AI-based coal bunker intelligent safety monitoring system and method

Abstract

The invention discloses an AI-based coal bunker intelligent safety monitoring system and method, relates to the technical field of artificial intelligence, and solves the problems that the detection range of a sensor for detecting gas is limited; the detection accuracy and integrity of the gas in the coal bunker are ensured; a large number of sensors need to be arranged for detection; a large amount of data is generated and needs to be calculated; and the detection efficiency of the hazardous gas in the coal bunker is reduced. The monitoring planning module is used for setting a monitoring scheme of each monitored gas in the coal bunker based on characteristic data; the data acquisition module is used for acquiring monitoring data of each gas sensor; the safety monitoring module is used for inputting the monitoring data into a concentration evaluation model to obtain a concentration risk score corresponding to the detected gas; the alarm module is used for generating a corresponding alarm signal based on the concentration risk score; the safety monitoring effect is improved through the positions of the sensors, and then the overall safety monitoring efficiency of the coal bunker is improved.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence, involves artificial intelligence technology, and specifically is an AI-based intelligent safety monitoring system and method for coal bunkers. Background Art

[0002] A coal bunker is an important device for storing coal, a closed or semi-closed structure for storing and transferring coal, mainly used to balance the supply and demand in the links of coal production, transportation and processing.

[0003] The prior art (CN118447666A) discloses the field of safety warning technology, and particularly involves a coal bunker safety monitoring and warning method and system based on artificial intelligence. The method includes: collecting environmental data points of each monitoring point in the coal bunker, constructing an isolation forest based on the environmental data points. In the constructed isolation trees, all environmental data points included in each child node are continuous in time series, and all environmental data points in the child node are less than the segmentation value or not less than the segmentation value in the corresponding feature dimension. According to the difference between the leaf node where the environmental data point at the current moment is located and the other leaf nodes in each isolation tree, the layer depth of the leaf node where it is located, and the number of environmental data points included in the leaf node where it is located, obtain the anomaly score of the environmental data point at the current moment, and conduct safety warning according to the size of the anomaly score; the present invention can identify continuous abnormal changes in environmental data, and can timely discover coal bunker safety hazards and give warnings.

[0004] The above-mentioned coal bunker safety monitoring and warning method and system based on artificial intelligence judge whether the anomaly of the environmental data point of the current monitoring point at the current moment is an anomaly caused by a sensor failure or a safety hazard caused by improper coal storage according to the difference between the anomaly scores of the environmental data points of different monitoring points at the current moment, and take different warning measures, so that coal bunker safety hazards can be discovered and disposed of more timely; however, in practical applications, the detection range of the sensors for detecting gases is limited; if you want to ensure the detection accuracy and integrity of the gases in the coal bunker; then a large number of sensors need to be set for detection; a large amount of data will be generated and need to be calculated; resulting in a reduction in the detection efficiency of dangerous gases in the coal bunker; therefore, an AI-based intelligent safety monitoring system and method for coal bunkers are needed. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art; for this purpose, this application proposes an AI-based intelligent safety monitoring system and method for coal bunkers, for the technical problem that the detection range of the sensors for detecting gases is limited; if you want to ensure the detection accuracy and integrity of the gases in the coal bunker; then a large number of sensors need to be set for detection; a large amount of data will be generated and need to be calculated; resulting in a reduction in the detection efficiency of dangerous gases in the coal bunker.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides an AI-based coal bunker intelligent safety monitoring system and method, including: a monitoring planning module, a data acquisition module, a safety monitoring module, an alarm module and a database;

[0007] Monitoring planning module: obtain characteristic data of several gases to be measured, set monitoring plans for each gas to be measured in the coal bunker based on the characteristic data, and install each gas sensor based on the monitoring plan;

[0008] Data acquisition module: collects monitoring data of each gas sensor based on data acquisition instructions;

[0009] Safety monitoring module: obtains a number of monitoring data corresponding to each measured gas, inputs the monitoring data into the concentration assessment model to obtain a concentration risk score for evaluating the danger level of the corresponding measured gas; sets data collection instructions corresponding to the measured gas based on the concentration risk score;

[0010] Alarm module: Generates corresponding alarm signals based on concentration risk scores and issues alarms.

[0011] In this application, the coal bunker is closed.

[0012] Since the distribution of different gases in the coal bunker has a certain regularity, such as methane, which is generally concentrated in the high places of the coal bunker due to its small relative molecular mass; sensors are set up in a targeted manner based on the regularity of the distribution of different dangerous gases; the number of sensors is greatly reduced, and thus the amount of data processing during subsequent monitoring data processing is greatly reduced; while ensuring the accuracy and completeness of the dangerous gas monitoring results, the detection efficiency of dangerous gases in the coal bunker is improved.

