Method and system for judging illegal production operation of coal mine enterprise based on machine learning

Abstract

The application discloses a kind of coal mine enterprise illegal production operation judgment method and system based on machine learning, it is related to electric power operation management technical field, including the following steps, based on enterprise historical electric power monitoring data and historical coal mine enterprise illegal production judgment result, establish enterprise power consumption anomaly identification model;From the result data of enterprise electric power consumption classification estimation calculation the power consumption data of each coal mine area obtained is input into enterprise power consumption anomaly identification model, determines power consumption anomaly area and abnormal power consumption;The abnormal power consumption of suspected mining operation point and personnel change feature are associated and analyzed, to determine whether suspected mining operation point is mining operation point.The present application carries out clustering analysis to power consumption anomaly area by personnel feature data, the abnormal power consumption of suspected mining operation point and personnel change feature are associated and analyzed, to determine whether suspected mining operation point is mining operation point, realize automatic identification and judge illegal behavior in coal mine production operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power operation management, and in particular to a coal mine enterprise illegal production operation judgment method and system based on machine learning. BACKGROUND

[0002] The hazards of illegal production of coal mine enterprises mainly include insufficient investment in safety facilities, electrical fire and electric shock hazards, and increased risk of major accidents in coal mines.

[0003] Insufficient investment in safety facilities: when coal mine enterprises face losses, they may reduce investment in safety facilities, resulting in necessary safety facilities not keeping up, thereby affecting safety production.

[0004] Electrical fire and electric shock hazards: if the electrical equipment used in the production system and auxiliary system of the coal mine is operated for a long time, it may generate a large amount of heat, causing the internal insulation of the electrical equipment to be damaged, the protection monitoring device to fail, and further causing electrical hazards such as fire, explosion, etc.

[0005] Increased risk of major accidents in coal mines: over-capacity production is a major risk factor for major accidents in coal mines. Overload production may cause problems in the safety monitoring system of the mine, such as not installing a safety monitoring system, a personnel location monitoring system or the system not functioning properly, and modifying, deleting and shielding system data, which may increase the risk of major accidents in coal mines.

[0006] In summary, illegal production of coal mine enterprises not only may result in insufficient investment in safety facilities, but also may cause electrical fire and electric shock hazards, and increase the risk of major accidents in coal mines, which is extremely detrimental to safety production, so it is particularly important to judge illegal production of coal mine enterprises. SUMMARY

[0007] In view of the problems existing in the prior art, the present application is proposed.

[0008] To solve the above technical problems, the present application provides the following technical solutions:

[0009] In a first aspect, the present application provides a coal mine enterprise illegal production operation judgment method based on machine learning, which includes the following steps:

[0010] Based on the historical electric power monitoring data of the enterprise and the historical coal mine enterprise illegal production judgment results, an enterprise power consumption anomaly identification model is established using machine learning.

[0011] The power consumption data of each coal mine area obtained from the result data calculated from the enterprise power consumption classification estimation is input into the enterprise power consumption anomaly recognition model to determine the power consumption anomaly area and the abnormal power consumption;

[0012] Personnel characteristic data of the power consumption anomaly area is obtained, and the power consumption anomaly area is analyzed based on the personnel characteristic data to obtain a suspected mining operation point;

[0013] The abnormal power consumption of the suspected mining operation point and the personnel change characteristics are associated and analyzed to determine whether the suspected mining operation point is a mining operation point.

[0014] As a preferred scheme of the coal mine enterprise illegal production operation judgment method based on machine learning, after determining that the mining operation point is a mining operation point, the method further includes,

[0015] When the abnormal power consumption of the mining operation point is greater than the first-level power consumption, an over-intensity illegal production warning information is generated;

[0016] When the abnormal power consumption of the mining operation point is less than the first-level power consumption and greater than the second-level power consumption, and the number of personnel in the mining operation point exceeds 20% of the preset number of personnel, an over-intensity illegal production warning information is generated.

