Generative coal mine production environment risk troubleshooting decision-making method

Abstract

The invention discloses a generative coal mine production environment risk troubleshooting decision-making method, which relates to the technical field of risk prediction and comprises the following steps of S1, determining a sensor number, S2, dividing a time sequence, S3, analyzing a space sequence, S4, calculating a risk level and S5, predicting risk data. According to the invention, the data collected by each sensor is divided according to two forms of time and space, so that an operator can check historical data according to different time intervals and determine risk levels corresponding to minutes, hours, days, months, quarters and years; concentration concepts corresponding to data related to each sensor, each working face, each mining area and each mining area are actively analyzed, so that an operator can intuitively obtain the risk level of the whole coal mining area according to different angles, and meanwhile, the risk level of the coal mining area can be predicted in advance; and an operator can take preventive measures in time to ensure the production safety.

Description

Technical Field

[0001] The present invention relates to the technical field of risk prediction, and in particular to a generative risk investigation and decision-making method for coal mine production environment. Background Technique

[0002] For coal mine risk investigation, in order to improve the safety production level of coal mines, prevent and reduce accidents, and ensure the stable and healthy development of coal mine enterprises, and at the same time identify and evaluate potential safety risks, which helps to discover and solve loopholes and weak links in management. In the invention patent with the application number 202311764824.1, "A safety risk assessment method and system for coal mine enterprises, which relates to the field of computer technology. Among them, the method includes: constructing a risk assessment index system for coal mine enterprises, and determining the static risk assessment score of the coal mine corresponding to the risk assessment index system of coal mine enterprises through fuzzy comprehensive evaluation; determining a set of coal mine risk points, and conducting dynamic risk assessment on the set of coal mine risk points to obtain the dynamic risk assessment score of the set of coal mine risk points; determining the safety risk assessment level of the coal mine enterprise according to the static risk assessment score and the dynamic risk assessment score of the coal mine. The present disclosure adopting the above solution can conduct safety risk assessment of coal mine enterprises and realize reasonable classification and control of coal mine enterprise risks."

[0003] The above-mentioned prior art solves problems such as unreasonable risk assessment in coal mine safety management. However, during the data collection process, the distance between each sensor and the working face is not statistically counted, making it impossible to determine the location of areas with high risk in a short time. At the same time, the data is not divided in both time and space forms, and the operator cannot intuitively obtain the corresponding risk levels of each area and different time intervals. This method also cannot predict the risk levels within a certain period in the future. Summary of the Invention

[0004] The purpose of the present invention is to provide a generative risk investigation and decision-making method for coal mine production environment to solve the problems raised in the above background technique.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A generative risk investigation and decision-making method for coal mine production environment, including the following steps:

[0006] S1. Determine the sensor numbers: Number each working face and heading face in all coal mining areas one by one, arrange multiple gas sensors near each working face and heading face, number the gas sensors, count the distances between each gas sensor and the working face and heading face, and transmit them to the database;

[0007] S2. Divide the time series: Store the collected data in the gas dataset in the form of a time series in sequence. After setting multiple time windows, use the time series segmentation method to divide the data in the gas dataset to obtain multiple time slice information. Then, use the cloud transformation algorithm to convert these time slice information into corresponding granular time series respectively, and transmit them to the storage device;

[0008] S3. Analyze the spatial sequence: Determine the working face with the shortest distance from each sensor, analyze the corresponding sensor numbers for each face. After setting multiple spatial windows, divide the dataset according to sensors, working faces, mining areas, and mining regions to obtain multiple spatial slice information. Then, use the cloud transformation algorithm to convert these spatial slice information into corresponding granular spatial series respectively, and transmit them to the storage device;