[0013] Furthermore, the data acquisition module is also used to obtain environmental data and coal quality data inside the coal bunker.

[0014] Further, the monitoring scheme of the detected gas in the coal bunker is set based on the characteristic data, including:

[0015] Obtain the relative molecular mass corresponding to each detected gas in the characteristic data; as well as the storage environment data and coal quality data of the coal bunker;

[0016] Divide the coal bunker into several monitoring areas according to height;

[0017] Generating the preliminary distribution score for representing the distribution of the measured gas in each monitoring area based on the relative molecular mass of the measured gas and the average molecular mass of air;

[0018] Based on the storage environment data and the coal quality data, the preliminary distribution score is modified to obtain the distribution score that conforms to the current actual storage situation;

[0019] Generate the number of gas sensors for each monitoring area based on the distribution score of the gas to be measured in each monitoring area;

[0020] Successively obtain the number of gas sensors for each gas to be measured in each monitoring area; and integrate them into a monitoring plan.

[0021] Further, generate the preliminary distribution score for representing the distribution of the gas to be measured in each monitoring area based on the relative molecular mass of the gas to be measured and the average molecular mass of air, including:

[0022] Obtain that the relative molecular mass of several gases to be measured is MAj, where j is the number of the gas to be measured; the average relative molecular mass of air is MK; j = 1, 2,..., n; n represents the number of gases to be measured; the total height of the internal area of the coal bunker is H, which is divided into h layers according to equal height; i = 1, 2,..., h; i represents the number of the monitoring area;

[0023] Through the formula:

[0024]

[0025] where Pij represents the preliminary distribution score corresponding to the gas j in the i-th monitoring area; represents the height corresponding to the i-th monitoring area;

[0026] After that, calculate the preliminary distribution scores of each gas to be measured in each monitoring area through the above formula;

[0027] For the gas to be measured with a relative molecular mass less than the average molecular mass of air, it will gather at the high place of the coal bunker in the coal bunker; therefore, the closer the height of the monitoring area is to the top of the coal bunker, the denser the distribution of the gas; and the greater the difference in molecular mass, the more concentrated the distribution; the larger the preliminary distribution score set in the corresponding area.

[0028] For the gas to be measured with a relative molecular mass greater than the average molecular mass of air, it will gather at the low place of the coal bunker in the coal bunker; therefore, the closer the height of the monitoring area is to the bottom of the coal bunker, the denser the distribution of the gas; and the greater the difference in molecular mass, the more concentrated the distribution; the larger the preliminary distribution score set in the corresponding area.

[0029] Further, correct the preliminary distribution score based on the storage environment data and coal quality data to obtain the distribution score that conforms to the current actual storage situation, including:

[0030] Extract the current values of several environmental items from the internal environment data of the coal bunker and label them as HJk, where k is the number of the environmental item; and the volatility of each measured gas in the coal quality data of the coal stacked inside the coal bunker and its corresponding optimal environmental conditions; extract the optimal values of each environmental item corresponding to the optimal environmental conditions; label the volatility of the measured gas as HFj; label the optimal values of each environmental item corresponding to it as ZJjk; the volatility of the measured gas is the ratio of the amount of the measured gas generated by the volatilization of the coal per unit volume to the amount of all gases generated under the condition that the values of each environmental item correspond to the optimal values; the larger the ratio, the easier the coal is to release the measured gas and the larger the release amount.

[0031] Through the formula:

[0032]

[0033] Where PFij represents the distribution score corresponding to the measured gas numbered j in the monitoring area numbered i; αk represents the weight coefficient corresponding to the environmental item parameter; MT is the volume of the stacked coal.

[0034] After that, calculate the distribution scores of each measured gas in each monitoring area through the above formula.

[0035] Furthermore, generate the number of gas sensors in each monitoring area based on the distribution scores of the measured gas in each monitoring area, including:

[0036] Obtain the distribution scores of the measured gas in each monitoring area; sum up the distribution scores corresponding to each monitoring area of the measured gas to obtain the total score corresponding to the measured gas; set the number of gas sensors corresponding to the measured gas in each monitoring area according to the ratio of the distribution score corresponding to the current monitoring area of the measured gas to the total score corresponding to the measured gas.

