Coal seam gas pressure measurement data maintenance optimization method and system

Abstract

The application provides a coal seam gas pressure measurement data maintenance optimization method and system, relates to the technical field of data processing, and can effectively optimize the maintenance of coal seam gas pressure measurement data. The method comprises the following steps: acquiring coal seam gas pressure measurement data, and acquiring data correlation parameters, the data correlation parameters comprising geological parameters of a data acquisition position, an acquisition time stamp and a sensor working state, determining a data maintenance state based on the coal seam gas pressure measurement data and the data correlation parameters, and performing a maintenance optimization operation on the coal seam gas pressure measurement data based on the data maintenance state.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for maintaining and optimizing coal seam gas pressure measurement data. Background Technology

[0002] Coal seam gas pressure data is crucial for coal mine safety, but it is prone to anomalies due to complex geology, harsh sensor environments, and highly dynamic data. Traditional processing methods are insufficient to handle these anomalies, necessitating efficient maintenance methods.

[0003] Currently, the identification and maintenance of anomalies in coal seam gas pressure data mainly suffer from the following limitations:

[0004] First, some methods rely on a single threshold or simple logical judgment. For example, an anomaly is only identified when the gas pressure exceeds a fixed safety threshold. This method cannot effectively handle soft anomalies caused by slow changes in geological conditions or gradual drift in sensor performance, and is prone to false negatives. At the same time, it may generate a large number of false alarms for normal pressure fluctuations caused by mining activities.

[0005] Second, the data dimensions are too narrow and lack comprehensive analysis. Existing technologies rarely incorporate key factors affecting gas pressure, such as geological parameters like coal seam thickness and fracture development, as well as operational parameters like sensor power supply voltage and sampling frequency, into anomaly detection models. This "data-only" approach makes judgments lack a basis and fails to distinguish between genuine geological changes and equipment malfunctions at their root.

[0006] Third, the maintenance strategies are static and lack dynamic adaptability. In traditional maintenance methods, the judgment thresholds and repair algorithms are rarely changed once set, which cannot adapt to the dynamic evolution of the geological environment and monitoring system during coal mine production, resulting in a significant decline in the effectiveness of data maintenance over time.

[0007] Therefore, there is an urgent need in this field for a coal seam gas pressure data optimization method that can integrate multi-source information, dynamically adjust strategies, and achieve intelligent maintenance throughout the entire life cycle.

[0008] Therefore, the present invention provides a method and system for maintaining and optimizing coal seam gas pressure measurement data. Summary of the Invention

[0009] This application provides a method and system for maintaining and optimizing coal seam gas pressure measurement data, which can effectively optimize the maintenance of coal seam gas pressure measurement data.

[0010] To achieve the above objectives, this application adopts the following technical solution:

[0011] This invention provides a method for maintaining and optimizing coal seam gas pressure measurement data, including:

[0012] Obtain coal seam gas pressure measurement data and data correlation parameters, including geological parameters of the data acquisition location, acquisition timestamp, and sensor operating status;

[0013] The data maintenance status was determined by conducting a multi-dimensional joint analysis of coal seam gas pressure measurement data and data correlation parameters.

[0014] Based on the data maintenance status, perform maintenance and optimization operations on the coal seam gas pressure measurement data.

[0015] Preferably, coal seam gas pressure measurement data is obtained, and data correlation parameters are obtained, including:

[0016] Real-time gas pressure values ​​at different monitoring points in the coal seam are obtained to obtain coal seam gas pressure measurement data;

[0017] The coal seam thickness, fracture development density, and distance from the fault corresponding to each monitoring point were obtained as geological parameters.

[0018] The sensor's built-in clock records the data acquisition time and generates a timestamp.

[0019] The sensor's operating status is determined by acquiring the power supply voltage, sampling frequency, and calibration records through the sensor's self-test module.

[0020] By integrating the original measurement data, geological parameters, timestamps, and sensor operating status, we obtain the coal seam gas pressure measurement data and data correlation parameters to be processed.

[0021] Preferably, the data maintenance states include: a first maintenance state and a second maintenance state; the first maintenance state is an abnormal data state, and the second maintenance state is a normal data state.

[0022] Preferably, the data maintenance status is determined based on coal seam gas pressure measurement data and data correlation parameters, including:

[0023] Obtain the initial judgment threshold;

[0024] The coal seam gas pressure measurement data and the initial judgment threshold are analyzed first.

[0025] The execution status of the second analysis is determined based on the results of the first analysis;

[0026] A second analysis is performed on the data association parameters based on the execution status of the second analysis;

[0027] The first maintenance state and the second maintenance state are determined based on the results of the first analysis and the second analysis.

[0028] Preferably, obtaining the initial judgment threshold includes:

[0029] A pressure value sequence matrix is ​​constructed based on coal seam gas pressure measurement data, and a first feature matrix is ​​generated by combining geological parameters, timestamps and sensor working status in the data association parameters.

[0030] The first feature matrix is ​​normalized to obtain the standardized feature matrix;

[0031] The local density and distance parameters of each data point in the standardized feature matrix are determined based on the standardized feature matrix.

[0032] The initial anomaly detection threshold is determined based on the local density and distance parameters of each data point.

[0033] Preferably, the local density and distance parameters of each data point in the standardized feature matrix are determined based on the standardized feature matrix, including:

[0034] Obtain the cutoff distance parameter;

[0035] The Euclidean distance between each first data point and all other data points is determined based on the truncation distance parameter.

[0036] The first number of points is determined based on Euclidean distance and cutoff distance, and the first number of points is used as the local density of the first data point.

