Abnormality judgment method and device for gas extraction monitoring data, medium and terminal

Abstract

The invention discloses an anomaly judgment method and device for gas extraction monitoring data, a medium and a terminal, relates to the technical field of coal mine safety production, and mainly aims to solve the problems of lag and increased misjudgment risk caused by high dependence on manual judgment in the prior art. Comprising the following steps: firstly, carrying out false monitoring data screening processing and monitoring failure judgment processing on current monitoring data to determine the validity of the monitoring data, and further, carrying out long-term trend anomaly judgment on the current monitoring data according to a long-term data trend change characteristic value to obtain a first anomaly judgment result; and performing data abrupt change abnormity judgment on the current monitoring data according to the data fluctuation characteristic value, the short-term data trend change characteristic and the mean value characteristic value to obtain a second abnormity judgment result, thereby generating a final abnormity judgment result.

Description

Technical Field

[0001] The present application relates to the technical field of coal mine production safety, and in particular to a method, device, medium, and terminal for distinguishing abnormalities in gas extraction monitoring data. Background Art

[0002] Gas extraction technology plays a crucial role in coal mine safety and is one of the core means of preventing coal and gas outburst accidents. Gas extraction extracts gas from coal seams, reducing its content and pressure, thereby effectively reducing the occurrence of coal and gas outburst accidents. With increasing mining depth and mining intensity, the gas content and pressure in coal seams continue to increase, increasing the risk of coal and gas outburst accidents. In the context of intelligent coal mine development, intelligent monitoring and identification of gas extraction have become key components in improving coal mine safety and promoting the transformation and upgrading of the coal industry.

[0003] Currently, the identification of anomalies in gas drainage monitoring data primarily relies on manual monitoring and basic data identification software. Simple charts display changing trends in gas drainage monitoring data, and staff regularly review these charts, combining their expertise and experience to determine whether anomalies exist. However, this heavy reliance on manual identification can easily lead to delays and misjudgments, increasing the risk of coal and gas outburst accidents. Summary of the Invention

[0004] In view of this, the present application provides a method and device, medium, and terminal for identifying abnormalities in gas extraction monitoring data, the main purpose of which is to improve the existing problem of lag and increased risk of misjudgment due to the high reliance on manual judgment.

[0005] According to one aspect of the present application, a method for identifying abnormalities in gas drainage monitoring data is provided, comprising:

[0006] Acquire a first monitoring data group of the target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods;

[0007] Performing pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and performing monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result for the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a slight change characteristic value and a data fluctuation characteristic value, the slight change characteristic value is used to represent the maximum duration during which the difference between the maximum value and the minimum value in the continuous monitoring data within a statistical period is less than a preset change amplitude threshold, and the data fluctuation characteristic value is used to represent the average of the fluctuation amplitudes of multiple monitoring periods;

[0008] If the monitoring failure judgment result is valid, a long-term trend abnormality judgment process is performed on the second monitoring data group to obtain a first abnormality judgment result, and a data mutation abnormality judgment process is performed on the second monitoring data group to obtain a second abnormality judgment result, wherein the long-term trend abnormality judgment is performed based on the long-term data trend change characteristic value, and the data mutation abnormality judgment process is performed based on the data fluctuation characteristic value, the short-term data trend change characteristic and the mean characteristic value. The long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group;

[0009] The first abnormality determination result and the second abnormality determination result are combined to generate an abnormality determination result of the gas extraction monitoring data for the target sensor.

[0010] Preferably, the pseudo monitoring data identification processing is performed on each monitoring data to filter out the pseudo monitoring data in the first monitoring data group to obtain the second monitoring data group, including:

[0011] If the first monitoring data group is discrete data, identifying abnormal monitoring data in the first monitoring data group based on the Laida criterion and filtering out the abnormal monitoring data to obtain a second monitoring data group;

[0012] If the first monitoring data group is continuous data, the first abnormal interval in the first monitoring data group is identified based on the box-and-line method, and the first abnormal interval is verified based on the second abnormal interval of the adjacent sensor, so as to screen out the first abnormal interval according to the verification result and obtain the second monitoring data group.

[0013] Preferably, performing monitoring failure determination processing on the second monitoring data group to obtain a monitoring failure determination result of the second monitoring data group includes:

[0014] Obtaining a preset small change duration threshold corresponding to the target sensor;

[0015] Counting a small change characteristic value of the second monitoring data group, and if the small change characteristic value is less than the preset small change duration threshold, determining that the preliminary monitoring failure determination result of the second monitoring data group is valid;

[0016] If the preliminary monitoring failure determination result is valid, obtaining a preset minimum fluctuation amplitude threshold corresponding to the target sensor;

[0017] Calculating a data fluctuation characteristic value of the second monitoring data group, and marking the current statistical period if the data fluctuation characteristic value is less than the preset minimum fluctuation amplitude threshold;

[0018] If the number of the continuous marked statistical periods does not exceed the preset statistical period number threshold, it is determined that the final monitoring failure determination result of the second monitoring data set is valid, and a monitoring failure determination result of the second monitoring data set is obtained.

[0019] Preferably, the long-term trend anomaly determination processing on the second monitoring data set is performed to obtain a first anomaly determination result, including:

[0020] Based on the least square method, curve fitting processing is performed on the second monitoring data set to obtain a curve equation corresponding to the second monitoring data set, and a slope of the curve equation is determined as a long-term data trend change characteristic value of the second monitoring data set.

[0021] If the long-term data trend change characteristic value is greater than a preset long-term trend change threshold, it is determined that the first anomaly determination result is abnormal.

[0022] Preferably, the data mutation anomaly determination processing on the second monitoring data set is performed to obtain a second anomaly determination result, including:

[0023] The data fluctuation characteristic value, the short-term data trend change characteristic, and the mean value characteristic value of the second monitoring data set are calculated respectively.

[0024] A preset maximum fluctuation amplitude threshold corresponding to the target sensor and a plurality of long-term mean value characteristic values corresponding to the target sensor are obtained, wherein the time lengths of the respective long-term mean value characteristic values are different.

[0025] The deviation values between the mean value characteristic value and the respective long-term mean value characteristic values are calculated respectively, and the maximum deviation value is screened out.

[0026] If the data fluctuation characteristic value is greater than the preset maximum fluctuation amplitude threshold, it is determined that the first data mutation anomaly determination result of the second monitoring data is abnormal.

[0027] If the short-term data trend change characteristic is an upward trend, it is determined that the second data mutation anomaly determination result of the second monitoring data is abnormal.

[0028] If the maximum deviation value is greater than a preset deviation value threshold, it is determined that the third data mutation anomaly determination result of the second monitoring data is abnormal.

[0029] If any one of the first data mutation anomaly determination result, the second data mutation anomaly determination result, and the third data mutation anomaly determination result is abnormal, it is determined that the second anomaly determination result is abnormal.

[0030] Preferably, before obtaining the first monitoring data set of the target sensor in the current statistical period, the method further includes:

[0031] Collect monitoring data from various sensors based on driver modules adapted to different types of sensors;

[0032] Generate a monitoring data element basic attribute table corresponding to each monitoring data according to the preset data element standardization rule, and map and store the monitoring data element basic attribute table with the corresponding monitoring data;

[0033] The obtaining of a first monitoring data set of the target sensor in the current statistical period includes:

[0034] According to the identification of the target sensor;

[0035] A matching monitoring data element basic attribute table is searched according to the identity identifier to obtain monitoring data corresponding to the monitoring data element basic attribute table.