[0037] Furthermore, the concentration evaluation model is obtained through the training of an artificial intelligence model, including:

[0038] Obtain several monitoring data, the corresponding number of each sensor, and the concentration risk score of each measured gas in each monitoring area from the database; the concentration risk score is evaluated by experts on several monitoring data and the corresponding number of each gas sensor; the greater the concentration of the measured gas obtained in each monitoring area, and the greater the number of gas sensors corresponding to the measured gas in each monitoring area, it indicates that the occurrence risk probability of the measured gas in this monitoring area is greater; more gas sensors are required for monitoring, so the concentration risk score corresponding to the measured gas in each monitoring area is greater; integrate the number of each gas sensor corresponding to the monitoring data with its corresponding concentration risk score into several training data and test data.

[0039] Import a number of training data into the artificial intelligence model for training, and test the trained artificial intelligence model with test data; specifically, input the monitoring data in the test data into the trained artificial intelligence model to output a concentration risk score. Check whether the absolute value of the difference between the concentration risk score and the concentration risk score recorded in the test data is within the acceptable range; if so, it means that this set of test data passes the test, and continue to test the next set of test data; if not, relevant parameters of the artificial intelligence model need to be adjusted, and continue to use this set of test data for testing; until a set proportion of the test data passes the test; finally, obtain a concentration evaluation model with the input being the monitoring data and the corresponding number of each sensor and its corresponding concentration risk score, and the output being the concentration risk score; where the artificial intelligence model is a BP neural network model, etc.

[0040] The monitoring data is the concentration of the measured gas in each monitoring area; the measured gas is methane, hydrogen sulfide, carbon monoxide, etc.

[0041] Further, set the data collection instruction for the corresponding measured gas based on the concentration risk score, including:

[0042] Obtain the corresponding concentration risk scores of the measured gas in each monitoring area for the monitoring data in this cycle; draw a distribution scatter plot based on the corresponding concentration risk scores of the measured gas in each monitoring area for the monitoring data and set the data collection instruction for the next cycle.

[0043] Further, draw a distribution scatter plot based on the corresponding concentration risk scores of the measured gas in each monitoring area for the monitoring data and set the data collection instruction for the next cycle, including:

[0044] Draw data points in the rectangular coordinate system in sequence through the corresponding concentration risk scores of the measured gas.

[0045] When the overall data point distribution of the gas distribution scatter plot shows an upward trend, shorten and adjust the data collection interval for this time according to the set time step to obtain the data collection interval for the next data collection; after the duration of the data collection interval, generate a data collection instruction and send it to the data collection module.

[0046] When the overall data point distribution of the gas distribution scatter plot shows a downward trend, extend and adjust the data collection interval for this time according to the set time step to obtain the data collection interval for the next data collection; after the duration of the data collection interval, generate a data collection instruction and send it to the data collection module.

[0047] Further, generate a corresponding alarm signal based on the concentration risk score, including:

[0048] Obtain the gas safety threshold from the database; obtain the ratio relationship between the concentration risk scores of the measured gas in each monitoring area and their corresponding gas safety thresholds through the monitoring data of this cycle; when the concentration risk score of the measured gas is greater than 1 compared to the corresponding gas safety threshold, a danger signal is issued; when the concentration risk score of the measured gas is less than 1 compared to the corresponding gas safety threshold, a safety signal is issued; the alarm signals include: danger signals and safety signals.

[0049] The second aspect of the embodiments of the present application provides an AI-based intelligent safety monitoring method for coal bunkers, including the following steps:

[0050] Step 1: Obtain the characteristic data of several measured gases, and set the monitoring plan for each measured gas in the coal bunker based on the characteristic data;

[0051] Step 2: Install each gas sensor based on the monitoring plan;

[0052] Step 3: Collect the monitoring data of each gas sensor based on the data collection instruction;

[0053] Step 4: Obtain several monitoring data corresponding to each measured gas, and input the monitoring data into the concentration evaluation model to obtain the concentration risk score for evaluating the danger level of the corresponding measured gas;

[0054] Step 5: Set the data collection instruction for the corresponding measured gas based on the concentration risk score;

[0055] Step 6: Generate the corresponding alarm signal based on the concentration risk score and issue an alarm.

[0056] Compared with the prior art, the beneficial effects of the present application are:

[0057] 1. In the present application, by obtaining the characteristic data of several measured gases, setting the monitoring plan for each measured gas in the coal bunker based on the characteristic data, installing each gas sensor based on the monitoring plan; collecting the monitoring data of each gas sensor based on the data collection instruction; obtaining several monitoring data corresponding to each measured gas, inputting the monitoring data into the concentration evaluation model to obtain the concentration risk score for evaluating the danger level of the corresponding measured gas; setting the data collection instruction for the corresponding measured gas based on the concentration risk score; generating the corresponding alarm signal based on the concentration risk score and issuing an alarm; setting sensors specifically according to the characteristics of different measured gases and the characteristics of the stored coal; greatly reducing the number of sensors, and thus greatly reducing the data processing volume during subsequent monitoring data processing; while ensuring the accuracy and integrity of the monitoring results of dangerous gases, improving the detection efficiency of dangerous gases in the coal bunker.