[0017] As a preferred scheme of the coal mine enterprise illegal production operation judgment method based on machine learning, the step of establishing the enterprise power consumption anomaly recognition model includes,

[0018] A corresponding relationship between the enterprise historical power monitoring data and the historical coal mine enterprise illegal production judgment result is established;

[0019] The enterprise historical power monitoring data and the historical coal mine enterprise illegal production judgment result are taken as training data based on the corresponding relationship, and an initial recognition model based on machine learning is trained to obtain a training result;

[0020] Based on the coal mine enterprise illegal production operation judgment standard, a model attribute evaluation index is determined, and when the training result meets the model attribute evaluation index, an enterprise power consumption anomaly recognition model is obtained.

[0021] As a preferred scheme of the coal mine enterprise illegal production operation judgment method based on machine learning, the step of determining the power consumption anomaly area and the abnormal power consumption includes,

[0022] The power consumption data of each coal mine area is determined through the result data, and the power consumption data is processed based on a time dimension to obtain a plurality of input data;

[0023] The input data is sequentially input into the enterprise power consumption anomaly identification model to obtain a power consumption identification result of the coal mine area;

[0024] Based on the power consumption identification result, an abnormal power consumption area and an abnormal power consumption amount are obtained.

[0025] As a preferred scheme of the coal mine enterprise illegal production operation judgment method based on machine learning, the method for obtaining the suspected mining operation point comprises,

[0026] According to the personnel positioning data, the personnel quantity data and the personnel type data in the personnel feature data;

[0027] The personnel area density is determined through the personnel positioning data and the personnel quantity data, the power consumption abnormal area is divided through the personnel area density, a plurality of density areas are obtained, and a target density area meeting a preset density is obtained from the plurality of density areas;

[0028] A density distribution sequence of the density size of the target density area is obtained, and the number of clusters and the cluster center are determined based on the density distribution sequence;

[0029] The region name information of each target density area is classified to obtain a plurality of name types, the personnel type data is classified to obtain a plurality of type types, and an evaluation index system is established based on the name type and the type type in combination with historical coal mine enterprise operation data;

[0030] The target density area is analyzed by clustering according to the number of clusters and the cluster center through a preset clustering algorithm in combination with the evaluation index system, and a region clustering result is obtained;

[0031] The average feature of each group of target density areas is obtained from the region clustering result, and the target density area in the target density area group meeting a preset operation feature is taken as a suspected mining operation point;

[0032] The determination method of the number of clusters and the cluster center comprises:

[0033] The center point, the mean point and the median point of the density distribution sequence are obtained;

[0034] The number of clusters is determined through the center point and the length of the density distribution sequence;

[0035] The average value of the mean point and the median point is taken as a cluster center, and the density distribution sequence is divided based on the number of clusters and the length of the density distribution sequence, and other cluster centers are obtained based on the division points.

[0036] As a preferred embodiment of the machine learning-based method for judging illegal production operations in coal mines according to the present invention, the method for judging whether a suspected mining operation point is indeed a mining operation point includes:

[0037] The abnormal power consumption at the mining operation point is analyzed along the time dimension to obtain the abnormal power consumption curve, and the fluctuation points and fluctuation amplitude are obtained from the abnormal power consumption curve.

[0038] Multiple sets of change characteristics are obtained by analyzing the change characteristics of personnel according to the change attributes, and each set of change characteristics is analyzed according to the time dimension to obtain the attribute change curve corresponding to the change attribute. The attribute fluctuation amplitude at the same time point as the fluctuation point is obtained from the attribute change curve.

[0039] Based on the degree of influence of the changing attributes on mining operations, the curve weights of the attribute change curves are determined, and the attribute fluctuation amplitude is weighted and analyzed based on the curve weights to obtain the target attribute fluctuation amplitude.

[0040] Obtain the correlation between the fluctuation range of target attributes and the fluctuation range. Based on the correlation of the fluctuation range of all target attributes, determine the comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining operation points.

[0041] Based on the magnitude of abnormal power consumption at suspected mining sites, a first score is determined. The average characteristics of personnel changes at suspected mining sites are obtained. Based on the characteristic information of the average characteristics, a second score is determined. Based on the comprehensive correlation between abnormal power consumption and personnel changes at suspected mining sites, a third score is determined.

[0042] Based on the first score, the second score and the third score, a comprehensive score is obtained for the suspected mining operation site. The suspected mining operation site with a comprehensive score greater than the preset score is selected as the mining operation site.