[0037] For each first data point, filter out all data points with a local density greater than the first data point, and take the minimum Euclidean distance among them as the distance parameter of the first data point;

[0038] A decision graph is constructed based on local density and distance parameters, and the parameter value corresponding to the density peak point in the decision graph is taken as the initial anomaly judgment threshold.

[0039] Preferably, when in the first maintenance state, maintenance optimization operations are performed on the coal seam gas pressure measurement data based on the data maintenance state, including:

[0040] Spatial interpolation correction is performed on the abnormal data based on data from one or more monitoring points that are spatially adjacent to the abnormal data and are in the second maintenance state.

[0041] Alternatively, based on a historical data sequence that is continuous in time, the outlier data can be fitted and corrected by time trend fitting.

[0042] Based on the corrected abnormal data, an early warning message is generated that includes the sensor number, abnormal parameter items, and recommended maintenance measures.

[0043] Preferably, when in the second maintenance state, maintenance optimization operations are performed on the coal seam gas pressure measurement data based on the data maintenance state, including:

[0044] A sliding window smoothing algorithm was used to filter the coal seam gas pressure measurement data to eliminate random noise;

[0045] Alternatively, a recursive filtering algorithm can be used to optimize the state estimation of the measured data in order to correct systematic errors.

[0046] Preferably, the present invention provides a method for maintaining and optimizing coal seam gas pressure measurement data, which further includes:

[0047] A data quality lifecycle assessment model is constructed. The inputs of the data quality lifecycle assessment model include the original data and maintenance data of coal seam gas pressure measurement data and data correlation parameters under maintenance status. The output is the credibility level of coal seam gas pressure measurement data and data correlation parameters.

[0048] The initial judgment threshold is dynamically adjusted based on the credibility level.

[0049] This invention provides a coal seam gas pressure measurement data maintenance and optimization system, the system comprising:

[0050] Data acquisition module: Acquires coal seam gas pressure measurement data and data association parameters, including geological parameters of the data acquisition location, acquisition timestamp, and sensor operating status;

[0051] Status acquisition module: Determines the data maintenance status by performing multi-dimensional joint analysis of coal seam gas pressure measurement data and data correlation parameters;

[0052] Data optimization module: Performs maintenance and optimization operations on coal seam gas pressure measurement data based on data maintenance status.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0054] By integrating coal seam gas pressure data with multi-dimensional correlation parameters such as geological parameters and sensor status, the accuracy of anomaly data identification is improved, reducing false positives and false negatives. Combined with dynamic threshold adjustment and spatiotemporal correlation verification, accurate anomaly data repair is achieved, improving data reliability. A full lifecycle assessment and closed-loop feedback mechanism is constructed, enabling maintenance strategies to adapt to geological changes and data quality fluctuations, enhancing the scientific basis of coal mine gas management decisions and providing reliable data support for safe production. Specifically, through the multi-dimensional fusion analysis and dynamic closed-loop feedback mechanism provided by this invention, the overall accuracy of anomaly data identification can be increased from approximately 75% using traditional methods to over 95% in simulation tests. Furthermore, through refined optimization of normal data, the accuracy of subsequent gas emission prediction models is improved by approximately 15%, thus significantly enhancing the scientific rigor and foresight of coal mine gas management decisions. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart illustrating the method for maintaining and optimizing coal seam gas pressure measurement data provided in this application embodiment;

[0057] Figure 2 A schematic diagram of the structure of the coal seam gas pressure measurement data maintenance and optimization system provided in the embodiments of this application.

[0058] Figure 3 This is a schematic diagram of the density peak in the decision graph provided in the embodiments of this application. Detailed Implementation

[0059] In the embodiments of this application, in order to clearly describe the technical solutions of the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different. The technical features described by "first" and "second" have no sequential or size order.

[0060] In the embodiments of this application, the words "exemplarily" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplarily" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0061] In the embodiments of this application, at least one can also be described as one or more, and multiple can be two, three, four or more, and this application does not impose any restrictions.

[0062] Furthermore, the network architecture and scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0063] For example, with the development of intelligent coal mines, gas pressure monitoring has gradually become automated, collecting data in real time through distributed sensor arrays. However, the monitoring process is affected by multiple factors, such as environmental interference, equipment limitations, and data characteristics. Currently, the reliability of gas pressure data directly affects decisions such as mine ventilation design and gas extraction scheme formulation. Therefore, it is urgent to establish a maintenance method that can comprehensively consider multiple factors and dynamically optimize data quality.

[0064] Existing coal seam gas pressure data maintenance technologies primarily employ single threshold methods or simple time-series analysis, failing to consider geological parameters and sensor operating status. This leads to a high rate of misjudgment of abnormal data. Data maintenance methods mainly rely on manual verification or simple interpolation corrections, lacking intelligent repair mechanisms. Furthermore, the technologies fail to effectively integrate multi-dimensional data such as geological parameters, timestamps, and sensor status, resulting in a single dimension for data quality assessment and an inability to comprehensively reflect the reliability of gas pressure data.

[0065] Based on this, this application provides a method and system for maintaining and optimizing coal seam gas pressure measurement data. By integrating coal seam gas pressure data with multi-dimensional related parameters such as geological parameters and sensor status, it improves the accuracy of abnormal data identification and reduces false positives and false negatives. Combined with dynamic threshold adjustment and spatiotemporal correlation verification, it achieves accurate repair of abnormal data and improves data credibility. It constructs a full life cycle assessment and closed-loop feedback mechanism to make the maintenance strategy adaptable to geological changes and data quality fluctuations, enhances the scientific nature of coal mine gas management decisions, and provides reliable data support for safe production.