[0058] 2. The present application corrects the preliminary distribution score by storing environmental data and coal quality data to obtain the distribution score; correcting the preliminary distribution score to obtain the distribution score greatly improves the accuracy of the distribution score, taking into account the correction of the preliminary distribution score by environmental-related factors and coal quality-related factors, and improving the accuracy for subsequent installation of gas sensors;

[0059] 3. The present application obtains the respective concentration risk scores of the measured gas corresponding to each monitoring area, draws a distribution scatter plot, and sets the data acquisition instruction for the next cycle; setting the acquisition time interval for the measured gas in advance helps to save time and adjust the monitoring frequency, improving the monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0061] Figure 1 It is a schematic diagram of the principle of the present application;

[0062] Figure 2 It is a schematic diagram of the method of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] The following will clearly and completely describe the technical solutions of the present application in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0064] Embodiment 1:

[0065] Please refer to Figure 1 , the first aspect embodiment of the present application provides an AI-based intelligent safety monitoring system for coal bunkers, including: a monitoring planning module, a data acquisition module, a safety monitoring module, an alarm module, and a database;

[0066] Monitoring planning module: Obtain the characteristic data of several measured gases, set the monitoring plans for each measured gas in the coal bunker based on the characteristic data, and install each gas sensor based on the monitoring plan;

[0067] Data acquisition module: Collect the monitoring data of each gas sensor based on the data acquisition instruction; wherein, the data acquisition instruction is a signal to start data acquisition;

[0068] Safety monitoring module: Obtain a number of monitoring data corresponding to each measured gas, input the monitoring data into the concentration evaluation model to obtain a concentration risk score for evaluating the risk level of the corresponding measured gas; Set the data acquisition instruction for the corresponding measured gas based on the concentration risk score;

[0069] Alarm module: Generate corresponding alarm signals based on the concentration risk score and issue an alarm.

[0070] In this application, the coal bunker is enclosed.

[0071] Specifically, the data acquisition module is further configured to obtain the environmental data and coal quality data inside the coal bunker.

[0072] Specifically, setting the monitoring plan for the measured gas in the coal bunker based on the characteristic data includes:

[0073] Obtain the relative molecular mass of each measured gas in the characteristic data; Obtain the storage environment data and coal quality data of the coal bunker through the data acquisition module;

[0074] Divide the coal bunker into several monitoring areas according to height;

[0075] Generate the preliminary distribution score indicating the distribution of the measured gas in each monitoring area based on the relative molecular mass of the measured gas and the average molecular mass of air;

[0076] Correct the preliminary distribution score based on the storage environment data and coal quality data to obtain the distribution score that conforms to the current actual storage situation;

[0077] Generate the number of gas sensors in each monitoring area based on the distribution score of the measured gas in each monitoring area;

[0078] Successively obtain the number of gas sensors of each measured gas in each monitoring area; and integrate them into a monitoring plan.

[0079] The environmental data includes the temperature, humidity, etc. inside the coal bunker; the coal quality data is the volatility of each measured gas inside the coal bunker under the optimal environmental conditions; the characteristic data includes: the relative molecular mass, density, etc. of the measured gas; the monitoring data is the concentration of the measured gas in each monitoring area; the measured gas is methane, hydrogen sulfide, carbon monoxide, etc.; the monitoring plan is the number of gas sensors installed in each monitoring area inside the coal bunker; the preliminary distribution score is the preliminary characteristic result of the measured gas in each area; the distribution score is the distribution result of the measured gas after correction to conform to the current storage situation; the data acquisition instruction is the instruction for the gas sensor to receive and collect the monitoring data; the gas sensor includes a coal pile sensor, a methane sensor, a temperature sensor, etc.

[0080] Specifically, generating the preliminary distribution score of the gas to be measured in each monitoring area based on the relative molecular mass of the gas to be measured and the average molecular mass of air includes:

[0081] Obtaining the relative molecular mass of several gases to be measured as MAj, where j is the number of the gas to be measured; the average relative molecular mass of air is MK; j = 1, 2,..., n; n represents the number of gases to be measured; the total height of the internal area of the coal bunker is H, which is divided into h layers at equal heights; i = 1, 2,..., h; i represents the number of the monitoring area;

[0082] Through the formula:

[0083]

[0084] where Pij represents the preliminary distribution score corresponding to the j-th gas to be measured in the i-th monitoring area; represents the height corresponding to the i-th monitoring area;

[0085] After that, calculate the preliminary distribution scores of each gas to be measured in each monitoring area through the above formula,

[0086] In this embodiment, it should be noted that for the gas to be measured with a relative molecular mass less than the average relative molecular mass of air, it will gather at the high place of the coal bunker; therefore, the closer the height of the monitoring area is to the top of the coal bunker, the denser the distribution of the gas; and the greater the difference in molecular mass, the more concentrated the distribution; and the greater the preliminary distribution score set in the corresponding area.