[0043] The method for determining the comprehensive correlation between abnormal power consumption and personnel change characteristics at suspected mining sites includes:

[0044] By summing the correlations of the fluctuation ranges of the target attributes, a comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining sites is obtained.

[0045] As a preferred embodiment of the machine learning-based method for judging illegal production operations in coal mines according to the present invention, the method further includes: real-time adjustment of the enterprise's abnormal power consumption identification model, specifically including the following steps:

[0046] Real-time change data of coal mine operations is obtained, and the real-time change data is input into the enterprise power consumption anomaly identification model to obtain the first result;

[0047] The first result is used as a parameter in the next real-time change data and input into the enterprise power consumption anomaly identification model to obtain the second result;

[0048] Based on the difference between the changing trends of the first and second results and the actual changing trends, the model parameters are adjusted in real time.

[0049] The steps for real-time adjustment of model parameters include:

[0050] Based on the difference between the changing trends of the first and second results and the actual changing trends, the adjustment range of the model parameters is determined, and the corrected changing trend of the changing trends of the first and second results under the adjustment range is determined.

[0051] Based on the difference between the corrected trend and the actual trend, the model parameters are iteratively adjusted until the difference between the latest corrected trend and the actual trend meets the difference requirement.

[0052] Secondly, embodiments of the present invention provide a machine learning-based system for judging illegal production operations in coal mines, which includes a data collection and preprocessing module, a feature engineering module, an anomaly detection module, and a correlation analysis module.

[0053] The data collection and preprocessing module is used to collect historical power monitoring data and historical judgment results of illegal production by coal mining enterprises, and to clean the data and process missing and outlier values.

[0054] The feature engineering module is used to extract features from electricity consumption data that help identify illegal production.

[0055] The anomaly detection module is used to analyze power consumption data in real time or in batches to identify abnormal patterns.

[0056] The correlation analysis module is used to perform correlation analysis between abnormal power consumption at suspected mining sites and personnel change characteristics to determine whether these sites are indeed carrying out mining operations.

[0057] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the above-mentioned machine learning-based method for judging illegal production operations in coal mines.

[0058] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the above-described machine learning-based method for judging illegal production operations in coal mines.

[0059] The beneficial effects of this invention are as follows: By using machine learning, an abnormal power consumption identification model is established based on historical power monitoring data and historical judgment results of illegal production in coal mines. Power consumption data of various coal mine areas obtained from the predicted power consumption classification data of enterprises are input into the abnormal power consumption identification model to determine the abnormal power consumption areas and abnormal power consumption, providing a basis for further identification of illegal production operation points. By obtaining personnel characteristic data of abnormal power consumption areas, cluster analysis is performed on the abnormal power consumption areas based on personnel characteristic data to obtain suspected mining operation points. Correlation analysis is performed on the abnormal power consumption and personnel change characteristics of suspected mining operation points to determine whether the suspected mining operation point is a mining operation point. This achieves automatic identification and judgment of illegal behavior in coal mine production operations, and realizes accurate judgment of illegal production operations in coal mine enterprises. Attached Figure Description

[0060] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0061] Figure 1 This is a flowchart of a machine learning-based method for judging illegal production operations in coal mines. Detailed Implementation

[0062] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0063] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0064] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0065] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0066] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0067] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0068] Example 1

[0069] Reference Figure 1 This is the first embodiment of the present invention, which provides a machine learning-based method for judging illegal production operations in coal mines, including the following steps:

[0070] S1. Based on historical power monitoring data of enterprises and judgment results of illegal production of coal mining enterprises in the past, an abnormal power consumption identification model of enterprises is established by using machine learning.

[0071] The steps for establishing the enterprise power consumption anomaly identification model include:

[0072] Establish a correlation between historical power monitoring data of enterprises and historical judgment results of illegal production by coal mining enterprises;

[0073] Using the aforementioned correspondence as a standard, the historical power monitoring data of enterprises and the historical judgment results of illegal production by coal mining enterprises are used as training data to train an initial identification model determined by machine learning, and the training results are obtained.

[0074] Based on the judgment criteria for illegal production operations of coal mining enterprises, the model attribute evaluation index is determined, and when the training results meet the model attribute evaluation index, the abnormal power consumption identification model of enterprises is obtained.

