current concentration of harmful gases has not yet reached the safety threshold, but if the increasing trend is obvious, it means that it may quickly approach or even exceed the safety limit in the future, thus posing a safety hazard. Therefore, by quickly capturing the changing trend of gas concentration, for those with relatively obvious changing trends, it can be predicted to determine whether the gas is likely to reach the safety threshold in the future. In addition, by comparing the calculated concentration growth rate with the preset predicted growth rate, it is possible to effectively distinguish between normal fluctuations and potentially dangerous growth trends.

[0037] Only when the concentration growth rate of harmful gases reaches the preset growth rate required for prediction, further prediction or response operations can be performed, which can effectively reduce the computing resources required for prediction and avoid unnecessary complex calculations and analysis, thereby reducing costs and improving the overall efficiency of the system. By accurately capturing and evaluating the changing trend of gas concentration, the system can achieve optimal resource allocation while ensuring safety.

[0038] S300. If the concentration growth rate reaches the predicted growth rate, the LSTM model is used to process the first concentration data or the second concentration data to obtain the predicted concentration data of the specified gas in a specified time period in the future.

[0039] When the concentration growth rate reaches the preset growth rate required for prediction, it indicates that the change trend of harmful gases is significant and it is necessary to make a prediction. However, since the first concentration data transmitted by the first detector has low precision, making predictions directly based on this data may lead to inaccurate results. Therefore, the LSTM model can be used to process the second concentration data transmitted by the second detector. The second concentration data can more truly reflect the changes in gas concentration due to its high precision. By processing the second concentration data with the LSTM model, the predicted concentration data of the specified gas in the specified time period in the future can be obtained, which is more accurate than the prediction results based on low-precision data, thereby improving the accuracy of the prediction.

[0040] The LSTM model is a special recurrent neural network (RNN) that is suitable for processing time series data. When processing the second concentration data, the LSTM model is able to capture the long-term dependencies in the time series, that is, to use historical concentration data to predict future concentrations. The model receives concentration data through the input layer, processes it through the hidden layer, and finally outputs the predicted concentration data through the output layer. Because the LSTM model has memory and can retain historical information, it can make time series predictions more accurately.

[0041] When training the LSTM model, the training set data is used to train the model, the gradient of the loss function to the model parameters is calculated through the back-propagation algorithm, and the optimizer is used to update the parameters. During the training process, it is necessary to monitor the performance of the model, such as the loss value and accuracy, in order to adjust the hyperparameters. After multiple iterations of training, the model can learn the complex relationship between concentration data and time, and can accurately predict the future concentration of harmful gases.

[0042] S400. Determine whether the predicted concentration data reaches a safety threshold, and if so, issue a warning message.

[0043] In S400, the safety threshold is a preset critical value used to determine whether the concentration of harmful gases is safe. After obtaining the predicted concentration data, the data is immediately evaluated to determine whether it has reached the safety threshold. If the safety threshold is reached, the early warning mechanism will be immediately triggered and early warning information will be issued, which can identify potential risks in advance and provide valuable response time for relevant departments. Through early warning, relevant departments can take preventive measures more effectively, such as strengthening monitoring, adjusting emissions, evacuating people, etc., so as to avoid or reduce environmental pollution and health hazards that may be caused by excessive concentrations of harmful gases, which can play an important role in environmental protection and public safety.

[0044] In S200, the concentration growth rate includes the short-term maximum concentration growth rate Gs and the long-term concentration growth rate Gl.

[0045] Calculation of the maximum short-term concentration growth rate Gs includes:

[0046] S201. Acquire all first concentration data Cs within a recent specified time period, and the detection time point t corresponding to the first concentration data Cs.

[0047] S202. Determine the first concentration data with the largest value among all the first concentration data Cs. and the first concentration data with the minimum value , and determine the first concentration data and the first concentration data The corresponding detection time points and detection time point .

[0048] S203. Calculate the maximum short-term concentration growth rate Gs, Gs is: .

[0049] In S201-S203, by obtaining all the first concentration data and their corresponding detection time points within the most recent specified time period, the maximum and minimum concentration values ​​within the time period, as well as their respective corresponding time points, can be determined. The calculation of the short-term concentration maximum growth rate Gs is essentially to divide the amplitude of the concentration change by the time span. This indicator can intuitively reflect the drastic changes that may occur in the concentration of harmful gases in a short period of time. If the Gs value is high, it means that the concentration of harmful gases has changed significantly in a short period of time, which is often associated with certain emergencies or abnormal situations. It can quickly identify potential risks and help to take corresponding countermeasures in a timely manner.