[0066] The solutions provided by the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0067] Example 1

[0068] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data, including:

[0069] The data on coal seam gas pressure measurement is obtained, and the data association parameters are obtained. The data association parameters include the geological parameters of the data acquisition location, the acquisition timestamp, and the sensor working status.

[0070] The data maintenance status is determined by performing multi-dimensional joint analysis on the coal seam gas pressure measurement data and the data correlation parameters.

[0071] Based on the aforementioned data maintenance status, maintenance and optimization operations are performed on the coal seam gas pressure measurement data. The beneficial effects of the above technical solution are: by integrating coal seam gas pressure data with multi-dimensional related parameters such as geological parameters and sensor status, the accuracy of abnormal data identification is improved, reducing false positives and false negatives; combined with dynamic threshold adjustment and spatiotemporal correlation verification, accurate repair of abnormal data is achieved, improving data reliability; and a full life-cycle assessment and closed-loop feedback mechanism is constructed, enabling the maintenance strategy to adapt to geological changes and data quality fluctuations, enhancing the scientific nature of coal mine gas management decisions, and providing reliable data support for safe production.

[0072] Example 2

[0073] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data, which involves acquiring coal seam gas pressure measurement data and obtaining data correlation parameters, including:

[0074] Real-time gas pressure values ​​at different monitoring points in the coal seam are obtained to obtain coal seam gas pressure measurement data;

[0075] The coal seam thickness, fracture development density, and distance from the fault corresponding to each monitoring point were obtained as geological parameters.

[0076] The sensor's built-in clock records the data acquisition time and generates a timestamp.

[0077] The sensor's operating status is determined by acquiring the power supply voltage, sampling frequency, and calibration records through the sensor's self-test module.

[0078] By integrating the original measurement data, geological parameters, timestamps, and sensor operating status, we obtain the coal seam gas pressure measurement data and data correlation parameters to be processed.

[0079] In this embodiment, the real-time gas pressure values ​​at different monitoring points in the coal seam can be obtained by deploying a distributed array of gas pressure sensors at different depths and levels in the coal seam. Specifically, this includes: deploying fiber optic pressure sensors or piezoelectric pressure sensors in key areas such as around the mining face, near fault zones, and gas-rich areas. The sensors transmit the real-time collected pressure signals to the ground monitoring terminal through wired or wireless transmission modules. The terminal filters and converts the original signals from analog to digital to generate a sequence of real-time gas pressure values ​​for each monitoring point.

[0080] In this embodiment, the coal seam thickness, fracture development density, and distance from the fault corresponding to each monitoring point can be obtained through multi-source geological data fusion.

[0081] Specifically, the coal seam thickness is determined by combining core drilling data with 3D seismic exploration technology to form a coal seam thickness distribution map at the location of the monitoring point. The fracture development density is based on fracture images obtained by downhole borehole inspection instruments. The number of fractures per unit length is counted by image recognition technology. The distance to the fault is calculated by using the mine geological profile map and GPS coordinates of the monitoring point to obtain the straight-line distance between each monitoring point and the nearest fault. The above parameters are linked to the spatial coordinates of the corresponding monitoring point through a geographic information system.

[0082] In this embodiment, the sensor's operating status refers to the sensor's operating parameters and health status during the data acquisition process.

[0083] Specifically, the power supply voltage reflects the stability of the sensor's power supply, the sampling frequency reflects the time interval of data acquisition, and the calibration record includes the most recent calibration time, calibration error value, and whether it is within the validity period. These parameters are monitored in real time by the self-test chip built into the sensor and uploaded synchronously with the gas pressure data to evaluate the reliability of the measurement data.

[0084] The beneficial effects of the above technical solution are as follows: by collecting and integrating data from multiple dimensions, the comprehensiveness of the data is ensured, real-time pressure values ​​are integrated with geological, time, and equipment status parameters, the limitations of single data are avoided, the correlation of data is improved, multi-source verification basis is provided for subsequent anomaly judgment, misjudgment is reduced, the foundation for data standardization is laid, the integrated data format is unified, which facilitates the efficient implementation of subsequent maintenance and optimization processes, and provides a reliable data source for the accurate identification and handling of gas pressure data anomalies.

[0085] Example 3

[0086] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data. The data maintenance states include: a first maintenance state and a second maintenance state; the first maintenance state is an abnormal data state, and the second maintenance state is a normal data state.

[0087] In this embodiment, the first maintenance state refers to a state in which the coal seam gas pressure measurement data has unreliable or abnormal characteristics, and needs to be repaired, verified or maintained by technical means.

[0088] Specifically, the determination of this state is based on the degree of deviation between the coal seam gas pressure measurement data and the initial determination threshold, the abnormal characteristics of the data correlation parameters, and the comprehensive analysis results of the two, covering the abnormality of the data itself and the correlation factors that cause the data abnormality.

[0089] For example, if the real-time gas pressure value at a certain monitoring point exceeds the upper limit of the initial judgment threshold by 20%, and the sensor self-test module shows that the power supply voltage is continuously lower than 85% of the rated value, it is determined to be in the first maintenance state, and the power supply of the sensor needs to be repaired immediately and the abnormal data corrected.

[0090] In this embodiment, the second maintenance state refers to the state in which the coal seam gas pressure measurement data meets the reliability standards, requires no additional repair or emergency treatment, and can be directly used for subsequent gas control analysis.

[0091] Specifically, under the second maintenance state, the gas pressure measurement data is within the initial judgment threshold range, and the data correlation parameters all meet the preset conditions, indicating high spatiotemporal consistency and reliability of the data.