[0075] Among them, the evaluation indicators for model attributes include accuracy, robustness, and overfitting;

[0076] By establishing a correspondence between historical power monitoring data of enterprises and historical judgment results of illegal production of coal mine enterprises, and using the correspondence as a standard, the historical power monitoring data of enterprises and historical judgment results of illegal production of coal mine enterprises are used as training data to train an initial identification model determined by machine learning, and the training results are obtained. Based on the judgment criteria for illegal production operations of coal mine enterprises, the model attribute evaluation index is determined. When the training results meet the model attribute evaluation index, the abnormal power consumption identification model of enterprises is obtained, which provides a basis for further identification of illegal production operation points.

[0077] S2. Input the electricity consumption data of each coal mine area obtained from the enterprise electricity consumption classification and prediction results into the enterprise electricity consumption anomaly identification model to determine the electricity consumption anomaly area and the abnormal electricity consumption.

[0078] The steps for determining the abnormal power consumption area and the abnormal power consumption include,

[0079] The power consumption data of each coal mining area is determined by the result data, and the power consumption data is processed based on the time dimension to obtain multiple sets of input data;

[0080] The input data is sequentially input into the enterprise power consumption anomaly identification model to obtain the power consumption identification results for the coal mine area;

[0081] Based on the power consumption identification results, abnormal power consumption areas and abnormal power consumption are obtained;

[0082] Specifically, the power consumption data is processed based on the time dimension by dividing it into multiple sets of input data according to a period of time.

[0083] The power consumption identification results for coal mining areas include two types: one is abnormal power consumption areas and abnormal power consumption, and the other is normal power consumption areas and normal power consumption.

[0084] By determining the power consumption data of each coal mine area based on the result data, and processing the power consumption data based on the time dimension, multiple sets of input data are obtained. The input data are then sequentially input into the enterprise power consumption anomaly identification model to obtain the power consumption identification results of the coal mine area. Based on the power consumption identification results, the power consumption anomaly area and the abnormal power consumption are obtained, providing a basis for further identification of illegal production operation points.

[0085] S3. Obtain personnel characteristic data in areas with abnormal power consumption, and perform cluster analysis on areas with abnormal power consumption based on personnel characteristic data to obtain suspected mining operation points.

[0086] The methods for obtaining the suspected mining operation site include:

[0087] Based on the personnel characteristic data, determine the personnel location data, personnel quantity data, and personnel type data;

[0088] By using personnel location data and personnel quantity data, the personnel area density is determined. The abnormal power consumption area is divided into multiple density areas based on the personnel area density. From the multiple density areas, a target density area whose density meets the preset density is obtained.

[0089] Obtain the density distribution sequence of the density magnitude of the target density region, and determine the number of clusters and cluster centers based on the density distribution sequence;

[0090] The regional name information of each target density area is classified to obtain multiple name types, and the personnel type data is classified to obtain multiple job types. Based on the name type and job type, combined with historical coal mine enterprise operation data, an evaluation index system is established.

[0091] By using a preset clustering algorithm and combining it with an evaluation index system, cluster analysis is performed on the target density region according to the number of clusters and the cluster centers to obtain the region clustering results;

[0092] The average features of each target density region are obtained from the regional clustering results. The target density regions in the target density region group whose average features meet the preset operation features are taken as suspected mining operation points.

[0093] Name types include, for example, well access type, transportation type, etc.

[0094] The evaluation index system evaluates the characteristics of an area based on its name type and job type, specifically the probability of it being a mining operation site.

[0095] By determining personnel location data, personnel quantity data, and personnel type data from the personnel characteristic data, and based on the personnel location data and personnel quantity data, the personnel area density is determined. Based on the personnel area density, abnormal power consumption areas are divided into multiple density areas. From these multiple density areas, target density areas whose density meets the preset density are obtained. The density distribution sequence of the target density areas is obtained, and based on the density distribution sequence, the number of clusters and cluster centers are determined. The area name information of each target density area is classified to obtain multiple name types. The personnel type data is classified to obtain multiple job types. Based on the name types and job types, combined with historical coal mine operation data, an evaluation index system is established. Based on the preset clustering algorithm and combined with the evaluation index system, the target density areas are clustered according to the number of clusters and cluster centers to obtain regional clustering results. The average characteristics of each group of target density areas are obtained from the regional clustering results. The target density areas in the target density area group whose average characteristics meet the preset operation characteristics are identified as suspected mining operation points, and violations in coal mine production operations are identified and judged, achieving accurate judgment of violations in coal mine production operations.