[0050] Calculation of the long-term concentration growth rate Gl includes:

[0051] S211. Obtain all first concentration data Cl and the detection time point t corresponding to the first concentration data Cl.

[0052] S212. Determine the nearest specified number of first concentration data among all first concentration data Cl. , and the first concentration data Corresponding detection time point , determine the earliest specified number of first concentration data , and the first concentration data Corresponding detection time point .

[0053] S213. Calculate all first concentration data respectively The average value of Car, all detection time points The average value tar, all first concentration data The average Cap of all detection time points The average value of tap.

[0054] S214. Calculate the long-term concentration growth rate Gl, Gl is: .

[0055] In S211-S214, the calculation of the long-term concentration growth rate Gl is intended to evaluate the changing trend of harmful gas concentrations over a longer time span. By selecting the earliest and most recent specified number of first concentration data and their corresponding detection time points, the average concentrations of these two time periods can be calculated respectively. The calculation formula of the long-term concentration growth rate Gl essentially reveals the rate of change of the average concentration over time, providing us with a quantitative tool to measure the long-term growth of harmful gas concentrations. It is worth noting that although this growth may not be obvious in the short term, or even difficult to be directly perceived, the long-term concentration growth rate Gl can capture this potential, slow growth trend, and can help timely detect the long-term growth risk of harmful gas concentrations, and provide early warning signals for long-term planning of environmental protection and public safety, thereby ensuring that timely and effective measures can be taken to respond to potential environmental and health threats.

[0056] In S200, judging whether the concentration growth rate reaches a preset predicted growth rate includes:

[0057] a. Determine whether the maximum short-term concentration growth rate Gs reaches the preset value k1. If so, determine whether the concentration growth rate reaches the preset predicted growth rate.

[0058] b. Determine whether the long-term concentration growth rate Gl reaches the preset value k2. If so, determine that the concentration growth rate reaches the preset predicted required growth rate.

[0059] When the maximum short-term concentration growth rate Gs reaches the preset value k1, it means that the concentration of harmful gases has changed significantly in a short period of time, which may be caused by sudden events or abnormal situations. This rapid response mechanism helps to quickly identify and respond to potential environmental risks. When the long-term concentration growth rate Gl reaches the preset value k2, even if this growth is not obvious in the short term, this trend can be captured through long-term monitoring, which helps to timely detect and deal with those slow but continuous growth of harmful gases, thereby ensuring environmental and public safety. By setting the preset values ​​k1 and k2, the growth trend of harmful gas concentrations can be monitored more finely, and comprehensive monitoring of the growth trend of harmful gases can be achieved. Whether it is sudden or slow growth, it can be responded to in a timely and effective manner.

[0060] In S300, if the maximum short-term concentration growth rate Gs reaches the preset value k1, the LSTM model processes the detection time point and detection time point if the long-term concentration growth rate Gl reaches the preset value k2, the LSTM model processes the first concentration data corresponding to the detection time points corresponding to all the second concentration data.

[0061] When the maximum short-term concentration growth rate Gs reaches the preset value k1, it indicates that the concentration of harmful gases has changed significantly in a short period of time. At this time, the LSTM model will focus on processing the detection time point. and detection time point The second concentration data between the two periods has a higher precision and can more accurately reflect the sharp change in concentration, thus helping the model to make more accurate predictions.

[0062] On the other hand, if the long-term concentration growth rate Gl reaches the preset value k2, it indicates that the concentration of harmful gases has been steadily increasing over a long time span. In this scenario, if the concentration growth is negligible, the environment can be considered to be in a relatively safe state; however, if the growth is significant, it needs to be closely monitored. In view of the huge amount of data generated by long-term monitoring, the LSTM model will choose to process the first concentration data at the detection time point corresponding to all the second concentration data. While reflecting the situation of more obvious concentration growth, it can also effectively reduce the amount of data, thereby ensuring the accuracy of the prediction while greatly saving computing resources.

[0063] Embodiment 2:

[0064] like Figure 2 As shown, based on Example 1, when the second concentration data is received, a reference interval is constructed based on the second concentration data, and it is determined whether the first concentration data received at the same time point is within the reference interval. If not, an abnormal detector is determined between the first detector and the second detector.

[0065] After receiving the second concentration data, a reference interval is immediately constructed based on the second concentration data. This reference interval is actually a range for evaluating the normality of data, which reflects the reasonable range that the concentration data of harmful gases should be in under the current conditions. Then check whether the first concentration data received at the same time point falls within this reference interval. If the first concentration data is within the reference interval, it can be considered that the first detector and the second detector are working normally and no abnormal situation occurs.

