An efficient method for collecting and monitoring coal mine environmental data

Abstract

The present invention relates to the technical field of data acquisition, and specifically relates to an efficient acquisition and monitoring method for coal mine environmental data; segment feature values are obtained according to the difference characteristics of the data change trends before and after the data points in the gas concentration time series; different concentration feature time periods are obtained according to the segment feature values; discrete data points are obtained according to the influence degree of the data points in the concentration feature time periods on the overall data change trend; target data points are obtained according to the discrete features of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval features of adjacent discrete data points within the preset neighborhood range. The present invention obtains an acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration feature time periods with target data points, the quantity characteristics of the target data points in the concentration feature time periods, and the data fluctuation characteristics; adaptive data acquisition and analysis are performed on the concentration feature time periods according to the acquisition frequency adjustment coefficient, improving the accuracy and efficiency of data analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of data acquisition, and particularly relates to a method for efficiently acquiring and monitoring coal mine environmental data. Background Art

[0002] During the process of coal mine resource extraction, harmful gas such as methane is likely to appear. When the methane concentration is too high, safety accidents such as explosions may occur. With the modernization of the extraction process and the application of big data, the methane concentration at multiple positions in the mine can be monitored during the extraction process for early warning to improve the extraction safety; after the extraction is completed, the methane concentration data at all monitoring points can be analyzed and evaluated to analyze the change law of methane concentration and the working condition of the ventilation system during the extraction process, so as to optimize and improve the subsequent extraction process.

[0003] Due to the complex structure underground in the mine, the number of monitoring points for monitoring methane concentration is large, the amount of data collected at each monitoring point is large, and the methane concentration data is relatively stable in most time periods; therefore, a large amount of redundant data will appear when analyzing the monitoring data of methane concentration, which affects the efficiency and accuracy of data analysis. Summary of the Invention

[0004] In order to solve the technical problem that the redundant data generated during the above monitoring process affects the efficiency and accuracy of data analysis, the purpose of the present invention is to provide a method for efficiently acquiring and monitoring coal mine environmental data, and the specific technical solution adopted is as follows:

[0005] Obtain the time series of methane concentration at preset positions in the mine within a preset historical period;

[0006] Obtain segment feature values according to the difference characteristics of the data change trends before and after the data points in the methane concentration time series; segment the methane concentration time series according to the segment feature values of the data points to obtain different concentration feature periods; obtain the data deviation degree according to the influence degree of the data points in the concentration feature period on the overall data change trend; obtain discrete data points according to the data deviation degree of the data points.

[0007] Obtain the credibility according to the discrete characteristics of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval characteristics of adjacent discrete data points within the preset neighborhood range; obtain target data points according to the credibility of the discrete data points; obtain the acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration feature periods with target data points, the quantity characteristics of the target data points in the concentration feature periods, and the data fluctuation characteristics.

[0008] Perform adaptive data acquisition on the concentration feature periods according to the acquisition frequency adjustment coefficient to obtain a concentration monitoring analysis data set.

[0009] Further, the step of obtaining the segmented feature value according to the difference feature of the data change trend before and after the data point in the gas concentration time series includes:

[0010] Calculate the dynamic time warping distance of the gas concentration segments of the same preset length before and after the data point and normalize it to obtain the first change difference degree; calculate the slope of the fitting straight line of the gas concentration segment of the preset length before the data point to obtain the first change feature value; calculate the slope of the fitting straight line of the gas concentration segment of the preset length after the data point to obtain the second change feature value; calculate the absolute value of the difference between the first change feature value and the second change feature value and normalize it to obtain the second change difference degree; calculate the average value of the first change difference degree and the second change difference degree to obtain the segmented feature value of the data point.

[0011] Further, the step of segmenting the gas concentration time series according to the segmented feature value of the data point to obtain different concentration feature periods includes:

[0012] Take the data point whose segmented feature value exceeds the preset segmentation threshold as the segmentation point, and segment the gas concentration time series at the segmentation point to obtain different concentration feature periods.

[0013] Further, the step of obtaining the data deviation degree according to the influence degree of the data point in the concentration feature period on the overall data change trend includes:

[0014] Calculate the slope of the fitting straight line including all data points in the concentration feature period to obtain the first slope value; calculate the slope of the fitting straight line when no data point is included in the concentration feature period to obtain the second slope value; calculate the absolute value of the difference between the first slope value and the second slope value and normalize it to obtain the data deviation degree of any data point.