[0087] For the gas to be measured with a relative molecular mass greater than the average relative molecular mass of air, it will gather at the low place of the coal bunker; therefore, the closer the height of the monitoring area is to the bottom of the coal bunker, the denser the distribution of the gas; and the greater the difference in molecular mass, the more concentrated the distribution; and the greater the preliminary distribution score set in the corresponding area.

[0088] In this embodiment, the relative molecular mass of methane gas is 16, and the average relative molecular mass of air is about 29; the relative molecular mass of methane gas is less than the average relative molecular mass of air, and methane will gather at the high place of the coal bunker; the relative molecular mass of hydrogen sulfide is 34, and the relative molecular mass of hydrogen sulfide is greater than the average relative molecular mass of air, so hydrogen sulfide will gather at the bottom of the coal bunker.

[0089] Specifically, correcting the preliminary distribution score based on the storage environment data and coal quality data to obtain the distribution score, including:

[0090] Extract the current values of several environmental items in the internal environment data of the coal bunker, and mark them as HJk, where k is the number of the environmental item; and the volatility of each measured gas in the coal quality data of the coal stacked inside the coal bunker and its corresponding optimal environmental conditions; extract the optimal values corresponding to each environmental item in the optimal environmental conditions; mark the volatility of the measured gas as HFj; mark the optimal values corresponding to each environmental item as ZJjk; the volatility of the measured gas is the value measured for each environmental item under the optimal environmental conditions, which is the ratio of the amount of the measured gas generated by the volatilization of the coal per unit volume to the total amount of all gases generated; the larger the ratio, the easier the coal is to release the measured gas and the larger the release amount; the current value is the value corresponding to each current environmental item in the internal environment data of the coal bunker; the optimal environmental condition is the optimal value corresponding to each environmental item for the measured gas; the optimal value is the optimal value corresponding to each environmental item in the internal environment data of the coal bunker

[0091] Through the formula:

[0092]

[0093] Wherein, PFij represents the distribution score corresponding to the measured gas numbered j in the monitoring area numbered i; αk represents the weight coefficient corresponding to the environmental item parameter; MT is the volume of the stacked coal; K represents the total number of environmental items; the weight coefficient corresponding to the environmental item parameter adjusts the volatility of the measured gas through the environmental item of the measured gas;

[0094] In this embodiment, it should be specifically noted that the greater the difference between the current value and the optimal value of each environmental item corresponding to the measured gas in each monitoring area in the above formula; the greater the volatility of the corresponding measured gas, and the denser the distribution of the measured gas; the greater the distribution score of each monitoring area corresponding to the measured gas, and vice versa;

[0095] After that, calculate the distribution scores of each measured gas in each monitoring area through the above formula.

[0096] Specifically, generating the number of gas sensors for each monitoring area based on the distribution scores of the measured gas in each monitoring area includes:

[0097] Obtain the distribution scores of the measured gas in each monitoring area; sum the distribution scores corresponding to each monitoring area of the measured gas to obtain the total score corresponding to the measured gas; set the number of gas sensors corresponding to the measured gas in each monitoring area by the ratio of the distribution score corresponding to the current monitoring area of the measured gas to the total score corresponding to the measured gas;

[0098] In this embodiment, the larger the ratio of the distribution score corresponding to the current monitoring area of the measured gas to the total score corresponding to the measured gas, the more gas sensors are installed in the monitoring area for the measured gas. For example, when the ratio of the distribution score corresponding to the current monitoring area of the measured gas to the total score corresponding to the measured gas is denoted as AK; for the gas sensor installation adjustment factor a, the number of gas sensors CGQ in this monitoring area is calculated by the formula CGQ = AK × a; when AK = 0.3 and a = 100, then CGQ = 30; specific data is set according to experience.