[0066] Whenever new second concentration data is received, the abnormality judgment process is repeated. This real-time abnormality detection mechanism ensures the accuracy and reliability of the detector data. Once the data deviates from the normal range, the system can quickly identify possible abnormal detectors and take timely measures to repair or replace them, ensuring the continuity and accuracy of monitoring work.

[0067] Identify anomaly detectors including:

[0068] S110. Send an opening instruction to the third detector, and receive third concentration data of the specified gas transmitted by the third detector.

[0069] S120. Calculate the degree of deviation of the average values ​​of the first concentration data, the second concentration data, and the third concentration data at the same time point.

[0070] S130. Determine the first detector or the second detector with the largest average value deviation as the abnormal detector.

[0071] In S110-S130, first, the average value of the concentration data of the three detectors at the same time point is calculated. , for: ,in, , , Respectively represent the concentration data of the first detector, the second detector and the third detector at the same time point.

[0072] Next, calculate the deviation of each detector concentration data from the average value , for: , where i=1 or 2, when i=1, represents the first detector, Represents the first concentration data, when i=2, represents the second detector, Represents the second concentration data.

[0073] Finally, compare the deviation of the first detector The deviation from the second detector , select the detector with a larger deviation as the anomaly detector. That is: if D1>D2, the first detector is the anomaly detector, otherwise the second detector.

[0074] When it is preliminarily determined that there may be an abnormality in the first detector and the second detector, the third detector is activated for further identification, which can avoid unnecessary use of detectors and ensure that the third detector is called only when necessary, thereby effectively reducing the frequency of use of the detector and extending its service life.

[0075] After the anomaly detector is determined, a third detector is used to replace the anomaly detector.

[0076] The third detector can also be a Honeywell RAEGuard 3 series fixed toxic and hazardous gas detector. After successfully switching to the same working mode as the anomaly detector, the third detector will seamlessly connect its monitoring tasks to ensure the continuity and accuracy of the monitoring system, thereby effectively replacing the original anomaly detector and continuing to provide strong support for safety monitoring work.

[0077] Embodiment three:

[0078] like Figure 3 As shown, this embodiment provides a harmful gas monitoring and early warning system, which adopts the harmful gas monitoring and early warning method in embodiment one or embodiment two, and includes: a data receiving module, a growth rate calculation module, a data prediction module and an early warning module.

[0079] A data receiving module, which is used to receive monitoring data, wherein the monitoring data includes: first concentration data and second concentration data of a specified gas transmitted by a first detector and a second detector respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data.

[0080] The growth rate calculation module is used to calculate the concentration growth rate of the first concentration data and determine whether the concentration growth rate reaches a preset predicted growth rate.

[0081] The data prediction module is used to use the LSTM model to process the first concentration data or the second concentration data if the concentration growth rate reaches the predicted growth rate, so as to obtain the predicted concentration data of the specified gas in a specified time period in the future.

[0082] The early warning module is used to determine whether the predicted concentration data reaches the safety threshold. If so, an early warning message is issued.

[0083] This embodiment has all the advantages of the harmful gas monitoring and early warning method in Embodiment 1 or Embodiment 2, and can automatically implement the steps of the harmful gas monitoring and early warning method.

[0084] Embodiment 4:

[0085] This embodiment provides a harmful gas monitoring and early warning device, which includes: a memory and a processor.

[0086] The memory is used to store computer programs; the processor is used to implement the steps of the harmful gas monitoring and early warning method in Example 1 or Example 2 when executing the computer program.

[0087] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

Claims

1. A harmful gas monitoring and early warning method, characterized in that: It includes the following steps: Receive monitoring data, the monitoring data including: first concentration data and second concentration data of a specified gas transmitted by a first detector and a second detector, respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data; Calculating the concentration growth rate of the first concentration data, and determining whether the concentration growth rate reaches a preset predicted required growth rate; If the concentration growth rate reaches the growth rate required for prediction, the LSTM model is used to process the first concentration data or the second concentration data to obtain the predicted concentration data of the specified gas in the specified time period in the future; Determine whether the predicted concentration data reaches a safety threshold, and if so, issue a warning message; Wherein, the concentration growth rate includes the short-term maximum concentration growth rate Gs and the long-term concentration growth rate Gl; Calculation of the maximum short-term concentration growth rate Gs includes: Acquire all first concentration data Cs within a recent specified time period, and the detection time point t corresponding to the first concentration data Cs; Among all the first density data Cs, determine the first density data with the largest value and the first concentration data with the minimum value , and determine the first concentration data and the first concentration data The corresponding detection time points and detection time point ; Calculate the maximum short-term concentration growth rate Gs, Gs is: ; Calculation of the long-term concentration growth rate Gl includes: Acquire all first concentration data Cl and the detection time point t corresponding to the first concentration data Cl; Among all the first density data Cl, determine the most recent specified number of first density data , and the first concentration data Corresponding detection time point , determine the earliest specified number of first concentration data , and the first concentration data Corresponding detection time point ; Calculate all first concentration data separately The average value of Car, all detection time points The average value tar, all first concentration data The average Cap of all detection time points The average value of tap; Calculate the long-term concentration growth rate Gl, Gl is: ; Determining whether the concentration growth rate reaches the preset predicted growth rate includes: Determine whether the short-term maximum concentration growth rate Gs reaches the preset value k1, and if so, determine whether the concentration growth rate reaches the preset predicted growth rate; Determine whether the long-term concentration growth rate Gl reaches a preset value k2, and if so, determine that the concentration growth rate reaches a preset predicted required growth rate; If the maximum short-term concentration growth rate Gs reaches the preset value k1, the LSTM model processes the detection time point and detection time point The second concentration data between; If the long-term concentration growth rate Gl reaches the preset value k2, the LSTM model processes the first concentration data corresponding to the detection time points corresponding to all the second concentration data.