[0096] The methods for determining the number of clusters and cluster centers include:

[0097] Obtain the center point, mean point, and median point of the density distribution sequence;

[0098] The number of clusters is determined by the center point and the length of the density distribution sequence;

[0099] Then, the average of the mean point and the median point is used as a cluster center, and the density distribution sequence is divided based on the number of clusters and the length of the density distribution sequence, and other cluster centers are obtained based on the division points;

[0100] By obtaining the center point, mean point, and median point of the density distribution sequence, the number of clusters is determined based on the center point and the length of the density distribution sequence. The average value of the mean point and the median point is used as a cluster center. Based on the number of clusters and the length of the density distribution sequence, the density distribution sequence is divided. Other cluster centers are obtained based on the division points, ensuring the accuracy of the obtained cluster centers and providing a basis for cluster analysis.

[0101] S4. Conduct correlation analysis between the abnormal power consumption and personnel change characteristics of suspected mining sites to determine whether the suspected mining site is indeed a mining site.

[0102] 6. The method for determining whether a suspected mining operation point is indeed a mining operation point includes,

[0103] The abnormal power consumption at the mining operation point is analyzed along the time dimension to obtain the abnormal power consumption curve, and the fluctuation points and fluctuation amplitude are obtained from the abnormal power consumption curve.

[0104] Multiple sets of change characteristics are obtained by analyzing the change characteristics of personnel according to the change attributes, and each set of change characteristics is analyzed according to the time dimension to obtain the attribute change curve corresponding to the change attribute. The attribute fluctuation amplitude at the same time point as the fluctuation point is obtained from the attribute change curve.

[0105] Based on the degree of influence of the changing attributes on mining operations, the curve weights of the attribute change curves are determined, and the attribute fluctuation amplitude is weighted and analyzed based on the curve weights to obtain the target attribute fluctuation amplitude.

[0106] Obtain the correlation between the fluctuation range of target attributes and the fluctuation range. Based on the correlation of the fluctuation range of all target attributes, determine the comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining operation points.

[0107] Based on the magnitude of abnormal power consumption at suspected mining sites, a first score is determined. The average characteristics of personnel changes at suspected mining sites are obtained. Based on the characteristic information of the average characteristics, a second score is determined. Based on the comprehensive correlation between abnormal power consumption and personnel changes at suspected mining sites, a third score is determined.

[0108] Based on the first score, the second score and the third score, a comprehensive score is obtained for the suspected mining operation site. The suspected mining operation site with a comprehensive score greater than the preset score is selected as the mining operation site.

[0109] The variable attributes include quantity, location, and job type;

[0110] A first score is determined based on the magnitude of abnormal power consumption at suspected mining sites. The average characteristics of personnel changes at these sites are then obtained. A second score is determined based on these average characteristics. A third score is determined based on the comprehensive correlation between abnormal power consumption and personnel changes at the suspected mining sites. A comprehensive score for each suspected mining site is obtained based on the first, second, and third scores. Sites with a comprehensive score greater than a preset score are selected as mining sites. The determination of mining sites is based on three aspects: power consumption, personnel changes, and the correlation between the two. This ensures the accuracy of the obtained mining sites and enables precise judgment of illegal production operations by coal mining enterprises.

[0111] The method for determining the comprehensive correlation between abnormal power consumption and personnel change characteristics at suspected mining sites includes:

[0112] The correlation between the fluctuation range of the target attributes is accumulated to obtain the comprehensive correlation between the abnormal power consumption and personnel change characteristics of the suspected mining operation point;

[0113] By summing the correlations of the fluctuation ranges of the target attributes, a comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining operation points is obtained. This ensures that the comprehensive correlation takes into account the fluctuation ranges of all target attributes, thus guaranteeing accuracy.