[0009] S4. Calculate the risk level: Obtain the data in the gas dataset. After performing distribution tests on these data, use the skewness analysis algorithm to calculate the skewness value corresponding to the distribution function. Analyze multiple cloud parameters based on the Gaussian distribution function, fit these cloud parameters to generate a gas concentration concept distribution map. Divide the concentration into four concepts according to the gas content. Use the gas concentration concept distribution map to determine the relevant parameters and risk levels corresponding to each concentration concept. After extracting the granular time series and granular spatial series from the storage device, analyze the risk levels corresponding to each time granular layer and spatial granular layer according to the relevant parameters corresponding to each concentration concept;

[0010] S5. Predict risk data: Obtain the relevant parameters corresponding to all granular time series, use these parameters for analysis to obtain a development matrix. Add the development matrix to the variable cloud prediction formula to obtain a corresponding prediction sequence. After determining the corresponding risk level, output the risk level and the prediction sequence through the visualization interface.

[0011] Preferably, the step S2 includes the following steps:

[0012] S201. After obtaining the data collected by each gas sensor, store these data in the gas dataset in the form of a time series in sequence;

[0013] S202. Since the data recording interval of the sensor is one minute, after setting multiple time windows, use the time series segmentation method to divide the data in the gas dataset according to minutes, hours, days, months, seasons, and years respectively to obtain minute time slice information, hour time slice information, day time slice information, month time slice information, season time slice information, and year time slice information.

[0014] Preferably, the step S2 further includes the following steps:

[0015] S203. Respectively convert the minute time slice information, hour time slice information, day time slice information, month time slice information, quarter time slice information, and year time slice information into corresponding granular time series through the cloud transformation algorithm, and transmit them to the storage device. The cloud transformation algorithm includes a certainty analysis algorithm, and the certainty analysis algorithm is specifically:

[0016]

[0017] where E n ′ represents a normal random number, Ex represents the expected value, x represents the quantitative value, y represents the certainty, and e represents the natural constant;

[0018] The cloud transformation algorithm includes a quantitative value analysis algorithm, and the quantitative value analysis algorithm is specifically:

[0019]

[0020] where x represents the quantitative value, Ex represents the expected value, En′ represents a normal random number, and y represents the certainty.

[0021] Preferably, the step S3 specifically includes the following steps:

[0022] S301. After obtaining the data collected by each gas sensor and the distance from the corresponding sensor to each working face, calculate the working face with the shortest distance for each sensor, and associate the corresponding face number with the sensor number, so as to obtain the sensor number corresponding to each face;

[0023] S302. After setting multiple spatial windows, divide the data set according to sensors, working faces, mining areas, and mining regions to obtain sensor spatial slice information, working face spatial slice information, mining area spatial slice information, and mining region spatial slice information.

[0024] Preferably, the step S3 specifically further includes the following steps:

[0025] S303. Respectively convert the sensor spatial slice information, working face spatial slice information, mining area spatial slice information, and mining region spatial slice information into corresponding granular spatial series through the cloud transformation algorithm, and transmit them to the storage device.

[0026] Preferably, the step S4 specifically includes the following steps:

[0027] S401. Obtain the data in the gas data set, perform distribution tests on these data, select the number of peaks of the distribution function corresponding to the data as the initial frequency, and use the skewness analysis algorithm to calculate the skewness value corresponding to the distribution function. If the skewness value is greater than the threshold, perform logarithmic processing on the current distribution function, otherwise do not perform logarithmic processing;

[0028] S402. Convert the current distribution function into multiple Gaussian distribution functions, calculate multiple cloud parameters based on the Gaussian distribution functions and the relevant parameters of the standard deviation. If the mixing degree of the cloud parameters is greater than the upper limit of the mixing degree, adjust the mixing degree; otherwise, fit these cloud parameters to obtain the conceptual distribution map of the gas concentration.