[0099] Specifically, inputting the monitoring data into the concentration evaluation model to obtain a concentration risk score for evaluating the risk level of the corresponding measured gas includes:

[0100] The concentration evaluation model is obtained through training of an artificial intelligence model;

[0101] Obtain a number of monitoring data, corresponding numbers of each sensor, and corresponding concentration risk scores of each measured gas in each monitoring area from the database; the concentration risk score is evaluated by experts based on a number of monitoring data and the corresponding numbers of each gas sensor; the greater the concentration of the measured gas in each monitoring area, the more gas sensors corresponding to the measured gas in each monitoring area, indicating that the risk probability of the measured gas occurring in this monitoring area is greater; more gas sensors are required for monitoring, so the greater the concentration risk score corresponding to the measured gas in each monitoring area; integrate the numbers of each gas sensor corresponding to the monitoring data with their corresponding concentration risk scores into a number of training data and test data;

[0102] Import a number of training data into the artificial intelligence model for training, and test the trained artificial intelligence model with the test data. Specifically, input the monitoring data in the test data into the trained artificial intelligence model to output a concentration risk score. Check whether the absolute value of the difference between the concentration risk score and the concentration risk score recorded in the test data is within an acceptable range; if yes, it means that this set of test data passes the test, and continue to test the next set of test data; if not, relevant parameters of the artificial intelligence model need to be adjusted, and continue to test with this set of test data; until a set proportion of the test data passes the test; finally, obtain a concentration evaluation model with the monitoring data, the corresponding numbers of each sensor, and their corresponding concentration risk scores as inputs and the concentration risk score as the output; where the artificial intelligence model is a BP neural network model, etc.

[0103] Specifically, generating a corresponding alarm signal based on the concentration risk score includes:

[0104] Obtain the gas safety threshold from the database; obtain the ratio relationship between the concentration risk scores of the measured gas in each monitoring area and their corresponding gas safety thresholds through the monitoring data of this cycle; when the concentration risk score of the measured gas is greater than 1 compared to the corresponding gas safety threshold, a danger signal is issued; when the concentration risk score of the measured gas is less than 1 compared to the corresponding gas safety threshold, a safety signal is issued; the alarm signals include: danger signals and safety signals.

[0105] Please refer to Figure 2 , the second aspect embodiment of this application provides an AI-based intelligent safety monitoring method for coal bunkers, including the following steps:

[0106] Step 1: Obtain the characteristic data of several measured gases, and set the monitoring plan for each measured gas in the coal bunker based on the characteristic data;

[0107] Step 2: Install each gas sensor based on the monitoring plan;

[0108] Step 3: Collect the monitoring data of each gas sensor based on the data acquisition instruction;

[0109] Step 4: Obtain several monitoring data corresponding to each measured gas, and input the monitoring data into the concentration evaluation model to obtain the concentration risk score for evaluating the danger level of the corresponding measured gas;

[0110] Step 5: Generate corresponding alarm signals based on the concentration risk score and issue an alarm.

[0111] Embodiment 2:

[0112] Different from Embodiment 1, the data acquisition instruction in this embodiment is obtained in the following manner:

[0113] Set the data acquisition instruction for the corresponding measured gas based on the concentration risk score, including:

[0114] Obtain the concentration risk scores corresponding to the measured gas in each monitoring area from the monitoring data of this cycle; draw a distribution scatter plot based on the concentration risk scores of the corresponding measured gas obtained from the monitoring data in each monitoring area and set the data acquisition instruction for the next cycle;

[0115] Specifically, draw a distribution scatter plot based on the concentration risk scores of the corresponding measured gas obtained from the monitoring data in each monitoring area and set the data acquisition instruction for the next cycle, including:

[0116] Draw data points in the rectangular coordinate system in sequence through the concentration risk scores corresponding to the measured gas;

[0117] When the overall data point distribution of the gas distribution scatter plot shows an upward trend, shorten and adjust the current data collection interval according to the set time step to obtain the data collection interval for the next data collection; after the duration of the data collection interval has passed, generate a data collection instruction and send it to the data collection module;

[0118] When the overall data point distribution of the gas distribution scatter plot shows a downward trend, extend and adjust the current data collection interval according to the set time step to obtain the data collection interval for the next data collection; after the duration of the data collection interval has passed, generate a data collection instruction and send it to the data collection module;

[0119] In this embodiment, when it shows an upward trend, fit the distribution scatter plot into a distribution curve, then multiply the slope of the curve by the time step, and subtract the product of the step size and the absolute value of the slope from the current collection interval; calculate the data collection interval for the next data collection, denoted as T1, through the formula T1 = T - |KD|×t; T represents the current collection interval; KD represents the slope of the distribution scatter plot corresponding to the measured gas;

[0120] When it shows a downward trend, fit the distribution scatter plot into a distribution curve, multiply the slope of the curve by the time step, and add the product of the step size and the absolute value of the slope to the current collection interval. Calculate the data collection interval for the next data collection, denoted as T2, through the formula T2 = T + |KD|×t.