[0092] For example, if the gas pressure value at a certain monitoring point is stable within the initial judgment threshold range, and the deviations of geological parameters such as coal seam thickness and fracture development density from those of the three adjacent monitoring points are all less than 5%, and the sensor power supply voltage and sampling frequency are normal and the calibration record is within the validity period, it is judged to be in the second maintenance state after comprehensive evaluation, and the data can be directly used for gas pressure distribution pattern analysis.

[0093] The beneficial effects of the above technical solution are as follows: By clearly defining the first and second maintenance states, precise classification management is achieved: abnormal data is processed first to avoid misleading decisions with erroneous data, while normal data is efficiently retained to reduce the cost of ineffective processing. The classification criteria combine the data itself with related parameters to ensure scientific status determination, providing a clear basis for subsequent targeted optimization, improving the accuracy and efficiency of gas pressure data maintenance, and ensuring the reliability of data-driven coal mine safety decisions.

[0094] Example 4

[0095] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data. Based on the coal seam gas pressure measurement data and data correlation parameters, the method determines the data maintenance status, including:

[0096] Obtain the initial judgment threshold;

[0097] The coal seam gas pressure measurement data and the initial judgment threshold are analyzed first.

[0098] The execution status of the second analysis is determined based on the results of the first analysis;

[0099] A second analysis is performed on the data association parameters based on the execution status of the second analysis;

[0100] The first maintenance state and the second maintenance state are determined based on the results of the first analysis and the second analysis.

[0101] In this embodiment, the first analysis refers to the preliminary anomaly screening analysis of coal seam gas pressure measurement data based on the initial judgment threshold. The core is to quickly distinguish whether the data has the characteristic of "significantly exceeding the reasonable range" by directly comparing the single or batch gas pressure measurement data with the preset initial judgment threshold, and initially classify the results into two categories: "suspected abnormal" or "temporarily normal", providing a preliminary judgment basis for whether further in-depth verification is needed.

[0102] The second analysis refers to the in-depth verification and supplementary judgment analysis based on the data correlation parameters of the first analysis. It performs secondary verification on the preliminary results of the first analysis from three dimensions: spatial correlation, temporal stability, and data acquisition reliability. It corrects the misjudgments that may be caused by relying solely on pressure values. Essentially, it is a multi-dimensional verification of the authenticity of data anomalies or the stability of normal data.

[0103] Specifically, the first analysis involves comparing the coal seam gas pressure measurement data at each monitoring point with the initial judgment threshold, such as the historical normal pressure range of the coal seam [Pmin, Pmax]. If the data is greater than Pmax or less than Pmin, the first analysis result is "suspected abnormality"; if the data is within the range of [Pmin, Pmax], the first analysis result is "provisionally normal".

[0104] Second analysis: If the result of the first analysis is "suspected anomaly": the cause of the anomaly can be further verified through the second analysis. It is not a problem with the data itself, but a problem with the acquisition environment or equipment. For example, check the working status of the sensor. If the sensor status is abnormal, equipment interference needs to be eliminated before re-judging. If the sensor status is normal, the "suspected anomaly" is confirmed to be a real anomaly, corresponding to the first maintenance status.

[0105] If the first analysis result is "temporarily normal", the data stability needs to be further verified through a second analysis: This involves spatial correlation verification using geological parameters. Using fracture development density as the core weight and coal seam thickness and distance from the fault as auxiliary weights, a gas pressure spatial distribution prediction surface is generated through Kriging interpolation. The absolute deviation between the measured value and the predicted value at the corresponding location of the "temporarily normal" data is calculated. If the deviation ratio (absolute deviation / preset deviation threshold) > 1, it indicates that the data does not match the reasonable pressure value under the surrounding geological conditions, and is determined to be in the first maintenance state. Temporal stability verification using timestamps is also performed. A fixed-duration sliding window is constructed based on the acquisition timestamp, and the fluctuation variance of the "temporarily normal" data within the window is calculated. If the variance is greater than the preset stability threshold, it indicates that the data exhibits irregular jumps in time, and is determined to be in the first maintenance state. If both spatial correlation verification and temporal stability verification pass, and the sensor's working status is normal, then "temporarily normal" is confirmed as truly normal, corresponding to the second maintenance state.

[0106] In this embodiment, the initial judgment threshold refers to the critical value used for preliminary screening of coal seam gas pressure measurement data, which is an abnormal data judgment benchmark determined based on multi-dimensional feature analysis.

[0107] In this embodiment, determining the first maintenance state and the second maintenance state based on the results of the first analysis and the second analysis includes: determining the coal seam gas pressure measurement data that exceeds the preset pressure threshold range as the first maintenance state;

[0108] Optionally, if the data exceeds the threshold range, it is determined to be in the first maintenance state; if the data is within the threshold range, further analysis is performed in conjunction with the data correlation parameters.

[0109] Optionally, a spatial correlation model can be constructed by combining geological parameters to calculate the spatial correlation coefficient between suspected abnormal data and data from surrounding monitoring points. If the coefficient is lower than a preset correlation threshold, it is determined to be in the first maintenance state.

[0110] Optionally, perform time-series stability analysis on the normal dataset: construct a sliding window based on timestamps, calculate the variance of data fluctuations within the window, and if the variance exceeds a preset stability threshold, determine it as the first maintenance state; otherwise, determine it as the second maintenance state.

[0111] Optionally, a spatial correlation model is constructed by combining geological parameters, including: constructing a three-dimensional geological weight model with coal seam thickness, fracture development density, and distance from fault as input variables, in which the weight coefficient of fracture development density is higher than that of coal seam thickness and distance from fault; performing spatial interpolation on the normal dataset based on the Kriging interpolation method to generate a gas pressure spatial distribution prediction surface; calculating the absolute deviation between the measured value of suspected abnormal data and the predicted value at the corresponding location of the spatial distribution prediction surface, and using the ratio of the absolute deviation to a preset deviation threshold as the spatial correlation coefficient.