[0114] After determining the mining operation point as the mining operation point, the process also includes,

[0115] When the abnormal power consumption at the mining operation site exceeds the first-level power consumption, an over-intensity violation production warning is generated.

[0116] When the abnormal power consumption at the collection point is less than the first-level power consumption but greater than the second-level power consumption, and the number of workers at the mining point exceeds 20% of the preset number, an over-intensity violation production warning message is generated.

[0117] Among them, the power consumption of Level 1 is greater than that of Level 2;

[0118] When the abnormal power consumption at the mining operation point exceeds the first-level power consumption, an over-intensity violation production warning is generated. When the abnormal power consumption at the mining operation point is less than the first-level power consumption but greater than the second-level power consumption, and the number of workers at the mining operation point exceeds 20% of the preset number, an over-intensity violation production warning is generated. This makes it easier for staff to detect over-intensity violation production in a timely manner and intervene promptly.

[0119] In this embodiment, the enterprise power consumption anomaly identification model is used to obtain abnormal power monitoring data from power monitoring data.

[0120] This also includes real-time adjustments to the enterprise power consumption anomaly identification model, with specific steps including...

[0121] Real-time change data of coal mine operations is obtained, and the real-time change data is input into the enterprise power consumption anomaly identification model to obtain the first result;

[0122] The first result is used as a parameter in the next real-time change data and input into the enterprise power consumption anomaly identification model to obtain the second result;

[0123] Based on the difference between the changing trends of the first and second results and the actual changing trends, the model parameters are adjusted in real time.

[0124] Model parameters include, for example, the parameters of convolutional layers, fully connected layers, and data normalization layers;

[0125] By acquiring real-time change data of coal mine operations, the real-time change data is input into the enterprise power consumption anomaly identification model to obtain a first result. The first result is then used as a parameter in the next real-time change data input into the enterprise power consumption anomaly identification model to obtain a second result. Based on the difference between the change trends of the first and second results and the actual change trends, the model parameters are adjusted in real time. Through batch parameter replacement and optimization, the adaptability of the model is improved.

[0126] The steps for real-time adjustment of model parameters include:

[0127] Based on the difference between the changing trends of the first and second results and the actual changing trends, the adjustment range of the model parameters is determined, and the corrected changing trend of the changing trends of the first and second results under the adjustment range is determined.

[0128] Based on the difference between the corrected trend and the actual trend, the model parameters are iteratively adjusted until the difference between the latest corrected trend and the actual trend meets the difference requirement.

[0129] By determining the adjustment range of the model parameters based on the difference between the changing trends of the first and second results and the actual changing trends, and determining the corrected changing trend of the changing trends of the first and second results under the adjustment range, the model parameters are iteratively adjusted based on the difference between the corrected changing trend and the actual changing trend until the difference between the latest corrected changing trend and the actual changing trend meets the difference requirement, thus improving the pillow and model correction effect.

[0130] In summary, by utilizing machine learning, a model for identifying abnormal power consumption in coal mines is established based on historical power monitoring data and historical judgments of illegal production practices. Power consumption data for each coal mine area, obtained from the estimated power consumption classification data, is input into this model to identify abnormal power consumption areas and amounts, providing a foundation for further identification of illegal production sites. By acquiring personnel characteristic data from these abnormal power consumption areas, cluster analysis is performed to identify suspected mining sites. Correlation analysis is then conducted between the abnormal power consumption and personnel change characteristics of these suspected mining sites to determine whether they are indeed mining sites. This approach enables automatic identification and judgment of illegal activities in coal mine production, achieving accurate assessment of illegal production practices in coal mines.

[0131] Example 2

[0132] Based on the first embodiment, this embodiment further provides a machine learning-based system for judging illegal production operations in coal mines, including 8. a data collection and preprocessing module, a feature engineering module, an anomaly detection module, and a correlation analysis module;

[0133] The data collection and preprocessing module is used to collect historical power monitoring data and historical judgment results of illegal production by coal mining enterprises, and to clean the data and process missing and outlier values.

[0134] The feature engineering module is used to extract features from electricity consumption data that help identify illegal production.

[0135] The anomaly detection module is used to analyze power consumption data in real time or in batches to identify abnormal patterns.

[0136] The correlation analysis module is used to perform correlation analysis between abnormal power consumption at suspected mining sites and personnel change characteristics to determine whether these sites are indeed carrying out mining operations.