[0121] An AI-based intelligent safety monitoring method for coal bunkers in this embodiment includes the following steps:

[0122] Step 1: Obtain the characteristic data of several measured gases, and set the monitoring schemes for each measured gas in the coal bunker based on the characteristic data;

[0123] Step 2: Install each gas sensor based on the monitoring scheme;

[0124] Step 3: Collect the monitoring data of each gas sensor based on the data collection instruction;

[0125] Step 4: Obtain several monitoring data corresponding to each measured gas, and input the monitoring data into the concentration evaluation model to obtain a concentration risk score for evaluating the danger level of the corresponding measured gas;

[0126] Step 5: Set the data collection instruction for the corresponding measured gas based on the concentration risk score;

[0127] Step 6: Generate a corresponding alarm signal based on the concentration risk score and give an alarm.

[0128] Some of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is obtained by software simulation of a large amount of collected data to get a formula that is closest to the actual situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0129] The working principle of this application: This application obtains the characteristic data of several measured gases, sets the monitoring scheme for each measured gas in the coal bunker based on the characteristic data, and installs each gas sensor based on the monitoring scheme; collects the monitoring data of each gas sensor based on the data collection instruction; obtains several monitoring data corresponding to each measured gas, inputs the monitoring data into the concentration evaluation model to obtain a concentration risk score for evaluating the risk level of the corresponding measured gas; sets the data collection instruction for the corresponding measured gas based on the concentration risk score; generates a corresponding alarm signal based on the concentration risk score and issues an alarm; sets the sensors specifically according to the characteristics of different measured gases and the characteristics of the stored coal; greatly reduces the number of sensors, and thus greatly reduces the amount of data processing in subsequent processing of monitoring data; while ensuring the accuracy and integrity of the monitoring results of dangerous gases, improves the detection efficiency of dangerous gases in the coal bunker.

[0130] The above embodiments are only used to illustrate the technical method of this application and not to limit it. Although this application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of this application can be modified or equivalently replaced without departing from the spirit and scope of the technical method of this application.

Claims

1. An AI-based intelligent safety monitoring system for coal bunkers, comprising: A monitoring planning module, a data acquisition module, a safety monitoring module, an alarm module, and a database, characterized in that The monitoring planning module: obtains characteristic data of a plurality of measured gases, sets monitoring plans for each measured gas in the coal bunker based on the characteristic data, and installs each gas sensor based on the monitoring plan; The data acquisition module: acquires monitoring data of each gas sensor based on a data acquisition instruction; obtains environmental data and coal quality data inside the coal bunker; wherein, the data acquisition instruction is a signal to start data acquisition; The safety monitoring module: obtains a plurality of monitoring data corresponding to each measured gas, and inputs the monitoring data into a concentration evaluation model to obtain a concentration risk score for evaluating the danger level of the corresponding measured gas; The alarm module: generates a corresponding alarm signal based on the concentration risk score and gives an alarm.

2. The intelligent safety monitoring system for coal bunker based on AI according to claim 1, characterized in that, Setting the monitoring plan for the detected gases in the coal bunker based on the characteristic data includes: Obtaining the relative molecular mass of each detected gas in the characteristic data; as well as the storage environment data and coal quality data of the coal bunker; Dividing the coal bunker into a plurality of monitoring areas according to height; Generating a preliminary distribution score for indicating the distribution of the measured gas in each monitoring area based on the relative molecular mass of the measured gas and the average molecular mass of air; Correcting the preliminary distribution score based on the storage environment data and coal quality data to obtain a distribution score that conforms to the current actual storage situation; Generating the number of gas sensors in each monitoring area based on the distribution score of the measured gas in each monitoring area; Sequentially obtaining the number of gas sensors of each measured gas in each monitoring area; and integrating them into a monitoring plan.

3. The intelligent safety monitoring system for coal bunker based on AI according to claim 2, characterized in that, Generating the preliminary distribution score for indicating the distribution of the measured gas in each monitoring area based on the relative molecular mass of the measured gas and the average molecular mass of air, including: Obtaining the relative molecular mass of a plurality of measured gases as MAj, where j is the number of the measured gas; the average relative molecular mass of air is MK; j = 1, 2,..., n; n represents the number of measured gases; the total height of the internal area of the coal bunker is H, which is divided into h layers according to equal height; i = 1, 2,..., h; i represents the number of the monitoring area; Through the formula: Among them, Pij represents the preliminary distribution score corresponding to the measured gas j in the i-th monitoring area; represents the height corresponding to the i-th monitoring area; After that, calculate the preliminary distribution score of each measured gas in each monitoring area through the above formula.