[0112] The beneficial effects of the above technical solution are: through the initial threshold screening and the two-layer analysis mechanism of related parameters, obvious abnormal data can be quickly identified, the judgment efficiency can be improved, the misjudgment of a single threshold can be avoided, and the maintenance status can be accurately classified by combining the two-dimensional results, providing clear guidance for subsequent optimization. This reduces the risk of missed judgment and the cost of ineffective processing, and significantly improves the accuracy and reliability of gas pressure data status judgment.

[0113] Example 5

[0114] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data, including obtaining an initial judgment threshold, comprising:

[0115] A pressure value sequence matrix is ​​constructed based on coal seam gas pressure measurement data, and a first feature matrix is ​​generated by combining geological parameters, timestamps and sensor working status in the data association parameters.

[0116] The first feature matrix is ​​normalized to obtain the standardized feature matrix;

[0117] The local density and distance parameters of each data point in the standardized feature matrix are determined based on the standardized feature matrix.

[0118] The initial anomaly detection threshold is determined based on the local density and distance parameters of each data point.

[0119] In this embodiment, the first feature matrix refers to the structured data set formed after integrating core data on coal seam gas pressure with multi-dimensional correlation parameters.

[0120] Specifically, rows represent data records at different monitoring times or different monitoring points, and columns contain the following features: coal seam gas pressure measurement values, geological parameters, timestamps, and sensor operating status parameters.

[0121] In this embodiment, the local density and distance parameters of each data point refer to two key indicators used to describe the clustering characteristics of data points in the standardized feature matrix.

[0122] In this embodiment, local density refers to the number of data points with similar characteristics to the current data point within a preset cutoff distance range, reflecting the degree of clustering of the data point in the feature space. The larger the number, the more the characteristics of the data point conform to the distribution pattern of most data.

[0123] In this embodiment, the distance parameter refers to the minimum Euclidean distance between the current data point and all data points with a local density higher than its own, reflecting the distance between the data point and the denser clustered area. The larger the distance, the more the characteristics of the data point deviate from the mainstream characteristics of high-density clustering.

[0124] The beneficial effects of the above technical solution are as follows: It determines the initial anomaly detection threshold through multi-dimensional feature fusion and density clustering analysis; it integrates multi-source information such as pressure data and geological parameters, avoiding threshold bias caused by a single data dimension; normalization processing eliminates differences in parameter dimensions, ensuring the comparability of feature matrices; and it captures data distribution patterns through local density and distance parameters, enabling the threshold to adapt to the gas pressure characteristics under complex geological environments. Ultimately, this improves the scientific validity and dynamic adaptability of the threshold, laying a reliable foundation for accurate identification of subsequent anomaly data.

[0125] Example 6

[0126] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data. Based on a standardized feature matrix, it determines the local density and distance parameters of each data point in the standardized feature matrix, including:

[0127] Obtain the cutoff distance parameter;

[0128] The Euclidean distance between each first data point and all other data points is determined based on the truncation distance parameter.

[0129] The first number of points is determined based on Euclidean distance and cutoff distance, and the first number of points is used as the local density of the first data point.

[0130] For each first data point, filter out all data points with a local density greater than the first data point, and take the minimum Euclidean distance among them as the distance parameter of the first data point;

[0131] A decision graph is constructed based on local density and distance parameters, and the parameter value corresponding to the density peak point in the decision graph is taken as the initial anomaly judgment threshold.

[0132] In this embodiment, the cutoff distance parameter refers to the critical distance threshold used to measure the similarity of data point features, which serves as the benchmark for dividing nearest neighbor data in the standardized feature matrix. Specifically, its value needs to be dynamically determined in conjunction with the data distribution density. Typically, it is determined by statistically analyzing the cumulative distribution of pairwise Euclidean distances among all data points in the standardized feature matrix, and taking a distance value with a cumulative probability of 1%-5% as the cutoff distance.

[0133] For example, when more than 95% of the data points are less than 0.8 in pairs, the cutoff distance can be set to 0.8 to ensure that the local clustering characteristics of the data points are captured while avoiding the inclusion of too many irrelevant data points.

[0134] In this embodiment, determining the first point number based on Euclidean distance and truncation distance means counting the number of other data points whose Euclidean distance to the current first data point is less than or equal to the truncation distance.

[0135] Specifically, for each first data point, calculate its Euclidean distance to all other data points in the standardized feature matrix. If the Euclidean distance between a data point and the first data point is less than or equal to the cutoff distance, it is considered a nearest neighbor. The total number of all nearest neighbors that meet this condition is taken as the first point number. This value directly represents the degree of clustering of the current data point in the local feature space and is the core basis for calculating the local density.

[0136] In this embodiment, taking the parameter value corresponding to the density peak point in the decision graph as the initial anomaly judgment threshold includes: the local density is significantly higher than that of the adjacent data points; the distance parameter is large, indicating that the point is far away from higher density points; the product of the local density and the distance parameter of the density peak point is greater than a preset threshold; the local density value corresponding to the density peak point is used as the initial anomaly judgment threshold, wherein the density peak point is an isolated point in the decision graph that simultaneously satisfies high local density and high distance parameter, and its local density value is significantly higher than that of the surrounding data points and the distance parameter is at a high level.

[0137] Figure 3An illustrative diagram of density peaks in a decision graph is provided, with density peaks marked in black. The horizontal axis represents local density, which is the degree of clustering of data points and reflects the tightness of clustering of data points in the corresponding region. The vertical axis represents the distance parameter, which is the distance to high-density points, indicating the distance between data points and points with higher local density.