[0137] This embodiment also provides a computer device applicable to the method for judging illegal production operations in coal mines based on machine learning, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for judging illegal production operations in coal mines based on machine learning as proposed in the above embodiment.

[0138] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0139] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the method for judging illegal production operations in coal mines based on machine learning as proposed in the above embodiments.

[0140] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0141] Example 3

[0142] Based on the previous two embodiments, this embodiment provides a machine learning-based method for judging illegal production operations in coal mines. To verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0143] Table 1: Comparison of Simulation Experiments Using Different Methods

[0144]

[0145] Experimental data shows that the method of this invention outperforms existing technologies in key indicators such as accuracy, recall, and F1 score. By combining an abnormal power consumption identification model and correlation analysis, this invention can more accurately identify and judge the illegal production behavior of coal mining enterprises, effectively improving the efficiency and accuracy of supervision.

[0146] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A machine learning-based method for judging illegal production operations in coal mines, characterized by: Includes the following steps, Based on historical power monitoring data of enterprises and historical judgment results of illegal production by coal mining enterprises, a model for identifying abnormal power consumption of enterprises is established using machine learning. The electricity consumption data of each coal mine area obtained from the enterprise electricity consumption classification and prediction results are input into the enterprise electricity consumption anomaly identification model to determine the electricity consumption anomaly area and the abnormal electricity consumption. Obtain personnel characteristic data in areas with abnormal power consumption, and perform cluster analysis on these areas based on the personnel characteristic data to identify suspected mining operation sites; The abnormal power consumption and personnel change characteristics of suspected mining sites are correlated and analyzed to determine whether the suspected mining sites are indeed mining sites. After confirming the suspected mining operation site as a mining operation site, the process also includes, When the abnormal power consumption at the mining operation site exceeds the first-level power consumption, an over-intensity violation production warning is generated. When the abnormal power consumption at the collection point is less than the first-level power consumption but greater than the second-level power consumption, and the number of workers at the mining point exceeds 20% of the preset number, an over-intensity violation production warning message is generated.

2. The method for judging illegal production operations in coal mines based on machine learning as described in claim 1, characterized in that: The steps for establishing the enterprise power consumption anomaly identification model include: Establish a correlation between historical power monitoring data of enterprises and historical judgment results of illegal production by coal mining enterprises; Using the aforementioned correspondence as a standard, the historical power monitoring data of enterprises and the historical judgment results of illegal production by coal mining enterprises are used as training data to train an initial identification model determined by machine learning, and the training results are obtained. Based on the judgment criteria for illegal production operations in coal mining enterprises, the model attribute evaluation index is determined, and when the training results meet the model attribute evaluation index, the abnormal power consumption identification model of enterprises is obtained.

3. The method for judging illegal production operations in coal mines based on machine learning as described in claim 2, characterized in that: The steps for determining the abnormal power consumption area and the abnormal power consumption include, The power consumption data of each coal mining area is determined by the result data, and the power consumption data is processed based on the time dimension to obtain multiple sets of input data; The input data is sequentially input into the enterprise power consumption anomaly identification model to obtain the power consumption identification results for the coal mine area; Based on the power consumption identification results, abnormal power consumption areas and abnormal power consumption are obtained.

4. The method for judging illegal production operations in coal mines based on machine learning as described in claim 3, characterized in that: The methods for obtaining the suspected mining operation site include: Based on the personnel characteristic data, determine the personnel location data, personnel quantity data, and personnel type data; By using personnel location data and personnel quantity data, the personnel area density is determined. The abnormal power consumption area is divided into multiple density areas based on the personnel area density. From the multiple density areas, a target density area whose density meets the preset density is obtained. Obtain the density distribution sequence of the density magnitude of the target density region, and determine the number of clusters and cluster centers based on the density distribution sequence; The regional name information of each target density area is classified to obtain multiple name types, and the personnel type data is classified to obtain multiple job types. Based on the name type and job type, combined with historical coal mine enterprise operation data, an evaluation index system is established. By using a preset clustering algorithm and combining it with an evaluation index system, cluster analysis is performed on the target density region according to the number of clusters and the cluster centers to obtain the region clustering results; The average features of each target density region are obtained from the regional clustering results. The target density regions in the target density region group whose average features meet the preset operation features are taken as suspected mining operation points. The methods for determining the number of clusters and cluster centers include: Obtain the center point, mean point, and median point of the density distribution sequence; The number of clusters is determined by the center point and the length of the density distribution sequence; The average of the mean point and the median point is then used as a cluster center. Based on the number of clusters and the length of the density distribution sequence, the density distribution sequence is divided, and other cluster centers are obtained based on the division points.