[0029] Preferably, the step S4 specifically further includes the following steps:

[0030] S403. Divide the concentration into four concepts according to the gas content, namely low concentration, medium concentration, medium-high concentration and high concentration. After determining the relevant parameters corresponding to each concentration concept according to the conceptual distribution map of the gas concentration, determine that the low concentration concept corresponds to the first risk level, the medium concentration concept corresponds to the second risk level, the medium-high concentration concept corresponds to the third risk level, and the high concentration concept corresponds to the fourth risk level;

[0031] S404. After extracting the particle time series in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all particle time series to determine the risk levels corresponding to the minute particle layer, hour particle layer, day particle layer, month particle layer, quarter particle layer and year particle layer;

[0032] S405. After extracting the particle space series in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all particle space series to determine the risk levels corresponding to the transmitter particle layer, working face particle layer, mining area particle layer and mining area particle layer.

[0033] Preferably, the step S5 specifically includes the following steps:

[0034] S501. Obtain the relevant parameters corresponding to all particle time series, where the relevant parameters include the expectation, entropy and hyper-entropy corresponding to the series. Use these parameters for analysis to obtain the development matrix, and add the development matrix to the variable cloud prediction formula to obtain the corresponding prediction series;

[0035] S502. Compare the relevant parameters corresponding to each concentration concept with the relevant parameters corresponding to the prediction series. After determining the corresponding risk level, output the risk level and the prediction series through the visualization interface;

[0036] S503. When the predicted risk level is the third level and the fourth level, give a prompt through the intelligent alarm, and determine the specific time and sensor number corresponding to the prediction series.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] The present invention numbers all the working faces and gas sensors in a coal mining area, and counts the distances between each gas sensor and the working face and the heading face, so as to accurately analyze the specific position of each gas sensor. During the risk prediction process, once the risk level corresponding to the predicted data is too high, the number of the sensor and the specific time can be quickly determined, and then the distance from the sensor to the nearest working face can be calculated, which is convenient for maintenance personnel to take timely measures for intervention. At the same time, the data collected by each sensor is divided in two forms: time and space, so that the operator can check the historical data at different time intervals, clarify the risk levels corresponding to minutes, hours, days, months, quarters and years, and actively analyze the concentration concepts corresponding to the data related to each sensor, working face, mining area and mining area, which is convenient for the operator to intuitively obtain the risk level of the entire coal mining area from different angles. At the same time, according to the existing sensor-related data, the relevant parameters in the future period can be predicted. Therefore, the risk level of the coal mining area can also be predicted in advance to ensure that the operator can take preventive measures in time to ensure the safety of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 FIG. is a flowchart of the overall method provided by an embodiment of the present invention;

[0040] Figure 2 FIG. is a flowchart of the method for dividing time series provided by an embodiment of the present invention;

[0041] Figure 3 FIG. is a flowchart of the method for analyzing spatial series provided by an embodiment of the present invention;

[0042] Figure 4 FIG. is a flowchart of the method for calculating risk level provided by an embodiment of the present invention;

[0043] Figure 5 FIG. is a flowchart of the method for predicting risk data provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0045] Please refer to Figures 1-5 , the present invention provides a technical solution: a generative decision-making method for risk investigation in coal mine production environment, including the following steps:

[0046] S1. Determine the sensor numbers: Number each working face and heading face in all mining areas of multiple coal mines one by one. Arrange multiple gas sensors near each working face and heading face, number the gas sensors, count the distances between each gas sensor and the working face and heading face, and transmit them to the database;

[0047] S2. Divide the time series: Store the collected data in the gas dataset in the form of a time series in sequence. After setting multiple time windows, use the time series segmentation method to divide the data in the gas dataset to obtain multiple time slice information. Respectively convert these time slice information into corresponding granular time series through the cloud transformation algorithm, and transmit them to the storage device;

[0048] S3. Analyze the spatial sequence: Determine the working face with the shortest distance from each sensor, analyze the corresponding sensor numbers for each face. After setting multiple spatial windows, divide the dataset according to sensors, working faces, mining areas, and mining regions to obtain multiple spatial slice information. Respectively convert these spatial slice information into corresponding granular spatial series through the cloud transformation algorithm, and transmit them to the storage device;