4. An AI-based intelligent safety monitoring system for coal bunkers according to claim 2, characterized in that, Correcting the preliminary distribution score based on the storage environment data and coal quality data to obtain the distribution score that conforms to the current actual storage situation, including: Extracting the current values of several environmental items in the internal environment data of the coal bunker and marking them as HJk, where k is the number of the environmental item; as well as the volatility of each measured gas and its corresponding optimal environmental conditions in the coal quality data of the coal stacked inside the coal bunker; extracting the optimal values of each environmental item corresponding to the optimal environmental conditions; marking the volatility of the measured gas as HFj; marking the optimal values of each environmental item corresponding to it as ZJjk; Through the formula: Wherein, PFij represents the distribution score corresponding to the measured gas numbered j in the monitoring area numbered i; αk represents the weight coefficient corresponding to the environmental item parameter; MT is the volume of the stacked coal; After that, the distribution scores of each measured gas in each monitoring area are calculated by the above formula.

5. An AI-based intelligent safety monitoring system for coal bunkers according to claim 2, characterized in that, Generating the number of gas sensors for each monitoring area based on the distribution scores of the measured gas in each monitoring area, including: Obtaining the distribution scores of the measured gas in each monitoring area; summing up the distribution scores corresponding to each monitoring area of the measured gas to obtain the total score corresponding to the measured gas; setting the number of gas sensors corresponding to each monitoring area of the measured gas by the ratio of the distribution score corresponding to the current monitoring area of the measured gas to the total score corresponding to the measured gas.

6. An AI-based intelligent safety monitoring system for coal bunkers according to claim 1, characterized in that, The concentration evaluation model is obtained by training an artificial intelligence model, including: Obtaining a number of monitoring data of each measured gas in each monitoring area from the database, and the corresponding number of each sensor and the concentration risk score; integrating the monitoring data and the corresponding concentration risk score into a number of training data and test data; Importing a number of training data into the artificial intelligence model for training, and testing the trained artificial intelligence model with the test data; finally obtaining a concentration evaluation model with the monitoring data and the corresponding number of each sensor and the corresponding concentration risk score as the input and the concentration risk score as the output; where the artificial intelligence model is a BP neural network model.

7. An AI-based intelligent safety monitoring system for coal bunkers according to claim 1, characterized in that, The safety monitoring module is also used for: setting the data acquisition instruction corresponding to the measured gas based on the concentration risk score.

8. An AI-based intelligent safety monitoring system for coal bunkers according to claim 1 or 7, characterized in that, Setting the data acquisition instruction corresponding to the measured gas based on the concentration risk score, including: Obtaining the concentration risk scores corresponding to the measured gas in each monitoring area from the monitoring data of this cycle; drawing a distribution scatter plot according to the concentration risk scores corresponding to the measured gas in each monitoring area of the monitoring data and setting the data acquisition instruction for the next cycle.

9. An AI-based intelligent safety monitoring system for coal bunkers according to claim 8, characterized in that Drawing a distribution scatter plot according to the concentration risk scores corresponding to the measured gas in each monitoring area of the monitoring data and setting the data acquisition instruction for the next cycle, including: Drawing data points in the rectangular coordinate system in sequence through the concentration risk scores corresponding to the measured gas; When the overall data point distribution of the gas distribution scatter plot shows an upward trend, shortening and adjusting the data acquisition interval of this time according to the set time step to obtain the data acquisition interval for the next data acquisition; after the duration of the data acquisition interval, generating a data acquisition instruction and sending it to the data acquisition module; When the overall data point distribution of the gas distribution scatter plot shows a downward trend, extending and adjusting the data acquisition interval of this time according to the set time step to obtain the data acquisition interval for the next data acquisition; after the duration of the data acquisition interval, generating a data acquisition instruction and sending it to the data acquisition module.

10. An AI-based intelligent safety monitoring method for coal bunkers, which is applied to the operation of an AI-based intelligent safety monitoring system described in any one of claims 1-6; characterized in that, Including the following steps: Step 1: Obtain the characteristic data of a number of measured gases, and set the monitoring scheme for each measured gas in the coal bunker based on the characteristic data; Step 2: Install each gas sensor based on the monitoring scheme; Step 3: Collect the monitoring data of each gas sensor based on the data acquisition instruction. Step 4: Obtain a number of monitoring data corresponding to each gas to be measured, and input the monitoring data into the concentration assessment model to obtain a concentration risk score for evaluating the risk level of the corresponding gas to be measured; Step 5: Generate a corresponding alarm signal based on the concentration risk score and issue an alarm.

Images