5. The method for judging illegal production operations in coal mines based on machine learning as described in claim 4, characterized in that: The method for determining whether a suspected mining operation site is indeed a mining operation site includes... The abnormal power consumption at the mining operation point is analyzed along the time dimension to obtain the abnormal power consumption curve, and the fluctuation points and fluctuation amplitude are obtained from the abnormal power consumption curve. Multiple sets of change characteristics are obtained by analyzing the change characteristics of personnel according to the change attributes, and each set of change characteristics is analyzed according to the time dimension to obtain the attribute change curve corresponding to the change attribute. The attribute fluctuation amplitude at the same time point as the fluctuation point is obtained from the attribute change curve. Based on the degree of influence of the changing attributes on mining operations, the curve weights of the attribute change curves are determined, and the attribute fluctuation amplitude is weighted and analyzed based on the curve weights to obtain the target attribute fluctuation amplitude. Obtain the correlation between the fluctuation range of target attributes and the fluctuation range. Based on the correlation of the fluctuation range of all target attributes, determine the comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining operation points. Based on the magnitude of abnormal power consumption at suspected mining sites, a first score is determined. The average characteristics of personnel changes at suspected mining sites are obtained. Based on the characteristic information of the average characteristics, a second score is determined. Based on the comprehensive correlation between abnormal power consumption and personnel changes at suspected mining sites, a third score is determined. Based on the first score, the second score and the third score, a comprehensive score is obtained for the suspected mining operation site. The suspected mining operation site with a comprehensive score greater than the preset score is selected as the mining operation site. The method for determining the comprehensive correlation between abnormal power consumption and personnel change characteristics at suspected mining sites includes: By summing the correlations of the fluctuation ranges of the target attributes, a comprehensive correlation between the abnormal power consumption and personnel change characteristics of suspected mining sites is obtained.

6. The method for judging illegal production operations in coal mines based on machine learning as described in claim 5, characterized in that: This also includes real-time adjustments to the enterprise power consumption anomaly identification model, with specific steps including... Real-time change data of coal mine operations is obtained, and the real-time change data is input into the enterprise power consumption anomaly identification model to obtain the first result; The first result is used as a parameter in the next real-time change data and input into the enterprise power consumption anomaly identification model to obtain the second result; Based on the difference between the changing trends of the first and second results and the actual changing trends, the model parameters are adjusted in real time. The steps for real-time adjustment of model parameters include: Based on the difference between the changing trends of the first and second results and the actual changing trends, the adjustment range of the model parameters is determined, and the corrected changing trend of the changing trends of the first and second results under the adjustment range is determined. Based on the difference between the corrected trend and the actual trend, the model parameters are iteratively adjusted until the difference between the latest corrected trend and the actual trend meets the difference requirement.

7. A machine learning-based system for judging illegal production operations in coal mines, based on the machine learning-based method for judging illegal production operations in coal mines as described in any one of claims 1 to 6, characterized in that: It includes a data collection and preprocessing module, a feature engineering module, an anomaly detection module, and a correlation analysis module; The data collection and preprocessing module is used to collect historical power monitoring data and historical judgment results of illegal production by coal mining enterprises, and to clean the data and process missing and outlier values. The feature engineering module is used to extract features from electricity consumption data that help identify illegal production. The anomaly detection module is used to analyze power consumption data in real time or in batches to identify abnormal patterns. The correlation analysis module is used to perform correlation analysis between abnormal power consumption at suspected mining sites and personnel change characteristics to determine whether these sites are indeed carrying out mining operations.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the machine learning-based method for judging illegal production operations in coal mines as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the machine learning-based method for judging illegal production operations in coal mines as described in any one of claims 1 to 6.

Images