[0015] Further, the step of obtaining the discrete data points according to the data deviation degree of the data point includes:

[0016] Take the data point whose data deviation degree exceeds the preset deviation threshold as the discrete data point.

[0017] Further, the step of obtaining the credibility according to the discrete feature of the difference between adjacent data points within the preset neighborhood range of the discrete data point and the interval feature of adjacent discrete data points within the preset neighborhood range includes:

[0018]

[0019] Wherein, R represents the credibility of the discrete data point, a represents a preset extremely small positive number, and N represents the number of interval segments formed by adjacent discrete data points within the preset neighborhood range. represents the number of data points in the nth interval segment, represents the average value of the number of data points in all interval segments, represents the density eigenvalue; Q represents the number of difference values of the difference sequence corresponding to the data within the preset neighborhood range, represents the average value from the first difference value to the qth difference value, represents from the first difference value to the average value of the difference values, represents the degree of difference fluctuation; represents the difference fluctuation eigenvalue.

[0020] Further, the step of obtaining the target data points according to the credibility of the discrete data points includes:

[0021] Taking the discrete data points whose credibility exceeds the preset credibility threshold as the target data points.

[0022] Further, the step of obtaining the acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration characteristic time periods with target data points, the quantity characteristics of the target data points in the concentration characteristic time periods, and the data fluctuation characteristics includes:

[0023] Calculating the ratio of the number of concentration characteristic time periods with target data points to the total number of concentration characteristic time periods to obtain the time period occupancy ratio; calculating the product of the proportion of the number of target data points in the concentration characteristic time periods and the time period occupancy ratio and performing a negative correlation mapping to obtain the redundancy eigenvalue; calculating the reciprocal of the variance of the difference sequence of the concentration characteristic time periods to obtain the stability eigenvalue; calculating the average value of the redundancy eigenvalue and the stability eigenvalue to obtain the acquisition frequency adjustment coefficient of the concentration characteristic time periods.

[0024] Further, the step of adaptively collecting data for the concentration characteristic time periods according to the acquisition frequency adjustment coefficient to obtain the concentration monitoring analysis data set includes:

[0025] Calculating the product of a preset constant and the acquisition frequency adjustment coefficient and rounding up to obtain the deletion quantity; starting from the first position in the concentration characteristic time periods for collection, collecting the next data point every other deletion quantity of data points until the traversal is completed to obtain the concentration monitoring analysis data set.

[0026] The present invention has the following beneficial effects:

[0027] In the present invention, obtaining the segmented eigenvalue can segment different concentration change trend sequence segments, so as to adaptively determine the data acquisition frequency of different concentration characteristic time periods, and initially improve the accuracy of data acquisition frequency calculation. Since the more fluctuating concentration data in the concentration characteristic time period can represent more mining characteristic information, obtaining the data deviation degree can determine the influence of different data points on the overall change trend of the concentration characteristic time period, so as to determine the amount of mining characteristic information reflected by different data points; obtaining discrete data points can determine the data points in the concentration characteristic time period that represent more mining characteristic information, and improve the accuracy of the adaptive data acquisition frequency. Since the reasons for the appearance of discrete data points may be caused by multiple factors, obtaining the credibility can screen the target data points caused by mining reasons among the discrete data points, and further improve the accuracy of the data acquisition frequency. Obtaining the acquisition frequency adjustment coefficient can accurately reflect the amount of data used for data analysis in the concentration characteristic time period. Finally, obtaining the concentration monitoring and analysis data set according to the acquisition frequency adjustment coefficient can improve the accuracy and efficiency of concentration data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0029] Figure 1 It is a flowchart of a method for efficiently collecting and monitoring coal mine environment data provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the drawings and preferred embodiments, describe in detail the specific implementation manner, structure, characteristics and effects of a method for efficiently collecting and monitoring coal mine environment data proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0032] The following will specifically describe the specific solution of a method for efficiently collecting and monitoring coal mine environment data provided by the present invention with reference to the drawings.

[0033] Please refer to Figure 1, which shows a flowchart of an efficient acquisition and monitoring method for coal mine environmental data provided by an embodiment of the present invention. The method includes the following steps:

[0034] Step S1, obtain the time series of gas concentration at a preset position in the mine within a preset historical period.

[0035] First, obtain the time series of gas concentration at a preset position in the mine within a preset historical period. The acquisition frequency, preset historical period, and preset position of the gas concentration time series are constructed by the implementer according to the implementation scenario. In the embodiment of the present invention, the acquisition frequency of the gas concentration time series is 1 time per second, and the calculation steps for other preset positions and preset historical periods are the same.