[0049] S4. Calculate the risk level: Obtain the data in the gas dataset. After performing distribution tests on these data, use the skewness analysis algorithm to calculate the skewness value corresponding to the distribution function. Analyze multiple cloud parameters according to the Gaussian distribution function, fit these cloud parameters to generate a gas concentration concept distribution map. Divide the concentration into four concepts according to the gas content. Use the gas concentration concept distribution map to determine the relevant parameters and risk levels corresponding to each concentration concept. After extracting the granular time series and granular spatial series from the storage device, analyze the risk levels corresponding to each time granular layer and spatial granular layer according to the relevant parameters corresponding to each concentration concept;

[0050] S5. Predict risk data: Obtain the relevant parameters corresponding to all granular time series, use these parameters for analysis to obtain a development matrix, add the development matrix to the variable cloud prediction formula to obtain a corresponding prediction sequence. After determining the corresponding risk level, output the risk level and the prediction sequence through the visualization interface.

[0051] Step S2 includes the following steps:

[0052] S201. After obtaining the data collected by each gas sensor, store these data in the gas dataset in the form of a time series in sequence.

[0053] S202. Since the interval for the sensor to record data is one minute, after setting multiple time windows, the data in the gas dataset is divided according to minutes, hours, days, months, quarters, and years respectively using the time series segmentation method, obtaining minute time slice information, hour time slice information, day time slice information, month time slice information, quarter time slice information, and year time slice information;

[0054] Step S2 further includes the following steps:

[0055] S203. Respectively convert the minute time slice information, hour time slice information, day time slice information, month time slice information, quarter time slice information, and year time slice information into corresponding granular time series through the cloud transformation algorithm, and transmit them to the storage device. The cloud transformation algorithm includes a certainty analysis algorithm, and the certainty analysis algorithm is specifically:

[0056]

[0057] where, E n ′ represents a normal random number, Ex represents the expected value, x represents the quantitative value, y represents the certainty, and e represents the natural constant;

[0058] The cloud transformation algorithm includes a quantitative value analysis algorithm, and the quantitative value analysis algorithm is specifically:

[0059]

[0060] where, x represents the quantitative value, Ex represents the expected value, En′ represents a normal random number, and y represents the certainty;

[0061] Step S3 specifically includes the following steps:

[0062] S301. After obtaining the data collected by each gas sensor and the distance from the corresponding sensor to each working face, calculate the working face with the shortest distance for each sensor, and associate the corresponding face number with the sensor number, thereby obtaining the sensor number corresponding to each face;

[0063] S302. After setting multiple spatial windows, divide the dataset according to sensors, working faces, mining areas, and mining regions, obtaining sensor spatial slice information, working face spatial slice information, mining area spatial slice information, and mining region spatial slice information;

[0064] Step S3 specifically further includes the following steps:

[0065] S303. Respectively convert the sensor spatial slice information, working face spatial slice information, mining area spatial slice information, and mining region spatial slice information into corresponding granular spatial series through the cloud transformation algorithm, and transmit them to the storage device;

[0066] Step S4 specifically includes the following steps:

[0067] S401. Obtain the data in the gas dataset. After performing a distribution test on these data, select the number of peaks of the distribution function corresponding to the data as the initial frequency, and use the skewness analysis algorithm to calculate the skewness value corresponding to the distribution function. If the skewness value is greater than the threshold, perform logarithmic processing on the current distribution function; otherwise, do not perform logarithmic processing. The skewness analysis algorithm is specifically as follows:

[0068]

[0069] Among them, Δp represents the distribution skewness, x represents the data in the dataset, E(x) represents the standard deviation, and D(x) represents the variance;

[0070] S402. Convert the current distribution function into multiple Gaussian distribution functions, and calculate multiple cloud parameters based on the Gaussian distribution function and related standard deviation parameters. If the mixture degree of the cloud parameters is greater than the upper limit of the mixture degree, adjust the mixture degree; otherwise, fit these cloud parameters to obtain the gas concentration concept distribution map;