[0138] The local density on the horizontal axis is dimensionless, and the distance parameter on the vertical axis is consistent with the unit of measurement of the original data. For example, if the unit of the original data is "meter (m)"; if the unit of the original data is the corresponding physical quantity unit, such as "megapascal (MPa)", it is determined by the measurement dimension of the data itself.

[0139] The beneficial effects of the above technical solution are as follows: Firstly, by scientifically quantifying the local clustering characteristics of data points to determine the initial threshold, the rationality of local density calculation is ensured. Secondly, the combination of Euclidean distance and truncation distance with the statistical count of the first point objectively reflects the degree of local clustering of data points. Thirdly, by selecting the minimum Euclidean distance of high local density data points as the distance parameter, abnormal isolated points in the feature space are effectively identified. Finally, the threshold is determined using the density peak points of the decision graph, enabling the threshold to adapt to the data distribution pattern and improving the objectivity and accuracy of anomaly detection.

[0140] Example 7

[0141] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data. When in the first maintenance state, maintenance and optimization operations are performed on the coal seam gas pressure measurement data based on the data maintenance state, including:

[0142] Based on data from one or more monitoring points that are spatially adjacent to the abnormal data and are in the second maintenance state, spatial interpolation correction is performed on the abnormal data;

[0143] Alternatively, based on a historical data sequence that is continuous in time, the abnormal data can be fitted and corrected using a time trend.

[0144] Based on the corrected abnormal data, an early warning message is generated that includes the sensor number, abnormal parameter items, and recommended maintenance measures.

[0145] In this embodiment, interpolation correction of abnormal data refers to the process of filling or correcting abnormal values ​​by using the spatial correlation of reliable data points around the abnormal data through mathematical interpolation algorithms.

[0146] Specifically, based on the geological parameters of the coal seam, data from 3-5 monitoring points in the second maintenance state around the abnormal data point are selected. Inverse distance weighted interpolation or Kriging interpolation is used to calculate the reasonable predicted value of the abnormal point, and the original abnormal data is replaced to ensure that the corrected data conforms to the spatial distribution pattern.

[0147] For example, if a monitoring point displays a gas pressure of 0 due to a sensor malfunction, an inverse distance weighted interpolation is performed using the pressure values ​​of 0.8MPa, 1.0MPa, 0.9MPa, and 1.1MPa from four normal monitoring points around it, to obtain a corrected value of 0.95MPa.

[0148] Trend fitting correction of abnormal data based on historical data sequences from the same collection location refers to a correction method that uses the temporal evolution pattern of historical data from the monitoring point to infer reasonable values ​​at abnormal times through trend models.

[0149] Specifically, gas pressure data for the past 30 days at this location is extracted, a time series model is constructed, the trend of data change over time is analyzed, the theoretical value of abnormal moments is predicted based on the model, the original abnormal data is replaced, and the corrected data is ensured to conform to the continuity of the time dimension.

[0150] For example, a monitoring point experienced a sudden increase in pressure due to signal interference during a rainstorm. Based on the historical data showing a daily average decrease of 0.02 MPa, a reasonable value for that moment was estimated to be 0.78 MPa through linear fitting, and the original outlier value was corrected by 1.5 MPa.

[0151] The beneficial effects of the above technical solution are as follows: For abnormal data in the first maintenance state, interpolation correction and trend fitting correction are used to accurately repair abnormal values, ensuring the spatiotemporal consistency of data. Simultaneously, early warnings containing sensor information, anomalies, and maintenance suggestions are generated. This improves data reliability to support safety decisions and reduces manual investigation costs through targeted early warnings, forming a closed-loop management system for abnormal data and enhancing the practicality and safety of the coal mine gas monitoring system.

[0152] Example 8

[0153] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data. When in the second maintenance state, maintenance and optimization operations are performed on the coal seam gas pressure measurement data based on the data maintenance state, including:

[0154] The coal seam gas pressure measurement data were filtered using a sliding window smoothing algorithm to eliminate random noise;

[0155] Alternatively, a recursive filtering algorithm can be used to optimize the state estimation of the measured data in order to correct systematic errors.

[0156] In this embodiment, the preset sliding window algorithm refers to an algorithm that smooths normal data within a window by setting a fixed time or data volume window to eliminate random noise.

[0157] Specifically, a continuous time window or data volume window is defined, and the mean, median or weighted average of the gas pressure data in the second maintenance state within the window is calculated. This statistical value is then used to replace the original data within the window, filtering out short-term random fluctuations and preserving the overall trend of the data.

[0158] For example, a 10-minute sliding window is set, and the average value of 0.845 MPa is calculated for six consecutive normal data points of 0.82, 0.85, 0.83, 0.86, 0.84, and 0.87 MPa at a certain monitoring point, which is used as the optimized value of the data within the window.

[0159] In this embodiment, the Kalman filter algorithm refers to a recursive filtering algorithm that optimizes data accuracy through a prediction-update iterative process based on the dynamic system state equation and observation equation. Specifically, for gas pressure data in the second maintenance state, the state equation is first established based on historical data, the theoretical value of the current state is estimated through the prediction step, and then the actual measured value is combined for updating and correction to reduce the impact of system error and observation noise.

[0160] For example, for normal pressure data at a certain monitoring point, the Kalman filter first predicts the theoretical pressure value at time t as 0.9 MPa. Combined with the measured value of 0.92 MPa, which includes ±0.01 MPa of noise, the optimized value of 0.91 MPa is obtained through iterative calculation, thus balancing the predicted trend with the measured data.