[0036] Preferably, the categories of sensors include: online metering equipment, gas extraction pump station equipment and drilling trajectory measurement equipment.

[0037] According to another aspect of the present application, a device for identifying abnormalities in gas drainage monitoring data is provided, comprising:

[0038] A data acquisition module, configured to acquire a first monitoring data group of a target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods;

[0039] a data confirmation module, configured to perform pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and perform monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result for the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a slight change characteristic value and a data fluctuation characteristic value, the slight change characteristic value being used to characterize the maximum duration during which the difference between the maximum and minimum values ​​in the continuous monitoring data within a statistical period is less than a preset variation amplitude threshold, and the data fluctuation characteristic value being used to characterize the average of the fluctuation amplitudes of multiple monitoring periods;

[0040] a data anomaly discrimination module, configured to, if the monitoring failure discrimination result is valid, perform a long-term trend anomaly discrimination process on the second monitoring data group to obtain a first anomaly discrimination result, and perform a data mutation anomaly discrimination process on the second monitoring data group to obtain a second anomaly discrimination result, wherein the long-term trend anomaly discrimination is performed based on the long-term data trend change characteristic value, and the data mutation anomaly discrimination process is performed based on the data fluctuation characteristic value, the short-term data trend change characteristic, and the mean characteristic value, the long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group;

[0041] The discrimination result generating module is used to combine the first abnormality discrimination result and the second abnormality discrimination result to generate an abnormality discrimination result of the gas extraction monitoring data for the target sensor.

[0042] Preferably, the data confirmation module includes a false monitoring data screening unit, which is used to:

[0043] If the first monitoring data group is discrete data, identifying abnormal monitoring data in the first monitoring data group based on the Laida criterion and filtering out the abnormal monitoring data to obtain a second monitoring data group;

[0044] If the first monitoring data group is continuous data, the first abnormal interval in the first monitoring data group is identified based on the box-and-line method, and the first abnormal interval is verified based on the second abnormal interval of the adjacent sensor, so as to screen out the first abnormal interval according to the verification result and obtain the second monitoring data group.

[0045] Preferably, the data confirmation module includes a monitoring failure determination unit configured to:

[0046] Obtaining a preset small change duration threshold corresponding to the target sensor;

[0047] Counting a small change characteristic value of the second monitoring data group, and if the small change characteristic value is less than the preset small change duration threshold, determining that the preliminary monitoring failure determination result of the second monitoring data group is valid;

[0048] If the preliminary monitoring failure determination result is valid, obtaining a preset minimum fluctuation amplitude threshold corresponding to the target sensor;

[0049] Calculating a data fluctuation characteristic value of the second monitoring data group, and marking the current statistical period if the data fluctuation characteristic value is less than the preset minimum fluctuation amplitude threshold;

[0050] If the number of consecutively marked statistical cycles does not exceed the preset statistical cycle number threshold, the final monitoring failure determination result of the second monitoring data group is determined to be valid, and the monitoring failure determination result of the second monitoring data group is obtained.

[0051] Preferably, the data anomaly identification module includes a long-term trend anomaly identification unit, which is used to:

[0052] performing curve fitting processing on the second monitoring data set based on the least squares method to obtain a curve equation corresponding to the second monitoring data set, and determining the slope of the curve equation as the long-term data trend change characteristic value of the second monitoring data set;

[0053] If the long-term data trend change characteristic value is greater than a preset long-term trend change threshold, the first abnormality discrimination result is determined to be abnormal.

[0054] Preferably, the data anomaly identification module includes data mutation anomaly identification, which is used to:

[0055] Calculate the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value of the second monitoring data group respectively;

[0056] Obtaining a preset maximum fluctuation amplitude threshold corresponding to the target sensor and a plurality of long-term average characteristic values ​​corresponding to the target sensor, wherein the time lengths of the respective long-term average characteristic values ​​are different;

[0057] Calculating the deviations between the mean characteristic value and each of the long-term mean characteristic values ​​respectively, and screening out the maximum deviation value;

[0058] If the data fluctuation characteristic value is greater than the preset maximum fluctuation amplitude threshold, determining that the first data mutation abnormality determination result of the second monitoring data is abnormal;

[0059] If the short-term data trend change characteristic is an upward trend, determining that the second data mutation abnormality determination result of the second monitoring data is abnormal;

[0060] If the maximum deviation value is greater than the preset deviation value threshold, the third data mutation abnormality determination result of the second monitoring data is determined to be abnormal;

[0061] If any one of the first data mutation abnormality determination result, the second data mutation abnormality determination result, and the third data mutation abnormality determination result is abnormal, the second abnormality determination result is determined to be abnormal.

[0062] Preferably, before the data acquisition module, the device further includes a data collection module for:

[0063] collecting monitoring data of each sensor based on a driving module matched with a different category of sensor;

[0064] generating a monitoring data element basic attribute table corresponding to each monitoring data according to a preset data element standardization rule, and mapping and storing the monitoring data element basic attribute table and the corresponding monitoring data;

[0065] The data acquisition module is configured to:

[0066] According to the identity of the target sensor;

[0067] According to the identity, the matching monitoring data element basic attribute table is searched to obtain the monitoring data corresponding to the monitoring data element basic attribute table.

[0068] Preferably, the category of the sensor includes an online metering device, a gas extraction pump station device, and a drilling trajectory measurement device.

[0069] According to another aspect of the present application, a storage medium is provided, and at least one executable instruction is stored in the storage medium, the executable instruction causing a processor to perform operations corresponding to the above-described gas extraction monitoring data abnormality determination method.

[0070] According to still another aspect of the present application, a terminal is provided, and the terminal includes a processor, a memory, a communication interface, and a communication bus, the processor, the memory, and the communication interface complete communication with each other through the communication bus.

[0071] The memory is configured to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the above-described gas extraction monitoring data abnormality determination method.

[0072] Through the above technical solution, the technical solution provided by the embodiments of the present application has at least the following advantages:

[0073] The present application provides a method and device, medium, and terminal for distinguishing abnormalities in gas extraction monitoring data, which first obtains a first monitoring data group of a target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods; secondly, pseudo-monitoring data identification processing is performed on each monitoring data to filter out pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and monitoring failure distinction processing is performed on the second monitoring data group to obtain a monitoring failure distinction result for the second monitoring data group, wherein the monitoring failure distinction processing is performed based on a slight change characteristic value and a data fluctuation characteristic value, the slight change characteristic value is used to characterize the maximum duration that the difference between the maximum value and the minimum value in the continuous monitoring data in the statistical period is less than a preset change amplitude threshold, and the data fluctuation characteristic value is used to characterize the average of the fluctuation amplitudes of multiple monitoring periods; again, if If the monitoring failure judgment result is valid, the second monitoring data group is subjected to long-term trend abnormality judgment processing to obtain a first abnormality judgment result, and the second monitoring data group is subjected to data mutation abnormality judgment processing to obtain a second abnormality judgment result, wherein the long-term trend abnormality judgment is based on the long-term data trend change characteristic value, and the data mutation abnormality judgment processing is based on the data fluctuation characteristic value, the short-term data trend change characteristic and the mean characteristic value. The long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group; finally, the first abnormality judgment result and the second abnormality judgment result are combined to generate the gas extraction monitoring data abnormality judgment result for the target sensor. Compared with the prior art, the embodiment of the present application first performs false monitoring data screening and monitoring failure judgment processing on the current monitoring data to determine the validity of the monitoring data, and further, performs long-term trend anomaly judgment on the current monitoring data based on the long-term data trend change characteristic value to obtain a first abnormal judgment result, and performs data mutation anomaly judgment on the current monitoring data based on the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value to obtain a second abnormal judgment result, thereby generating a final abnormal judgment result, realizing the automated processing of abnormal judgment of gas extraction monitoring data, avoiding the problems of lag and misjudgment caused by reliance on manual judgment, and reducing the risk of coal and gas outburst accidents; and by calculating the characteristics of the monitoring data such as the small change characteristic value, long-term data trend change characteristic value, data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value, the inherent correlation and law between the data are revealed, so that the abnormal judgment result for the monitoring data is more accurate, further reducing the risk of coal and gas outburst accidents.

[0074] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0076] Figure 1 A flow chart of a method for identifying abnormalities in gas drainage monitoring data provided by an embodiment of the present application is shown;

[0077] Figure 2 A flow chart of another method for identifying abnormalities in gas drainage monitoring data provided by an embodiment of the present application is shown;

[0078] Figure 3 A block diagram of a device for identifying abnormalities in gas drainage monitoring data provided by an embodiment of the present application is shown;

[0079] Figure 4 A schematic diagram of the structure of a terminal provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0080] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0081] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0082] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0083] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0084] It should be noted that like reference numerals and characters refer to like elements throughout the following detailed description and the claims; additionally, the first time an item is mentioned in a figure is also indicated by the number in that figure where that item is first mentioned.

[0085] Embodiments of the present application can be applied to a computer system / server, which can operate in association with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and / or configurations that can be suitable for use with computer system / server include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputers, mainframe computer systems, and distributed cloud computing environments that include any of the above systems, and the like.

[0086] The computer system / server can be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules can include routines, programs, objects, components, logic, data structures, and the like, that perform particular tasks or implement particular abstract data types. Computer system / server can operate in a distributed cloud computing environment where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules can be located in local or remote computer system storage media including memory storage devices.

[0087] Embodiments of the present application provide a method for judging the abnormality of gas extraction monitoring data, as shown in the method includes: Figure 1

[0088] 101, obtaining a first monitoring data group of a target sensor in a current statistical period.

[0089] Wherein, the target sensor can be a sensor of an online metering device, a sensor of a gas extraction pump station device, and a sensor of a borehole trajectory measuring device, etc.; the current statistical period contains a plurality of monitoring periods, for example, the monitoring period is 1 min, and the statistical period is 30 min, so that the statistical period contains 30 monitoring periods, and the first monitoring data group contains 30 monitoring data; the first monitoring data group is used to represent the original monitoring data of the target sensor obtained from the database. In the embodiments of the present application, the current execution end can be a data abnormality judging module in the gas extraction system.

[0090] 102, respectively performing pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group, obtaining a second monitoring data group, and performing monitoring failure judgment processing on the second monitoring data group, obtaining a monitoring failure judgment result of the second monitoring data group. ​

[0091] Among them, the pseudo-monitoring data identification processing is used to characterize the operation of identifying pseudo-monitoring data from the first monitoring data group, with the purpose of ensuring that the data used for abnormality judgment are all data generated by real situations; the monitoring failure judgment processing is used to characterize the operation of judging whether the target sensor has failed, with the purpose of identifying the monitoring failure caused by the sensor itself failure or being blocked or dragged. It is based on the small change characteristic value and the data fluctuation characteristic value. The small change characteristic value is used to characterize the maximum duration of the difference between the maximum and minimum values ​​in the continuous monitoring data within the statistical period being less than the preset change amplitude threshold. The data fluctuation characteristic value is used to characterize the average of the fluctuation amplitudes of multiple monitoring periods; the second monitoring data group is used to characterize the monitoring data group after screening out the pseudo-monitoring data.

[0092] 103. If the monitoring failure judgment result is valid, the second monitoring data group is subjected to long-term trend abnormality judgment processing to obtain a first abnormality judgment result, and the second monitoring data group is subjected to data mutation abnormality judgment processing to obtain a second abnormality judgment result.

[0093] Among them, the long-term trend anomaly discrimination processing is used to characterize the operation of judging whether the long-term overall change trend of the second monitoring data group exceeds the preset long-term overall change trend, which is based on the long-term data trend change characteristic value, and the long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group; the data mutation anomaly discrimination processing is used to characterize the operation of judging whether a mutation anomaly occurs in the second monitoring data group, which is based on the data fluctuation characteristic value, the short-term data trend change characteristic and the mean characteristic value, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group; the first anomaly discrimination result is used to characterize the anomaly discrimination result obtained by the second monitoring data group after the long-term trend anomaly discrimination processing; the second anomaly discrimination result is used to characterize the anomaly discrimination result obtained by the second monitoring data group after the data mutation anomaly discrimination processing.

[0094] 104. Combine the first abnormality determination result and the second abnormality determination result to generate an abnormality determination result of the gas extraction monitoring data for the target sensor.

[0095] In the embodiment of the present application, the abnormality judgment result of the gas extraction monitoring data of the target sensor is the accumulation of two abnormality judgment results: long-term trend abnormality and data mutation abnormality.

[0096] Compared with the prior art, the embodiment of the present application first performs false monitoring data screening and monitoring failure judgment processing on the current monitoring data to determine the validity of the monitoring data, and further, performs long-term trend anomaly judgment on the current monitoring data based on the long-term data trend change characteristic value to obtain a first abnormal judgment result, and performs data mutation anomaly judgment on the current monitoring data based on the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value to obtain a second abnormal judgment result, thereby generating a final abnormal judgment result, realizing the automated processing of abnormal judgment of gas extraction monitoring data, avoiding the problems of lag and misjudgment caused by reliance on manual judgment, and reducing the risk of coal and gas outburst accidents; and by calculating the characteristics of the monitoring data such as the small change characteristic value, long-term data trend change characteristic value, data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value, the inherent correlation and law between the data are revealed, so that the abnormal judgment result for the monitoring data is more accurate, further reducing the risk of coal and gas outburst accidents.

[0097] The present application embodiment provides another method for identifying abnormalities in gas extraction monitoring data, such as Figure 2 As shown, the method includes:

[0098] 201. Based on a driving module adapted to different types of sensors, the monitoring data of each sensor is collected; according to a preset data element standardization rule, a basic attribute table of monitoring data elements corresponding to each monitoring data is generated, and the basic attribute table of monitoring data elements is mapped and stored with the corresponding monitoring data.

[0099] Among them, different categories of sensors may include sensors of online metering equipment, sensors of gas extraction pump station equipment, and sensors of drilling trajectory measurement equipment; the preset data element standardization rules may adopt existing industry standardization rules; the basic attribute table of monitoring data elements is shown in Table 1.