[0071] Step S4 specifically further includes the following steps:

[0072] S403. Divide the concentration into four concepts according to the gas content, namely low concentration, medium concentration, medium-high concentration, and high concentration. After determining the relevant parameters corresponding to each concentration concept based on the gas concentration concept distribution map, determine that the low concentration concept corresponds to the first risk level, the medium concentration concept corresponds to the second risk level, the medium-high concentration concept corresponds to the third risk level, and the high concentration concept corresponds to the fourth risk level;

[0073] S404. After extracting the granular time series in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all granular time series to determine the risk levels corresponding to the minute granular layer, hour granular layer, daily granular layer, monthly granular layer, quarterly granular layer, and annual granular layer;

[0074] S405. After extracting the granular space series in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all granular space series to determine the risk levels corresponding to the transmitter granular layer, working face granular layer, mining area granular layer, and mining area granular layer;

[0075] Step S5 specifically includes the following steps:

[0076] S501. Obtain the relevant parameters corresponding to all granular time series, where the relevant parameters include the expectation, entropy, and hyperentropy corresponding to the series. Use these parameters for analysis to obtain the development matrix, and add the development matrix to the variable cloud prediction formula to obtain the corresponding prediction series;

[0077] S502. Compare the relevant parameters corresponding to each concentration concept with the relevant parameters corresponding to the prediction sequence. After determining the corresponding risk level, output the risk level and the prediction sequence through the visualization interface;

[0078] S503. When the predicted risk level is the third level and the fourth level, give a prompt through intelligent alarm, and determine the specific time and sensor number corresponding to the prediction sequence.

[0079] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0080] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A generative coal mine production environment risk screening and decision-making method, characterized in that: The method comprises the following steps: S1. Determine the sensor number: number all the working faces and access faces in multiple coal mining areas one by one, arrange multiple gas sensors near each working face and access face, number the gas sensors, count the distance between each gas sensor and the working face and access face, and transmit the distance to the database; S2. Divide the time series: the collected data are stored in the gas data set in the form of time series. After setting multiple time windows, the time series segmentation method is used to divide the data in the gas data set to obtain multiple time slice information. The cloud transformation algorithm is used to convert these time slice information into corresponding granular time series and transmit them to the storage device. S3, Analyze spatial sequence: determine the working face with the shortest distance to each sensor, analyze the sensor number corresponding to each face, set multiple spatial windows, divide the data set according to sensors, working faces, mining areas and mining areas, obtain multiple spatial slice information, convert these spatial slice information into corresponding granular spatial sequences through cloud transformation algorithm, and transmit them to storage devices; S4. Calculate the risk level: obtain the data in the gas data set, perform a distribution test on the data, calculate the skewness value corresponding to the distribution function using the skewness analysis algorithm, obtain multiple cloud parameters based on the Gaussian distribution function analysis, fit these cloud parameters, and generate a gas concentration concept distribution map. The concentration is divided into four concepts based on the gas content. The gas concentration concept distribution map is used to determine the relevant parameters and risk level corresponding to each concentration concept. After extracting the particle time series and particle space series in the storage device, analyze the risk level corresponding to each time particle layer and space particle layer according to the relevant parameters corresponding to each concentration concept. S5. Predict risk data: Obtain relevant parameters corresponding to all particle time series, use these parameters to analyze and obtain the development matrix, add the development matrix to the variable cloud prediction formula, obtain the corresponding prediction sequence, determine the corresponding risk level, and output the risk level and prediction sequence through a visual interface.

2. A generative coal mine production environment risk screening and decision-making method according to claim 1, characterized in that: The step S2 comprises the following steps: S201, after acquiring the data collected by each gas sensor, the data are sequentially stored in a gas data set in the form of a time series; S202. Since the interval time for sensor data recording is one minute, after setting multiple time windows, the time series segmentation method is used to divide the data in the gas data set according to minutes, hours, days, months, seasons and years, and obtain minute time slice information, hour time slice information, day time slice information, month time slice information, season time slice information and year time slice information.