[0161] In this embodiment, state estimation optimization refers to the process of using algorithms to make optimal predictions about the true state of normal data, thereby reducing measurement errors and system interference and improving data stability.

[0162] Specifically, to address the minor fluctuations in the second maintenance state data, a mathematical model is constructed based on the time-series characteristics of the data to estimate the theoretical value of the data under ideal conditions, making the optimized data closer to the real gas pressure state while retaining reasonable trend changes.

[0163] For example, normal data at a certain monitoring point may have high-frequency noise of ±0.03MPa due to power grid fluctuations. After optimization through state estimation, the data fluctuation range is reduced to ±0.01MPa, which not only eliminates the noise but also maintains the downward trend of 0.02MPa per day.

[0164] The beneficial effects of the above technical solutions are as follows: For normal data in the second maintenance state, the short-term random noise is smoothed by a preset sliding window algorithm to preserve the overall trend of the data, or the state estimation is dynamically optimized by a Kalman filter algorithm to reduce system errors. Both of these methods improve the stability and accuracy of the data, avoid the interference of subtle fluctuations in normal data with the analysis, provide high-quality basic data for the long-term trend judgment of gas pressure and the training of early warning models, and enhance the accuracy and foresight of coal mine gas control decisions.

[0165] Example 9

[0166] To address the problem that existing data maintenance strategies are static and unable to adapt to dynamic changes in operating conditions, this invention creatively proposes a data quality lifecycle assessment and feedback mechanism. The core idea of ​​this mechanism is that a data maintenance system should not be static but should possess the ability to "learn" and "evolve." By continuously assessing data quality and feeding the assessment results back to dynamically adjust core judgment parameters (i.e., initial judgment thresholds), the entire maintenance system achieves intelligent self-adaptation.

[0167] This embodiment provides a method for maintaining and optimizing coal seam gas pressure measurement data, and also includes:

[0168] A data quality lifecycle assessment model is constructed. The inputs of the data quality lifecycle assessment model include the original data and maintenance data of coal seam gas pressure measurement data and data correlation parameters under maintenance status. The output is the credibility level of coal seam gas pressure measurement data and data correlation parameters.

[0169] The initial judgment threshold is dynamically adjusted based on the credibility level.

[0170] The dynamic adjustment of the initial judgment threshold based on the credibility level specifically includes: when the credibility level of a monitoring point is excellent for three consecutive months, the initial judgment threshold of the monitoring point is relaxed by 10%-15%; when the credibility level is poor, the initial judgment threshold of the monitoring point is tightened by 5%-8%.

[0171] In this embodiment, a data quality lifecycle assessment model is constructed, including: Input data:

[0172] Coal seam gas pressure measurement data under maintenance status; original data and maintenance data of corresponding data association parameters. Output indicators: L1: Reliability level of coal seam gas pressure measurement data; L2: Reliability level of data association parameters; reliability levels are uniformly divided into four levels: excellent, good, average, and poor. For example,

[0173] ;

[0174] in, This is an adjustment factor for coal seam gas pressure measurement data.

[0175] For example, the specific parameter values ​​could be: =1.10 is the improvement coefficient corresponding to the excellent level. =0.95 is the improvement coefficient corresponding to the poor level, and n and m are the number of consecutive months to maintain the excellent and poor levels corresponding to the adjustment coefficient of coal seam gas pressure measurement data.

[0176] For example,

[0177] ;

[0178] in, It is the adjustment coefficient for data association parameters; p represents The number of consecutive months maintained at the "Excellent" level; q represents The number of consecutive months maintained at the "poor" level;

[0179] For example, the specific parameter values ​​could be: =1.15, =0.92, which are the basic coefficients corresponding to the excellent and poor performance of the coal seam gas pressure measurement data adjustment coefficients.

[0180] Specifically, the final threshold calculation formula is as follows:

[0181] ;

[0182] in, This is the adjusted initial judgment threshold. This is the initial judgment threshold, the basic threshold before confidence adjustment; It is a weighting coefficient, representing the degree of importance attached to the reliability of coal seam gas pressure measurement data, and .

[0183] The beneficial effects of the above technical solution are as follows: By constructing a data quality lifecycle assessment model, the reliability level of pressure data and related parameters is output, thereby dynamically adjusting the initial judgment threshold. This allows the threshold to adapt to changes in data quality; when the reliability is high, it can be appropriately relaxed to reduce misjudgments, and when the reliability is low, it can be tightened to improve accuracy. This enhances the timeliness and adaptability of the threshold, provides a reliable benchmark for determining the data maintenance status, and improves the scientific nature and dynamic adaptability of gas pressure data management.

[0184] Example 10

[0185] This embodiment provides a coal seam gas pressure measurement data maintenance and optimization system, including:

[0186] Data acquisition module: Acquires coal seam gas pressure measurement data and data association parameters, including geological parameters of the data acquisition location, acquisition timestamp, and sensor operating status;

[0187] Status acquisition module: Determines the data maintenance status by performing multi-dimensional joint analysis of coal seam gas pressure measurement data and data correlation parameters;

[0188] Data optimization module: Performs maintenance and optimization operations on coal seam gas pressure measurement data based on data maintenance status.

[0189] The beneficial effects of the above technical solution are as follows: by integrating coal seam gas pressure data with geological parameters, sensor status and other multi-dimensional related parameters, the accuracy of abnormal data identification is improved and false positives and false negatives are reduced; by combining dynamic threshold adjustment and spatiotemporal correlation verification, the abnormal data can be accurately repaired and the data credibility is improved; and by constructing a full life cycle assessment and closed-loop feedback mechanism, the maintenance strategy can adapt to geological changes and data quality fluctuations, enhance the scientific nature of coal mine gas management decisions, and provide reliable data support for safe production.