[0036] Step S2, obtain the segmented feature values according to the difference characteristics of the data change trends before and after the data points in the gas concentration time series; segment the gas concentration time series according to the segmented feature values of the data points to obtain different concentration feature periods; obtain the data deviation degree according to the influence degree of the data points in the concentration feature period on the overall data change trend; obtain the discrete data points according to the data deviation degree of the data points.

[0037] When the gas concentration in a certain period of the gas concentration time series is relatively stable, the gas concentration data in this period reflects similar characteristic states during the mining process, such as similar coal seam characteristics and similar ventilation states. Therefore, such stable data can be considered redundant data in the data analysis process. If there is a large amount of redundant data, it will affect the accuracy and efficiency of data analysis. To improve the accuracy and efficiency of gas concentration data analysis, it can be adaptively deleted according to the data characteristics of the gas concentration time series to reduce the amount of redundant data. During the mining process, the differences in processes, the complexity of gas distribution, and the changes in ventilation states will all cause the gas concentration at the preset position to show different change trends; the degree of data deletion in different change trends is different, so it is first necessary to split the data sequences with different change trends and obtain the segmented feature values according to the difference characteristics of the data change trends before and after the data points in the gas concentration time series.

[0038] Preferably, in the embodiments of the present invention, the steps of obtaining the segmented feature values include: calculating and normalizing the dynamic time warping distance of the gas concentration segments of the same preset length before and after the data point to obtain the first change difference degree; in the embodiments of the present invention, the preset length is the length constructed by 60 data points, and the implementer can determine it according to the implementation scenario; it should be noted that the dynamic time warping distance is obtained through the existing dynamic time warping algorithm. The smaller the dynamic time warping distance is when two sequences are more similar, so the larger the first change difference degree is, which means the greater the difference between the gas concentration segments before and after the data point, and the more likely the data point is the segmentation point of different change trends of the gas concentration. Calculating the slope of the fitting line of the gas concentration segment of the preset length before the data point to obtain the first change feature value; calculating the slope of the fitting line of the gas concentration segment of the preset length after the data point to obtain the second change feature value; calculating and normalizing the absolute value of the difference between the first change feature value and the second change feature value to obtain the second change difference degree; when the difference between the change trends of the gas concentration before and after the data point is greater, the difference between the first change feature value and the second change feature value is greater. When the second change difference degree is greater, it means the more likely the data point is the segmentation point of different gas concentration change trends. Calculating the average value of the first change difference degree and the second change difference degree to obtain the segmented feature value of the data point; when the segmented feature value is greater, it means that the gas concentration change trends before and after the data point are more likely to be different.

[0039] Further, the gas concentration time series can be segmented according to the segmented feature value of the data point to obtain different concentration feature periods; preferably, in the embodiments of the present invention, the steps of obtaining the concentration feature periods include: taking the data points with the segmented feature values exceeding the preset segmentation threshold as segmentation points, and splitting the gas concentration time series at the segmentation points to obtain different concentration feature periods; there are differences in the change trends of adjacent concentration feature periods. In the embodiments of the present invention, the preset segmentation threshold is 0.7, and the implementer can determine it according to the implementation scenario.

[0040] Gas exists in coal seams. When mining reaches the gas area, it will cause the gas concentration to rise. With the operation of the ventilation system, the gas concentration will slowly decrease. Such obvious fluctuating characteristics of gas concentration changes play an important role in data analysis, and can reflect the gas dissipation trend and ventilation efficiency. At the same time, the distribution of coal seams and the mining speed will also cause changes in gas concentration. Therefore, when analyzing gas concentration data, when the concentration data changes significantly, a larger amount of data is required to improve the accuracy of mining analysis. During the mining interval, the gas concentration will remain at a low level for a long time. Such stable gas concentration data has little effect on data analysis. If a large amount of redundant data appears, it will lead to a decrease in data analysis efficiency. Therefore, in data analysis, more data can be retained for concentration characteristic periods with more fluctuations that can reflect gas concentration changes, while less data can be retained for concentration characteristic periods with relatively stable concentrations, so as to improve the efficiency and accuracy of gas concentration data analysis. If any data point in the concentration characteristic period affects the data change trend, it means that the arbitrary data point is relatively discrete and the fluctuation is obvious. Therefore, the data deviation degree is obtained according to the influence degree of the data point in the concentration characteristic period on the overall data change trend.