3. A generative coal mine production environment risk screening and decision-making method according to claim 2, characterized in that: The step S2 further comprises the following steps: S203, converting minute time slice information, hour time slice information, day time slice information, month time slice information, season time slice information and year time slice information into corresponding granular time series respectively through cloud transformation algorithm, and transmitting them to storage device.

4. A generative coal mine production environment risk screening and decision-making method according to claim 1, characterized in that: The step S3 specifically comprises the following steps: S301, after obtaining the data collected by each gas sensor and the distance from the corresponding sensor to each working surface, calculate the working surface with the shortest distance for each sensor, associate the corresponding surface number with the sensor number, and thus obtain the sensor number corresponding to each surface; S302. After setting multiple spatial windows, the data set is divided according to sensors, working faces, mining areas and mining areas to obtain sensor spatial slice information, working face spatial slice information, mining area spatial slice information and mining area spatial slice information.

5. A generative coal mine production environment risk screening and decision-making method according to claim 4, characterized in that: The step S3 specifically further comprises the following steps: S303, respectively converting the sensor space slice information, the working face space slice information, the mining area space slice information and the mining area space slice information into corresponding particle space sequences through a cloud transformation algorithm, and transmitting them to a storage device.

6. A generative coal mine production environment risk screening and decision-making method according to claim 1, characterized in that: The step S4 specifically comprises the following steps: S401, obtaining data from a gas data set, performing a distribution test on the data, selecting the peak number of the distribution function corresponding to the data as the initial frequency, and using a skewness analysis algorithm to calculate the skewness value corresponding to the distribution function. If the skewness value is greater than a threshold, logarithmic processing is performed on the current distribution function, otherwise logarithmic processing is not performed; S402. Convert the current distribution function into multiple Gaussian distribution functions, and calculate multiple cloud parameters based on the Gaussian distribution function and standard deviation related parameters. If the mixing degree of the cloud parameters is greater than the upper limit of the mixing degree, adjust the mixing degree. Otherwise, fit these cloud parameters to obtain a conceptual distribution diagram of the gas concentration.

7. A generative coal mine production environment risk screening and decision-making method according to claim 6, characterized in that: The step S4 specifically further comprises the following steps: S403. According to the gas content, the concentration is divided into four concepts, namely low concentration, medium concentration, medium-high concentration and high concentration. After determining the relevant parameters corresponding to each concentration concept according to the gas concentration concept distribution map, it is determined that the low concentration concept corresponds to the first risk level, the medium concentration concept corresponds to the second risk level, the medium-high concentration concept corresponds to the third risk level, and the high concentration concept corresponds to the fourth risk level; S404, after extracting the particle time series in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all particle time series to determine the risk levels corresponding to the minute particle layer, hour particle layer, daily particle layer, monthly particle layer, seasonal particle layer and annual particle layer; S405. After extracting the particle space sequence in the storage device, compare the relevant parameters corresponding to each concentration concept with the relevant parameters of all particle space sequences to determine the risk levels corresponding to the conveyor particle layer, the working face particle layer, the mining area particle layer and the mining area particle layer.

8. A generative coal mine production environment risk screening and decision-making method according to claim 1, characterized in that: The step S5 specifically comprises the following steps: S501, obtaining relevant parameters corresponding to all particle time series, wherein the relevant parameters include expectation, entropy and super entropy corresponding to the series, using these parameters to analyze and obtain a development matrix, and after adding the development matrix to the variable cloud prediction formula, obtaining the corresponding prediction sequence; S502, comparing the relevant parameters corresponding to each concentration concept with the relevant parameters corresponding to the prediction sequence, and after determining the corresponding risk level, outputting the risk level and the prediction sequence through a visual interface; S503: When the predicted risk level is the third level or the fourth level, a prompt is given through an intelligent alarm, and a specific time and sensor number corresponding to the predicted sequence are determined.

Images