[0190] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer programs or instructions. When a computer program or instruction is loaded and executed on a computer, the processes or functions of the embodiments of this application are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable module. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, a computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); or it can be a semiconductor medium, such as a solid-state drive (SSD). The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for maintaining and optimizing coal seam gas pressure measurement data, characterized in that, The method includes: The data on coal seam gas pressure measurement is obtained, and the data association parameters are obtained. The data association parameters include the geological parameters of the data acquisition location, the acquisition timestamp, and the sensor working status. By performing multi-dimensional joint analysis of the coal seam gas pressure measurement data and the data correlation parameters, the data maintenance status is determined; the data maintenance status includes: a first maintenance status and a second maintenance status; the first maintenance status is an abnormal data status, and the second maintenance status is a normal data status; wherein, determining the data maintenance status includes: obtaining an initial judgment threshold; The coal seam gas pressure measurement data is compared with the initial judgment threshold for the first analysis. The execution status of the second analysis is determined based on the results of the first analysis; The data association parameters are analyzed a second time based on the execution status of the second analysis; The first maintenance state and the second maintenance state are determined based on the results of the first analysis and the results of the second analysis. Based on the aforementioned data maintenance status, maintenance and optimization operations are performed on the coal seam gas pressure measurement data.

2. The method according to claim 1, characterized in that, The acquisition of coal seam gas pressure measurement data and the acquisition of data correlation parameters include: Real-time gas pressure values ​​at different monitoring points in the coal seam are obtained to obtain coal seam gas pressure measurement data; The coal seam thickness, fracture development density, and distance from the fault corresponding to each monitoring point are obtained as geological parameters. The sensor's built-in clock records the data acquisition time and generates a timestamp. The sensor's operating status is determined by acquiring the power supply voltage, sampling frequency, and calibration records through the sensor's self-test module. By integrating the coal seam gas pressure measurement data, geological parameters, timestamps, and sensor operating status, the coal seam gas pressure measurement data to be processed and the data correlation parameters are obtained.

3. The method according to claim 1, characterized in that, The process of obtaining the initial determination threshold includes: Based on the coal seam gas pressure measurement data, a pressure value sequence matrix is ​​constructed, and a first feature matrix is ​​generated by combining the geological parameters, timestamps and sensor working status in the data association parameters. The first feature matrix is ​​normalized to obtain a standardized feature matrix; Based on the standardized feature matrix, determine the local density and distance parameters of each data point in the standardized feature matrix; The initial anomaly detection threshold is determined based on the local density and distance parameters of each data point.

4. The method according to claim 3, characterized in that, The step of determining the local density and distance parameters of each data point in the standardized feature matrix based on the standardized feature matrix includes: Obtain the cutoff distance parameter; The Euclidean distance between each first data point and all other data points is determined based on the cutoff distance parameter. The first number of points is determined based on the Euclidean distance and the cutoff distance, and the first number of points is used as the local density of the first data points. For each first data point, filter out all data points with a local density greater than the first data point, and take the minimum Euclidean distance among them as the distance parameter of the first data point; A decision graph is constructed based on the local density and the distance parameter, and the parameter value corresponding to the density peak point in the decision graph is taken as the initial anomaly judgment threshold.

5. The method according to claim 1, characterized in that, When in the first maintenance state, the maintenance optimization operation performed on the coal seam gas pressure measurement data based on the data maintenance state includes: Based on data from one or more monitoring points that are spatially adjacent to the abnormal data and are in the second maintenance state, spatial interpolation correction is performed on the abnormal data; Alternatively, based on a historical data sequence that is continuous in time, the abnormal data can be fitted and corrected using a time trend. Based on the corrected abnormal data, an early warning message is generated that includes the sensor number, abnormal parameter items, and recommended maintenance measures.

6. The method according to claim 1, characterized in that, When in the second maintenance state, the maintenance optimization operation performed on the coal seam gas pressure measurement data based on the data maintenance state includes: The coal seam gas pressure measurement data were filtered using a sliding window smoothing algorithm to eliminate random noise; Alternatively, a recursive filtering algorithm can be used to optimize the state estimation of the measured data in order to correct systematic errors.

7. The method according to claim 1 or 3, characterized in that, Also includes: A data quality lifecycle assessment model is constructed. The inputs of the data quality lifecycle assessment model include the original data and maintenance data of the coal seam gas pressure measurement data and the data correlation parameters under the maintenance state. The output is the reliability level of the coal seam gas pressure measurement data and the data correlation parameters. The initial judgment threshold is dynamically adjusted based on the aforementioned credibility level.

8. A coal seam gas pressure measurement data maintenance and optimization system, characterized in that, The system includes: Data acquisition module: acquires coal seam gas pressure measurement data and acquires data association parameters, including geological parameters of the data acquisition location, acquisition timestamp, and sensor working status; Status acquisition module: This module determines the data maintenance status by performing multi-dimensional joint analysis of the coal seam gas pressure measurement data and the data association parameters. The data maintenance status includes: a first maintenance status and a second maintenance status; the first maintenance status is an abnormal data status, and the second maintenance status is a normal data status. Determining the data maintenance status includes: acquiring an initial judgment threshold; performing a first analysis on the coal seam gas pressure measurement data and the initial judgment threshold; determining the execution status of a second analysis based on the results of the first analysis; performing a second analysis on the data association parameters based on the execution status of the second analysis; and determining the first maintenance status and the second maintenance status based on the results of the first and second analyses. Data optimization module: Performs maintenance and optimization operations on the coal seam gas pressure measurement data based on the data maintenance status.

Images