[0100] Table 1 Basic attributes of monitoring data elements

[0101]

[0102] Among them, the identification number represents the unique identifier of the metadata type; the Chinese name represents the Chinese name of the metadata element, metadata subset, or metadata entity; the English name represents the English name of the metadata entity, metadata element, or metadata subset, which is the English name of a certain characteristic of the information resource, generally using the full English name; the definition represents the basic content and attributes of the metadata subset, metadata entity, or metadata element; the data type represents the data type of the metadata subset, metadata entity, or metadata element, and the value domain description that allows operations on values ​​within its value domain, such as composite type, integer type, Boolean type, string type, date type, etc.; the value domain represents the range of values ​​that the metadata element can take.

[0103] 202. Obtain a first monitoring data set of a target sensor in a current statistical period.

[0104] Accordingly, step 202 of the embodiment specifically includes: according to the identity identifier of the target sensor; searching for a matching basic attribute table of monitoring data elements according to the identity identifier to obtain monitoring data corresponding to the basic attribute table of monitoring data elements.

[0105] Among them, the identity identifier can include a data element identification number, so as to find the corresponding monitoring data element basic attribute table according to the data element identification number, and then obtain the corresponding monitoring data based on the mapping relationship to realize the fusion of multi-source heterogeneous data.

[0106] 203. Perform pseudo monitoring data identification processing on each monitoring data to filter out the pseudo monitoring data in the first monitoring data group to obtain a second monitoring data group.

[0107] Correspondingly, step 203 of the embodiment specifically includes: if the first monitoring data group is discrete data, then based on the Laida criterion, the abnormal monitoring data in the first monitoring data group is identified and filtered out to obtain the second monitoring data group; if the first monitoring data group is continuous data, then based on the box-and-line method, the first abnormal interval in the first monitoring data group is identified, and the first abnormal interval is verified based on the second abnormal interval of the adjacent sensor, so as to filter out the first abnormal interval according to the verification result and obtain the second monitoring data group.

[0108] In an embodiment of the present application, when pseudo-monitoring data identification and processing is performed on each monitoring data separately, it is divided into two cases according to the type of monitoring data. First, the monitoring data is discrete data. At this time, the abnormal monitoring data in the first monitoring data group is identified based on the Laida criterion. Specifically, the mean and standard deviation of the historical monitoring data are calculated, and the range of the mean ±n times the standard deviation is taken as the normal interval. The monitoring data in the first monitoring data group that exceeds this interval is determined to be abnormal monitoring data and is screened out to obtain the second monitoring data group. Secondly, the monitoring data is discrete data. At this time, the first abnormal interval in the first monitoring data group is identified based on the box-and-plot method. Specifically, the abnormal value boundary is determined based on the box-and-plot method, and the interval exceeding the abnormal value boundary is taken as the first abnormal interval; similarly, the abnormal value boundary of the adjacent sensor is determined based on the box-and-plot method as the second abnormal interval; further, if the first abnormal interval and the second abnormal interval are in the same period, the first abnormal interval is determined to be normal monitoring data and the data is retained; if the first abnormal interval and the second abnormal interval are in different periods, or there is no second abnormal interval, the first abnormal interval is determined to be abnormal monitoring data, screened out, and the second monitoring data group is obtained.

[0109] 204. Perform monitoring failure determination processing on the second monitoring data group to obtain a monitoring failure determination result of the second monitoring data group.

[0110] Correspondingly, step 204 of the embodiment specifically includes: obtaining a preset small change duration threshold corresponding to the target sensor; counting the small change characteristic value of the second monitoring data group, if the small change characteristic value is less than the preset small change duration threshold, then judging that the preliminary monitoring failure judgment result of the second monitoring data group is valid; if the preliminary monitoring failure judgment result is valid, then obtaining the preset minimum fluctuation amplitude threshold corresponding to the target sensor; calculating the data fluctuation characteristic value of the second monitoring data group, if the data fluctuation characteristic value is less than the preset minimum fluctuation amplitude threshold, then marking the current statistical period; if the number of consecutively marked statistical periods does not exceed the preset statistical period number threshold, then judging that the final monitoring failure judgment result of the second monitoring data group is valid, and obtaining the monitoring failure judgment result of the second monitoring data group.

[0111] The preset slight fluctuation duration threshold and the preset minimum fluctuation amplitude threshold are set based on the slight fluctuation characteristic value and the data fluctuation characteristic value of historical monitoring data. In the embodiment of the application, the slight fluctuation characteristic value of the second monitoring data set is counted. Specifically, the initial length of the sliding window is first set, which can be the length of the monitoring period, and is gradually expanded to the length of the statistical period, and a fluctuation amplitude threshold is set. Further, the sliding window is slid by a preset step from the starting time of the statistical period, and the fluctuation amplitude of the maximum monitoring data and the minimum monitoring data in each window is calculated. If the fluctuation amplitude is less than the fluctuation amplitude threshold, the starting time and the ending time of the current window are recorded. Further, if the adjacent window also satisfies the condition that the fluctuation amplitude is less than the fluctuation amplitude threshold, they are combined into the same continuous period, and the longest continuous time is updated. The above window sliding and threshold comparison steps are repeated until the window covers the entire statistical period, and the longest continuous time, i.e. the slight fluctuation characteristic value of the second monitoring data set, is obtained. Further, if the slight fluctuation characteristic value of the second monitoring data set is less than the preset slight fluctuation duration threshold (the historical slight fluctuation characteristic value of the historical monitoring data is counted according to the above steps, and the preset slight fluctuation duration threshold is set according to the historical slight fluctuation characteristic value, and the preset slight fluctuation duration threshold must be greater than the minimum change time of 80% of the monitoring points), it is determined that the preliminary monitoring failure discrimination result of the second monitoring data set is valid. It should be noted that in the embodiment of the application, the slight fluctuation characteristic value is used instead of the data constant characteristic value (i.e. the maximum duration of continuous monitoring data without change in the statistical period), which can effectively avoid the data fluctuation caused by sensor measurement error. If the preliminary monitoring failure discrimination result is valid, further, the data fluctuation characteristic value of the second monitoring data set is calculated. Specifically, the true fluctuation amplitude of the monitoring data in each monitoring period is first calculated,

[0112] TR t =x t,high -x t,low ,

[0113] wherein TR t represents the true fluctuation amplitude in the monitoring period t, x t,high represents the maximum monitoring data in the monitoring period t, and x t,low represents the minimum monitoring data in the monitoring period t.

[0114] Further, the data fluctuation characteristic value of the second monitoring data set is calculated based on the following formula,

[0115]

[0116] wherein ATR T represents the data fluctuation characteristic value in the statistical period T, and ATR T-1It represents the data fluctuation characteristic value within the statistical period T-1, and N represents the number of monitoring periods;

[0117] It should be noted that the data fluctuation characteristic value for the first monitoring period can be calculated based on the following formula:

[0118]

[0119] Among them, ATR1 represents the data fluctuation characteristic value in the first statistical period.

[0120] Furthermore, if the data fluctuation characteristic value of the second monitoring data group is less than the preset minimum fluctuation amplitude threshold (the historical data fluctuation characteristic values ​​of the historical monitoring data are statistically calculated according to the above steps, and then the preset minimum fluctuation amplitude threshold is set according to the historical data fluctuation characteristic values, and the preset minimum fluctuation amplitude threshold must be less than the historical data fluctuation characteristic values ​​of eighty percent of the monitoring points), then the final monitoring failure judgment result of the second monitoring data group is determined to be valid, and the final monitoring failure judgment result is used as the monitoring failure judgment result of the second monitoring data group.