[0041] Preferably, in the embodiment of the present invention, the steps of obtaining the data deviation degree include: calculating the slope of the fitting line including all data points in the concentration characteristic period to obtain the first slope value; calculating the slope of the fitting line when not including any data point in the concentration characteristic period to obtain the second slope value; if the distribution of the arbitrary data point in the concentration characteristic period is more discrete, it means that the concentration characteristic at this moment is more fluctuating, and the difference between the first slope value and the second slope value is greater. Calculate the absolute value of the difference between the first slope value and the second slope value and normalize it to obtain the data deviation degree of the arbitrary data point; when the data deviation degree of the arbitrary data point is greater, it means that the arbitrary data point is more discrete and can better represent the mining state. Furthermore, discrete data points can be obtained according to the data deviation degree. Preferably, in the embodiment of the present invention, the steps of obtaining discrete data points include: taking the data points with a data deviation degree exceeding the preset deviation threshold as discrete data points. In the embodiment of the present invention, the preset deviation threshold is 0.5, and the implementer can determine it according to the implementation scenario.

[0042] Step S3, obtain the credibility according to the discrete characteristics of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval characteristics of adjacent discrete data points within the preset neighborhood range; obtain the target data points according to the credibility of the discrete data points; obtain the acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration characteristic periods with target data points, the quantity characteristics of the target data points in the concentration characteristic periods, and the data fluctuation characteristics.

[0043] Since the deviation data points may be caused by noise data or sensor acquisition errors, in order to improve the accuracy of data acquisition frequency adjustment, it is necessary to further screen the deviation data points to determine the number of data points that truly reflect the concentration fluctuation characteristics in the concentration characteristic period. When the deviation data points represent the concentration data fluctuations, there will be multiple other deviation data points in the adjacent periods of the deviation data points, and the data distribution in the adjacent periods of the deviation data points is relatively discrete; while for the deviation data points caused by noise or acquisition errors, they are relatively random, there are fewer other deviation data points in their adjacent periods, and the data distribution in the adjacent periods is relatively stable. Therefore, the credibility can be obtained according to the discrete characteristics of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval characteristics of adjacent discrete data points within the preset neighborhood range; preferably, in the embodiments of the present invention, the steps and formulas for obtaining the credibility include:

[0044]

[0045] In the formula, R represents the credibility of the discrete data point, a represents a preset extremely small positive number, which is 0.1 in the embodiments of the present invention, and the purpose is to avoid the denominator being 0. N represents the number of interval segments formed by adjacent discrete data points within the preset neighborhood range. In the embodiments of the present invention, the preset neighborhood range is the range formed by the 100 other data points closest to the discrete data point, and the implementer can determine it according to the implementation scenario; represents the number of data points in the nth interval segment. When the number of data points in the interval segment is more, it means that the distribution of other discrete data points in the adjacent period of the discrete data point is sparser, and the possibility that the discrete data point reflects the gas concentration fluctuation characteristics is smaller; represents the average value of the number of data points in all interval segments, represents the density eigenvalue; when the density eigenvalue is larger, it means that the distribution of other discrete data points in the adjacent period of the discrete data point is denser, and it is more likely to represent the true gas concentration fluctuation characteristics. Q represents the number of difference values of the difference sequence corresponding to the data within the preset neighborhood range, represents the average value from the first difference value to the qth difference value, represents the average value from the first difference value to the th difference value, represents the differential fluctuation degree; if the concentration fluctuation in the adjacent period of the discrete data point is more obvious, the distribution of the difference values is more discrete, and the differential fluctuation degree is larger; if the concentration change in the adjacent period of the discrete data point is smaller, the differential fluctuation degree is smaller, and the discrete data point is more likely to be caused by noise and acquisition errors. represents the differential fluctuation eigenvalue. The larger the differential fluctuation eigenvalue, the more likely the discrete data point is caused by the gas concentration fluctuation.

[0046] Further, target data points can be obtained based on the credibility of discrete data points, specifically including: regarding discrete data points with credibility exceeding a preset credibility threshold as target data points. In the embodiments of the present invention, the preset credibility threshold is 0.6, and the implementer can determine it according to the implementation scenario by himself. The target data points can characterize the real concentration fluctuation characteristics; when there are more target data points in the concentration characteristic period and the difference values in the difference sequence of the concentration characteristic period are more discrete, it means that the gas concentration fluctuation in this period is more obvious, and the more mining characteristic information is characterized. When analyzing the gas concentration data after mining, the more data is required in this concentration characteristic period, so as to improve the accuracy of concentration data analysis. At the same time, the more concentration characteristic periods with target data points in this gas concentration time series, it means that the preset historical period where this gas concentration time series is located is more likely to be in the peak mining period, and the change of gas concentration is more frequent. Therefore, more data needs to be provided in this concentration characteristic period to ensure the accuracy of data analysis. Therefore, the acquisition frequency adjustment coefficient is obtained according to the quantity characteristics of the concentration characteristic periods with target data points, the quantity characteristics of the target data points in the concentration characteristic periods, and the data fluctuation characteristics.