2. The harmful gas monitoring and early warning method according to claim 1, characterized in that: When the second concentration data is received, a reference interval is constructed based on the second concentration data, and it is determined whether the first concentration data received at the same time point is within the reference interval. If not, an abnormal detector is determined between the first detector and the second detector.

3. The harmful gas monitoring and early warning method according to claim 2, characterized in that: Identify anomaly detectors including: sending an opening instruction to the third detector, and receiving third concentration data of the specified gas transmitted by the third detector; Calculate the average deviation degree of the first concentration data, the second concentration data and the third concentration data at the same time point; The first detector or the second detector having the largest deviation from the average value is determined as the abnormal detector.

4. The harmful gas monitoring and early warning method according to claim 3, characterized in that: Also includes: After the anomaly detector is determined, a third detector is used to replace the anomaly detector.

5. A harmful gas monitoring and early warning system, which adopts the harmful gas monitoring and early warning method according to any one of claims 1 to 4, characterized in that: It includes: A data receiving module, which is used to receive monitoring data, wherein the monitoring data includes: first concentration data and second concentration data of a specified gas transmitted by the first detector and the second detector, respectively, and detection time points corresponding to the first concentration data and the second concentration data; wherein the accuracy of the first concentration data is less than that of the second concentration data, and the first detector is configured to realize instant transmission of data, and the second detector is configured to realize intermittent transmission of data; A growth rate calculation module, which is used to calculate the concentration growth rate of the first concentration data and determine whether the concentration growth rate reaches a preset predicted growth rate; A data prediction module, which is used to use an LSTM model to process the first concentration data or the second concentration data if the concentration growth rate reaches the predicted growth rate, so as to obtain predicted concentration data of the specified gas in a specified time period in the future; An early warning module is used to determine whether the predicted concentration data reaches a safety threshold, and if so, issue an early warning message; Wherein, the concentration growth rate includes the short-term maximum concentration growth rate Gs and the long-term concentration growth rate Gl; Calculating the maximum short-term concentration growth rate Gs includes: obtaining all first concentration data Cs within the most recent specified time period and the detection time point t corresponding to the first concentration data Cs; determining the first concentration data with the maximum value among all the first concentration data Cs; and the first concentration data with the minimum value , and determine the first concentration data and the first concentration data The corresponding detection time points and detection time point ; Calculate the maximum short-term concentration growth rate Gs, Gs is: ; Calculating the long-term concentration growth rate Gl includes: obtaining all first concentration data Cl and the detection time point t corresponding to the first concentration data Cl; determining the most recent specified number of first concentration data Cl among all first concentration data Cl; , and the first concentration data Corresponding detection time point , determine the earliest specified number of first concentration data , and the first concentration data Corresponding detection time point ; Calculate all first concentration data separately The average value of Car, all detection time points The average value tar, all first concentration data The average Cap of all detection time points The average value of tap; calculate the long-term concentration growth rate Gl, Gl is: ; Determining whether the concentration growth rate reaches the preset predicted growth rate includes: determining whether the short-term concentration maximum growth rate Gs reaches the preset value k1, if so, determining that the concentration growth rate reaches the preset predicted growth rate; determining whether the long-term concentration growth rate Gl reaches the preset value k2, if so, determining that the concentration growth rate reaches the preset predicted growth rate; If the maximum short-term concentration growth rate Gs reaches the preset value k1, the LSTM model processes the detection time point and detection time point if the long-term concentration growth rate Gl reaches the preset value k2, the LSTM model processes the first concentration data corresponding to the detection time points corresponding to all the second concentration data.

Images