[0121] 205. If the monitoring failure determination result is valid, a long-term trend abnormality determination process is performed on the second monitoring data group to obtain a first abnormality determination result.

[0122] Correspondingly, step 205 of the embodiment specifically includes: based on the least squares method, performing curve fitting processing on the second monitoring data group to obtain the curve equation corresponding to the second monitoring data group, and determining the slope of the curve equation as the long-term data trend change characteristic value of the second monitoring data group; if the long-term data trend change characteristic value is greater than the preset long-term trend change threshold, the first abnormality discrimination result is determined to be abnormal.

[0123] The preset long-term trend change threshold is set based on the long-term data trend change characteristic value of the historical monitoring data, which is divided into two cases. First, when there is historical gas outburst accident monitoring data, the gas outburst accident monitoring data is subjected to curve fitting processing to obtain the accident curve equation, and the slope of the accident curve equation is set as the preset long-term trend change threshold; Second, when there is no historical gas outburst accident monitoring data, the preset long-term trend change threshold is set according to the long-term data trend change characteristic value of the conventional monitoring data. It should be noted that the preset long-term trend change threshold is greater than the maximum value of the long-term data trend change characteristic value of the conventional monitoring data. In the embodiment of the present application, the long-term trend abnormality discrimination processing is performed on the monitoring data group of a single sensor. In addition, the long-term trend abnormality discrimination processing can also be performed on the monitoring data of a certain area. Specifically, according to the location information of the target sensor, the monitoring data groups of other sensors in the same area as the target sensor are obtained, the slopes of the monitoring data groups of each other sensor are determined, and other sensors whose slopes are greater than the preset regional long-term trend change threshold are marked. If the number of marked other sensors is greater than the preset number threshold, the area is determined to be an abnormal area, wherein the preset regional long-term trend change threshold is less than the preset long-term trend change threshold.

[0124] 206. Perform data mutation abnormality discrimination processing on the second monitoring data group to obtain a second abnormality discrimination result.

[0125] Correspondingly, step 206 of the embodiment specifically includes: respectively calculating the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value of the second monitoring data group; obtaining the preset maximum fluctuation amplitude threshold corresponding to the target sensor and multiple long-term mean characteristic values ​​corresponding to the target sensor, wherein the time length of each long-term mean characteristic value is different; respectively calculating the deviation value between the mean characteristic value and each long-term mean characteristic value, and screening out the maximum deviation value; if the data fluctuation characteristic value is greater than the preset maximum fluctuation amplitude threshold, then determining that the first data mutation anomaly judgment result of the second monitoring data is abnormal; if the short-term data trend change characteristic is an upward trend, then determining that the second data mutation anomaly judgment result of the second monitoring data is abnormal; if the maximum deviation value is greater than the preset deviation value threshold, then determining that the third data mutation anomaly judgment result of the second monitoring data is abnormal; if any data mutation anomaly judgment result among the first data mutation anomaly judgment result, the second data mutation anomaly judgment result and the third data mutation anomaly judgment result is abnormal, then determining that the second anomaly judgment result is abnormal.

[0126] In the embodiment of the present application, the mean characteristic value of the second monitoring data group is first calculated according to the following formula:

[0127]

[0128] Among them, SMA T represents the mean characteristic value within the statistical period T, n represents the number of monitoring data contained in the statistical period T, x t-i Represents the monitoring data within the monitoring period ti, i = 0, 1, 2, ..., n-1;

[0129] It should be noted that if the mean characteristic value of the previous statistical period is known, the mean characteristic value of the current statistical period can be calculated according to the following formula:

[0130]

[0131] Among them, SMA T-1 represents the mean eigenvalue in the previous statistical period T-1, x t represents the monitoring data within the monitoring period t, x t-n Indicates the monitoring data within the monitoring period tn.

[0132] Furthermore, multiple long-term mean characteristic values ​​corresponding to the target sensor are obtained, such as for 1 day, 3 days, and 7 days, and the deviation between the mean characteristic value of the second monitoring data group and each long-term mean characteristic value is calculated. That is, the ratio of the difference between the mean characteristic value of the second monitoring data group and the long-term mean characteristic value to the long-term mean characteristic value is calculated. The maximum ratio is selected from the ratios corresponding to the long-term mean characteristic values ​​as the maximum deviation value. Furthermore, if the maximum deviation value is greater than a preset deviation value threshold, the third data mutation abnormality discrimination result of the second monitoring data is determined to be abnormal.

[0133] Secondly, calculate the short-term data trend change characteristics of the second monitoring data group. Specifically, first, for the second monitoring data group, by comparing the increase and decrease relationships of all data pairs in the time series data, an order sequence is constructed to reflect the overall trend direction of the data.

[0134]

[0135] Among them, S k represents the order sequence, k=1,2,...,n, x i represents the i-th monitoring data in the second monitoring data group, x j Represents the jth monitoring data in the second monitoring data group.

[0136] Furthermore, the order column S is calculated k The variance of , is divided into two cases, one is no repeated value,

[0137]

[0138] Among them, Var(S k) represents the order column S k variance;

[0139] Second, there are duplicate values.

[0140]

[0141] Among them, p represents the number of repeated values, g represents the number of non-repeated values, and t p Indicates the number of repetitions of the p-th repetition value.

[0142] Furthermore, the trend discrimination parameters are calculated.

[0143]

[0144] Among them, Z MK Represents the trend discrimination parameter.

[0145] Furthermore, if the error tolerance α is set to (0, 0.5), the confidence level of the trend determination result is

[0146] when When, there is no trend;

[0147] when And Z MK >0, upward trend;

[0148] when And Z MK <0, downward trend;

[0149] in, Indicates that the confidence level in the normal distribution is The value of the x-axis at time t is determined. Thus, the trend of the short-term data trend change feature of the second monitoring data group is determined. If it is an upward trend, the second data mutation abnormality determination result of the second monitoring data is determined to be abnormal.

[0150] Next, the data fluctuation characteristic value of the second monitoring data set is calculated. For details regarding step 204 in the aforementioned embodiment, which will not be repeated here, refer to the detailed description. Furthermore, a preset maximum fluctuation amplitude threshold corresponding to the target sensor is obtained. If the data fluctuation characteristic value of the second monitoring data set is greater than the preset maximum fluctuation amplitude threshold, the first data mutation abnormality determination result of the second monitoring data is determined to be abnormal.

[0151] Finally, if any one of the first data mutation abnormality discrimination result, the second data mutation abnormality discrimination result, and the third data mutation abnormality discrimination result is abnormal, the second abnormality discrimination result is determined to be abnormal.

[0152] 207. Combine the first abnormality determination result and the second abnormality determination result to generate an abnormality determination result of the gas extraction monitoring data for the target sensor.

[0153] For step 207 of the embodiment, reference may be made to the detailed description of step 104 of the aforementioned embodiment, which will not be repeated here.