[0047] Preferably, in the embodiments of the present invention, the steps of obtaining the acquisition frequency adjustment coefficient include: calculating the ratio of the number of concentration characteristic periods with target data points to the total number of concentration characteristic periods to obtain the period occupancy ratio; when the period occupancy ratio is larger, it means that the preset historical period is more likely to be in the peak mining period, and more data is required in data analysis. Calculate the product of the ratio of the number of target data points in the concentration characteristic period to the period occupancy ratio and perform a negative correlation mapping to obtain the redundancy characteristic value; when the ratio of the number of target data points is larger, it means that there are more target data points in this concentration characteristic period, and more data should be provided in data analysis. Calculate the reciprocal of the variance of the difference sequence of the concentration characteristic period to obtain the stability characteristic value; when the stability characteristic value is smaller, it means that the concentration change in this period is stronger and can better characterize different situations in the mining process. Calculate the average value of the redundancy characteristic value and the stability characteristic value to obtain the acquisition frequency adjustment coefficient of the concentration characteristic period; when the acquisition frequency adjustment coefficient is larger, it means that the concentration data in this concentration characteristic period is more stable and there is more redundant data, and less data can be provided in data analysis to ensure the analysis efficiency; when the acquisition frequency adjustment coefficient is smaller, it means that the concentration data in this concentration characteristic period is more fluctuating, and more data characterizing different situations in mining is required, and more data needs to be provided in data analysis to ensure the analysis accuracy.

[0048] Step S4, adaptively collect data for the concentration characteristic periods according to the acquisition frequency adjustment coefficient to obtain the concentration monitoring and analysis data set.

[0049] After obtaining the acquisition frequency adjustment coefficient for the concentration characteristic period, the concentration characteristic period can be adaptively data-acquired according to the acquisition frequency adjustment coefficient to obtain a concentration monitoring and analysis data set. Preferably, in the embodiment of the present invention, the steps of obtaining the concentration monitoring and analysis data set include: calculating the product of a preset constant and the acquisition frequency adjustment coefficient and rounding up to obtain the deletion quantity. In the embodiment of the present invention, the preset constant is 10, and the implementer can determine it according to the implementation scenario. Starting from the first position in the concentration characteristic period, the next data point is acquired after every deletion quantity of data points. The larger the deletion quantity, the more data points are spaced apart, until the traversal is completed to obtain the concentration monitoring and analysis data set. The data volume in the concentration monitoring and analysis data set that can characterize the mining characteristics is larger, while the redundant data volume of the stable concentration data is smaller. Furthermore, data analysis is performed on the concentration monitoring and analysis data sets of all concentration characteristic periods during the mining process, improving the efficiency and accuracy of concentration data analysis.

[0050] In summary, the embodiment of the present invention provides an efficient acquisition and monitoring method for coal mine environmental data. The segmented characteristic values are obtained according to the difference characteristics of the data changes before and after the data points in the gas concentration time series. Different concentration characteristic periods are obtained according to the segmented characteristic values. Discrete data points are obtained according to the influence degree of the data points in the concentration characteristic period on the overall data change trend. The target data points are obtained according to the discrete characteristics of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval characteristics of adjacent discrete data points within the preset neighborhood range. The present invention obtains the acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration characteristic periods with target data points, the quantity characteristics of the target data points in the concentration characteristic periods, and the data fluctuation characteristics. Adaptive data acquisition and analysis are performed on the concentration characteristic periods according to the acquisition frequency adjustment coefficient, improving the accuracy and efficiency of data analysis.