[0154] In one embodiment of the present application, preferably, the second monitoring data group can be used as historical monitoring data to update the preset small change duration threshold, the preset minimum fluctuation amplitude threshold, the preset long-term trend change threshold, the preset maximum fluctuation amplitude threshold, and the preset deviation value threshold.

[0155] The application provides an abnormality discrimination method for gas extraction monitoring data. First, a first monitoring data group of a target sensor in a current statistical period is obtained, wherein the current statistical period comprises a plurality of monitoring periods; second, each monitoring data is subjected to pseudo-monitoring data identification processing to screen out pseudo-monitoring data in the first monitoring data group, to obtain a second monitoring data group, and the second monitoring data group is subjected to monitoring failure discrimination processing to obtain a monitoring failure discrimination result of the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a small variation characteristic value and a data fluctuation characteristic value, the small variation characteristic value is used to represent a maximum continuous time length in which a difference between a maximum value and a minimum value in continuous monitoring data in a statistical period is less than a preset variation amplitude threshold, and the data fluctuation characteristic value is used to represent a mean value of fluctuation amplitudes of the plurality of monitoring periods; third, if the monitoring failure discrimination result is valid, the second monitoring data group is subjected to long-term trend abnormality discrimination processing to obtain a first abnormality discrimination result, and the second monitoring data group is subjected to data mutation abnormality discrimination processing to obtain a second abnormality discrimination result, wherein the long-term trend abnormality discrimination is performed based on a long-term data trend change characteristic value, the data mutation abnormality discrimination processing is performed based on the data fluctuation characteristic value, a short-term data trend change characteristic and a mean value characteristic value, the long-term data trend change characteristic value is used to represent a slope of a fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to represent a recent dynamic change trend, and the mean value characteristic value is used to represent a mean value of the monitoring data group; and finally, the first abnormality discrimination result and the second abnormality discrimination result are combined to generate a gas extraction monitoring data abnormality discrimination result for the target sensor. Compared with the prior art, the application first performs pseudo-monitoring data screening processing and monitoring failure discrimination processing on current monitoring data to determine the validity of the monitoring data, further performs long-term trend abnormality discrimination on the current monitoring data based on a long-term data trend change characteristic value to obtain a first abnormality discrimination result, and performs data mutation abnormality discrimination on the current monitoring data based on a data fluctuation characteristic value, a short-term data trend change characteristic and a mean value characteristic value to obtain a second abnormality discrimination result, thereby generating a final abnormality discrimination result, realizing automatic processing of gas extraction monitoring data abnormality discrimination, avoiding lag and misjudgment caused by relying on manual discrimination, and reducing the risk of coal and gas outburst accidents; and by calculating the small variation characteristic value, the long-term data trend change characteristic value, the data fluctuation characteristic value, the short-term data trend change characteristic and the mean value characteristic value and the like, the internal correlation and rules between data are revealed, the abnormality discrimination result for the monitoring data is more accurate, and the risk of coal and gas outburst accidents is further reduced.

[0156] Further, as to the above Figure 1The embodiment of the present application provides a device for distinguishing abnormalities in gas extraction monitoring data, such as Figure 3 As shown, the device includes:

[0157] Data acquisition module 31, data confirmation module 32, data anomaly determination module 33, determination result generation module 34;

[0158] The data acquisition module 31 is used to acquire a first monitoring data group of the target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods;

[0159] a data confirmation module 32 for performing pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and performing monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result for the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a slight change characteristic value and a data fluctuation characteristic value. The slight change characteristic value is used to represent the maximum duration during which the difference between the maximum and minimum values ​​of the continuous monitoring data within a statistical period is less than a preset change amplitude threshold. The data fluctuation characteristic value is used to represent the average of the fluctuation amplitudes of multiple monitoring periods;

[0160] a data anomaly discrimination module 33 for, if the monitoring failure discrimination result is valid, performing a long-term trend anomaly discrimination process on the second monitoring data group to obtain a first anomaly discrimination result, and performing a data mutation anomaly discrimination process on the second monitoring data group to obtain a second anomaly discrimination result, wherein the long-term trend anomaly discrimination is performed based on the long-term data trend change characteristic value, and the data mutation anomaly discrimination process is performed based on the data fluctuation characteristic value, the short-term data trend change characteristic, and the mean characteristic value. The long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group;

[0161] The discrimination result generating module 34 is configured to combine the first abnormality discrimination result and the second abnormality discrimination result to generate an abnormality discrimination result of the gas drainage monitoring data for the target sensor.

[0162] In a specific application scenario, the data confirmation module includes a false monitoring data screening unit, which is used to:

[0163] If the first monitoring data group is discrete data, identifying abnormal monitoring data in the first monitoring data group based on the Laida criterion and filtering out the abnormal monitoring data to obtain a second monitoring data group;

[0164] If the first monitoring data set is continuous data, a first abnormal interval in the first monitoring data set is identified based on a box plot method, and the first abnormal interval is verified based on a second abnormal interval of an adjacent sensor to exclude the first abnormal interval according to a verification result, to obtain a second monitoring data set.

[0165] In a specific application scenario, the data confirmation module comprises a monitoring failure discrimination unit, configured to:

[0166] Obtain a preset micro-variation duration threshold corresponding to the target sensor;

[0167] Statistically obtain a micro-variation characteristic value of the second monitoring data set, and if the micro-variation characteristic value is less than the preset micro-variation duration threshold, determine that a preliminary monitoring failure discrimination result of the second monitoring data set is valid.

[0168] If the preliminary monitoring failure discrimination result is valid, obtain a preset minimum fluctuation amplitude threshold corresponding to the target sensor;

[0169] Calculate a data fluctuation characteristic value of the second monitoring data set, and if the data fluctuation characteristic value is less than the preset minimum fluctuation amplitude threshold, mark a current statistical period;

[0170] If the number of continuously marked statistical periods does not exceed a preset statistical period number threshold, determine that a final monitoring failure discrimination result of the second monitoring data set is valid, to obtain a monitoring failure discrimination result of the second monitoring data set.

[0171] In a specific application scenario, the data anomaly discrimination module comprises a long-term trend anomaly discrimination unit, configured to:

[0172] Based on a least square method, perform curve fitting processing on the second monitoring data set to obtain a curve equation corresponding to the second monitoring data set, and determine a slope of the curve equation as a long-term data trend change characteristic value of the second monitoring data set;

[0173] If the long-term data trend change characteristic value is greater than a preset long-term trend change threshold, determine that the first anomaly discrimination result is abnormal.

[0174] In a specific application scenario, the data anomaly discrimination module comprises a data mutation anomaly discrimination unit, configured to:

[0175] Respectively calculate a data fluctuation characteristic value, a short-term data trend change characteristic value, and a mean value characteristic value of the second monitoring data set;

[0176] Obtaining a preset maximum fluctuation amplitude threshold corresponding to the target sensor and a plurality of long-term average characteristic values ​​corresponding to the target sensor, wherein the time lengths of the respective long-term average characteristic values ​​are different;

[0177] Calculating the deviations between the mean characteristic value and each of the long-term mean characteristic values ​​respectively, and screening out the maximum deviation value;

[0178] If the data fluctuation characteristic value is greater than the preset maximum fluctuation amplitude threshold, determining that the first data mutation abnormality determination result of the second monitoring data is abnormal;

[0179] If the short-term data trend change characteristic is an upward trend, determining that the second data mutation abnormality determination result of the second monitoring data is abnormal;

[0180] If the maximum deviation value is greater than the preset deviation value threshold, the third data mutation abnormality determination result of the second monitoring data is determined to be abnormal;

[0181] If any one of the first data mutation abnormality determination result, the second data mutation abnormality determination result, and the third data mutation abnormality determination result is abnormal, the second abnormality determination result is determined to be abnormal.