[0051] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for efficiently collecting and monitoring coal mine environmental data, characterized in that: The method comprises the following steps: Obtaining the time series of gas concentration at a preset location in the mine within a preset historical period; Obtaining segmented characteristic values ​​according to the difference characteristics of the data change trends before and after the data points in the gas concentration time series; segmenting the gas concentration time series according to the segmented characteristic values ​​of the data points to obtain different concentration characteristic time periods; obtaining data deviation according to the influence of the data points in the concentration characteristic time period on the overall data change trend; obtaining discrete data points according to the data deviation of the data points; Obtain credibility based on the discrete features of the differences between adjacent data points within a preset neighborhood of the discrete data points and the interval features of adjacent discrete data points within the preset neighborhood; obtain target data points based on the credibility of the discrete data points; obtain acquisition frequency adjustment coefficients based on the quantity features of the concentration feature time period in which the target data points exist, the quantity features of the target data points in the concentration feature time period, and data fluctuation features; Adaptively collect data for the concentration characteristic period according to the collection frequency adjustment coefficient to obtain a concentration monitoring and analysis data set; The step of obtaining the credibility according to the discrete features of the differences between adjacent data points within the preset neighborhood range of the discrete data points and the interval features of adjacent discrete data points within the preset neighborhood range comprises: , where R represents the credibility of the discrete data point, a represents a preset minimum positive number, and N represents the number of intervals formed by adjacent discrete data points within the preset neighborhood. represents the number of data points in the nth interval, represents the average number of data points in all intervals, represents the density eigenvalue; Q represents the number of differential values ​​of the differential sequence corresponding to the data within the preset neighborhood range, represents the average value from the first difference value to the qth difference value, represents the average value from the first difference value to the q+1th difference value, Indicates the degree of differential fluctuation; represents the differential fluctuation eigenvalue; The step of obtaining the target data point according to the credibility of the discrete data point comprises: The discrete data points whose credibility exceeds a preset credibility threshold are used as the target data points.

2. A method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of obtaining segmented characteristic values ​​according to the difference characteristics of the data change trends before and after the data points in the gas concentration time series includes: Calculate the dynamic time warping distance of the gas concentration segment of the same preset length before and after the data point and normalize it to obtain a first change difference; calculate the slope of the fitting straight line of the gas concentration segment of the preset length before the data point to obtain a first change characteristic value; calculate the slope of the fitting straight line of the gas concentration segment of the preset length after the data point to obtain a second change characteristic value; calculate the absolute value of the difference between the first change characteristic value and the second change characteristic value and normalize it to obtain a second change difference; calculate the average value of the first change difference and the second change difference to obtain the segmented characteristic value of the data point.

3. The method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of segmenting the gas concentration time series according to the segmented characteristic values ​​of the data points to obtain different concentration characteristic time periods comprises: The data points where the segmentation characteristic values ​​exceed the preset segmentation threshold are used as segmentation points, and the gas concentration time series is segmented at the segmentation points to obtain different concentration characteristic time periods.

4. The method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of obtaining the data deviation according to the influence of the data points in the concentration characteristic period on the overall data change trend comprises: The slope of the fitted straight line containing all data points in the concentration characteristic time period is calculated to obtain a first slope value; the slope of the fitted straight line when the concentration characteristic time period does not contain any data point is calculated to obtain a second slope value; the absolute value of the difference between the first slope value and the second slope value is calculated and normalized to obtain the data deviation of the arbitrary data point.

5. The method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of obtaining discrete data points according to the data deviation of the data points comprises: The data points whose data deviation exceeds a preset deviation threshold are regarded as the discrete data points.

6. A method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of obtaining the acquisition frequency adjustment coefficient according to the quantity characteristics of the concentration characteristic time period in which the target data points exist, the quantity characteristics of the target data points in the concentration characteristic time period, and the data fluctuation characteristics comprises: Calculate the ratio of the number of concentration feature time periods in which target data points exist to the total number of concentration feature time periods to obtain a time period proportion value; calculate the product of the number proportion of target data points in the concentration feature time period and the time period proportion value and negatively correlate them to obtain a redundant eigenvalue; calculate the inverse of the variance of the differential sequence of the concentration feature time period to obtain a stable eigenvalue; calculate the average of the redundant eigenvalue and the stable eigenvalue to obtain the acquisition frequency adjustment coefficient of the concentration feature time period.

7. The method for efficiently collecting and monitoring coal mine environmental data according to claim 1, characterized in that: The step of adaptively collecting data for the concentration characteristic period according to the collection frequency adjustment coefficient to obtain a concentration monitoring and analysis data set comprises: Calculate the product of a preset constant and the acquisition frequency adjustment coefficient and round it up to obtain the number of deletions; start collecting from the first position in the concentration characteristic period, collect the next data point after each interval of deletion of the number of data points, until the traversal is completed, and obtain the concentration monitoring and analysis data set.

Images