[0182] In a specific application scenario, before the data acquisition module, the device further includes a data acquisition module for:

[0183] Collect monitoring data from various sensors based on driver modules adapted to different types of sensors;

[0184] Generate a monitoring data element basic attribute table corresponding to each monitoring data according to the preset data element standardization rule, and map and store the monitoring data element basic attribute table with the corresponding monitoring data;

[0185] The data acquisition module is used to:

[0186] According to the identification of the target sensor;

[0187] A matching monitoring data element basic attribute table is searched according to the identity identifier to obtain monitoring data corresponding to the monitoring data element basic attribute table.

[0188] In specific application scenarios, sensor categories include: online metering equipment, gas extraction pump station equipment, and drilling trajectory measurement equipment.

[0189] The present application provides an abnormality discrimination device for gas extraction monitoring data, which first obtains a first monitoring data group of a target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods; secondly, performs pseudo-monitoring data identification processing on each monitoring data to filter out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and performs monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result of the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a small change characteristic value and a data fluctuation characteristic value, the small change characteristic value is used to characterize the maximum duration that the difference between the maximum value and the minimum value in the continuous monitoring data in the statistical period is less than a preset change amplitude threshold, and the data fluctuation characteristic value is used to characterize the average of the fluctuation amplitudes of multiple monitoring periods; again, if the monitoring If the failure judgment result is valid, the second monitoring data group is subjected to long-term trend abnormality judgment processing to obtain a first abnormality judgment result, and the second monitoring data group is subjected to data mutation abnormality judgment processing to obtain a second abnormality judgment result, wherein the long-term trend abnormality judgment is based on the long-term data trend change characteristic value, and the data mutation abnormality judgment processing is based on the data fluctuation characteristic value, the short-term data trend change characteristic and the mean characteristic value. The long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group; finally, the first abnormality judgment result and the second abnormality judgment result are combined to generate the gas extraction monitoring data abnormality judgment result for the target sensor. Compared with the prior art, the embodiment of the present application first performs false monitoring data screening and monitoring failure judgment processing on the current monitoring data to determine the validity of the monitoring data, and further, performs long-term trend anomaly judgment on the current monitoring data based on the long-term data trend change characteristic value to obtain a first abnormal judgment result, and performs data mutation anomaly judgment on the current monitoring data based on the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value to obtain a second abnormal judgment result, thereby generating a final abnormal judgment result, realizing the automated processing of abnormal judgment of gas extraction monitoring data, avoiding the problems of lag and misjudgment caused by reliance on manual judgment, and reducing the risk of coal and gas outburst accidents; and by calculating the characteristics of the monitoring data such as the small change characteristic value, long-term data trend change characteristic value, data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value, the inherent correlation and law between the data are revealed, so that the abnormal judgment result for the monitoring data is more accurate, further reducing the risk of coal and gas outburst accidents.

[0190] According to one embodiment of the present application, a storage medium is provided, wherein the storage medium stores at least one executable instruction. The computer-executable instruction can execute the method for determining abnormality of gas drainage monitoring data in any of the above method embodiments.

[0191] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0192] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present application is shown. The specific embodiment of the present application does not limit the specific implementation of the terminal.

[0193] like Figure 4 As shown, the terminal may include: a processor (processor) 402 , a communications interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0194] The processor 402 , the communication interface 404 , and the memory 406 communicate with each other via a communication bus 408 .

[0195] The communication interface 404 is used to communicate with other devices such as clients or other servers.

[0196] The processor 402 is configured to execute the program 410 , and specifically to execute the relevant steps in the above-mentioned embodiment of the method for determining abnormality of gas drainage monitoring data.

[0197] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0198] Processor 402 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in a computer device may be processors of the same type, such as one or more CPUs, or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0199] a memory 406 for storing a program 410. The memory 406 can include a high-speed RAM memory and can also include a non-volatile memory, such as at least one disk memory.

[0200] The program 410 can specifically be used to cause the processor 402 to perform the following operations:

[0201] obtain a first monitoring data set of the target sensor in a current statistical period, wherein the current statistical period contains a plurality of monitoring periods;

[0202] respectively perform pseudo-monitoring data identification processing on each monitoring data to screen out pseudo-monitoring data in the first monitoring data set to obtain a second monitoring data set, and perform monitoring failure discrimination processing on the second monitoring data set to obtain a monitoring failure discrimination result of the second monitoring data set, wherein the monitoring failure discrimination processing is performed based on a small variation characteristic value and a data fluctuation characteristic value, the small variation characteristic value is used to represent a maximum continuous time length in which a difference between a maximum value and a minimum value in continuous monitoring data in a statistical period is less than a preset variation amplitude threshold, and the data fluctuation characteristic value is used to represent a mean value of fluctuation amplitudes of a plurality of monitoring periods;

[0203] if the monitoring failure discrimination result is valid, perform long-term trend anomaly discrimination processing on the second monitoring data set to obtain a first anomaly discrimination result, and perform data mutation anomaly discrimination processing on the second monitoring data set to obtain a second anomaly discrimination result, wherein the long-term trend anomaly discrimination is performed based on a long-term data trend change characteristic value, the data mutation anomaly discrimination processing is performed based on the data fluctuation characteristic value, a short-term data trend change characteristic, and a mean value characteristic value, the long-term data trend change characteristic value is used to represent a slope of a fitting curve equation of the monitoring data set, the short-term data trend change characteristic is used to represent a recent dynamic change trend, and the mean value characteristic value is used to represent a mean value of the monitoring data set;

[0204] combine the first anomaly discrimination result and the second anomaly discrimination result to generate a gas extraction monitoring data anomaly discrimination result for the target sensor.

[0205] The storage medium can also include an operating system, a network communication module. The operating system is a program for managing the hardware and software resources of the entity device of the above-described gas extraction monitoring data anomaly discrimination method, supporting the running of the information processing program and other software and / or programs. The network communication module is used to realize the communication between the components in the storage medium and the communication between other hardware and software in the information processing entity device.

[0206] Each embodiment in this specification is described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. References to the same or similar parts between the various embodiments are sufficient. For system embodiments, since they largely correspond to method embodiments, their description is relatively simple. For relevant parts, references to the description of the method embodiments are sufficient.

[0207] The methods and systems of the present application may be implemented in many ways. For example, the methods and systems of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present application. Therefore, the present application also covers recording media that store programs for executing the methods according to the present application.

[0208] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices. Alternatively, they can be implemented using program code executable by the computing device, so that they can be stored in a storage device and executed by the computing device. In some cases, the steps shown or described can be performed in a different order than herein, or they can be made into separate integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0209] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Persons skilled in the art will readily appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for identifying abnormalities in gas extraction monitoring data, characterized in that: include: Acquire a first monitoring data group of the target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods; Performing pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and performing monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result for the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a slight change characteristic value and a data fluctuation characteristic value, the slight change characteristic value is used to represent the maximum duration during which the difference between the maximum value and the minimum value in the continuous monitoring data within a statistical period is less than a preset change amplitude threshold, and the data fluctuation characteristic value is used to represent the average of the fluctuation amplitudes of multiple monitoring periods; If the monitoring failure judgment result is valid, a long-term trend abnormality judgment process is performed on the second monitoring data group to obtain a first abnormality judgment result, and a data mutation abnormality judgment process is performed on the second monitoring data group to obtain a second abnormality judgment result, wherein the long-term trend abnormality judgment is performed based on the long-term data trend change characteristic value, and the data mutation abnormality judgment process is performed based on the data fluctuation characteristic value, the short-term data trend change characteristic and the mean characteristic value. The long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group; The first abnormality determination result and the second abnormality determination result are combined to generate an abnormality determination result of the gas extraction monitoring data for the target sensor.

2. The method according to claim 1, characterized in that The pseudo monitoring data identification processing is performed on each monitoring data to filter out the pseudo monitoring data in the first monitoring data group to obtain the second monitoring data group, including: If the first monitoring data group is discrete data, identifying abnormal monitoring data in the first monitoring data group based on the Laida criterion and filtering out the abnormal monitoring data to obtain a second monitoring data group; If the first monitoring data group is continuous data, the first abnormal interval in the first monitoring data group is identified based on the box-and-line method, and the first abnormal interval is verified based on the second abnormal interval of the adjacent sensor, so as to screen out the first abnormal interval according to the verification result and obtain the second monitoring data group.

3. The method according to claim 1, characterized in that The performing monitoring failure determination processing on the second monitoring data group to obtain a monitoring failure determination result of the second monitoring data group includes: Obtaining a preset small change duration threshold corresponding to the target sensor; Counting a small change characteristic value of the second monitoring data group, and if the small change characteristic value is less than the preset small change duration threshold, determining that the preliminary monitoring failure determination result of the second monitoring data group is valid; If the preliminary monitoring failure determination result is valid, obtaining a preset minimum fluctuation amplitude threshold corresponding to the target sensor; Calculating a data fluctuation characteristic value of the second monitoring data group, and marking the current statistical period if the data fluctuation characteristic value is less than the preset minimum fluctuation amplitude threshold; If the number of consecutively marked statistical cycles does not exceed the preset statistical cycle number threshold, the final monitoring failure determination result of the second monitoring data group is determined to be valid, and the monitoring failure determination result of the second monitoring data group is obtained.

4. The method according to claim 1, wherein The performing long-term trend abnormality discrimination processing on the second monitoring data group to obtain a first abnormality discrimination result includes: performing curve fitting processing on the second monitoring data set based on the least squares method to obtain a curve equation corresponding to the second monitoring data set, and determining the slope of the curve equation as the long-term data trend change characteristic value of the second monitoring data set; If the long-term data trend change characteristic value is greater than a preset long-term trend change threshold, the first abnormality discrimination result is determined to be abnormal.

5. The method according to claim 1, wherein The performing data mutation abnormality discrimination processing on the second monitoring data group to obtain a second abnormality discrimination result includes: Calculate the data fluctuation characteristic value, short-term data trend change characteristic and mean characteristic value of the second monitoring data group respectively; Obtaining a preset maximum fluctuation amplitude threshold corresponding to the target sensor and a plurality of long-term average characteristic values ​​corresponding to the target sensor, wherein the time lengths of the respective long-term average characteristic values ​​are different; Calculating the deviations between the mean characteristic value and each of the long-term mean characteristic values ​​respectively, and screening out the maximum deviation value; If the data fluctuation characteristic value is greater than the preset maximum fluctuation amplitude threshold, determining that the first data mutation abnormality determination result of the second monitoring data is abnormal; If the short-term data trend change characteristic is an upward trend, determining that the second data mutation abnormality determination result of the second monitoring data is abnormal; If the maximum deviation value is greater than the preset deviation value threshold, the third data mutation abnormality determination result of the second monitoring data is determined to be abnormal; If any one of the first data mutation abnormality determination result, the second data mutation abnormality determination result, and the third data mutation abnormality determination result is abnormal, the second abnormality determination result is determined to be abnormal.

6. The method according to claim 1, characterized in that Before obtaining the first monitoring data group of the target sensor in the current statistical period, the method further includes: Collect monitoring data from various sensors based on driver modules adapted to different types of sensors; Generate a monitoring data element basic attribute table corresponding to each monitoring data according to the preset data element standardization rule, and map and store the monitoring data element basic attribute table with the corresponding monitoring data; The obtaining of a first monitoring data set of the target sensor in the current statistical period includes: According to the identification of the target sensor; A matching monitoring data element basic attribute table is searched according to the identity identifier to obtain monitoring data corresponding to the monitoring data element basic attribute table.

7. The method according to claim 6, characterized in that Sensor categories include: online metering equipment, gas extraction pump station equipment, and drilling trajectory measurement equipment.

8. A method for identifying abnormalities in gas extraction monitoring data, characterized in that: include: A data acquisition module, configured to acquire a first monitoring data group of a target sensor in a current statistical period, wherein the current statistical period includes multiple monitoring periods; a data confirmation module, configured to perform pseudo-monitoring data identification processing on each monitoring data to screen out the pseudo-monitoring data in the first monitoring data group to obtain a second monitoring data group, and perform monitoring failure discrimination processing on the second monitoring data group to obtain a monitoring failure discrimination result for the second monitoring data group, wherein the monitoring failure discrimination processing is performed based on a slight change characteristic value and a data fluctuation characteristic value, the slight change characteristic value being used to characterize the maximum duration during which the difference between the maximum and minimum values ​​in the continuous monitoring data within a statistical period is less than a preset variation amplitude threshold, and the data fluctuation characteristic value being used to characterize the average of the fluctuation amplitudes of multiple monitoring periods; a data anomaly discrimination module, configured to, if the monitoring failure discrimination result is valid, perform a long-term trend anomaly discrimination process on the second monitoring data group to obtain a first anomaly discrimination result, and perform a data mutation anomaly discrimination process on the second monitoring data group to obtain a second anomaly discrimination result, wherein the long-term trend anomaly discrimination is performed based on the long-term data trend change characteristic value, and the data mutation anomaly discrimination process is performed based on the data fluctuation characteristic value, the short-term data trend change characteristic, and the mean characteristic value, the long-term data trend change characteristic value is used to characterize the slope of the fitting curve equation of the monitoring data group, the short-term data trend change characteristic is used to characterize the recent dynamic change trend, and the mean characteristic value is used to characterize the mean of the monitoring data group; The discrimination result generating module is used to combine the first abnormality discrimination result and the second abnormality discrimination result to generate an abnormality discrimination result of the gas extraction monitoring data for the target sensor.

9. A storage medium storing at least one executable instruction, characterized in that: The executable instructions enable the processor to perform operations corresponding to the method for determining abnormality of gas extraction monitoring data according to any one of claims 1 to 7.

10. A terminal comprising: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, wherein the executable instruction enables the processor to perform operations corresponding to the abnormality judgment method for gas extraction monitoring data according to any one of claims 1 to 7.

Images