Mine gas detection and intelligent emission control system based on big data
By introducing an adaptive distance measurement function and a dynamic clustering center selection mechanism in the mine gas detection system, combined with individual fitness-driven optimization strategies and outlier detection mechanism, the problem of system identification of abnormal emissions under multi-dimensional features is solved, and the problem of improving the Test efficiency and accuracy.
Patent Information
- Application Number
- CN202411437863.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2024-12-27
- Estimated Expiration
- 2044-10-15
AI Technical Summary
The existing mine gas detection and intelligent control systems cannot accurately identify similar detection point emission patterns under multi-dimensional characteristics. The initial clustering center cannot identify abnormal emission gathering points with higher density, resulting in the inability to identify abnormal transactions or malicious detection point emissions. , and improper parameter setting leads to low detection efficiency.
Introduce adaptive weight and maximum normalization to design the adaptive distance measurement function, adopt dynamic cluster center selection mechanism based on maximum local density and multi-layer marking initialization mechanism, combine individual fitness-driven position update strategy, and use inverse error function outlier detection mechanism, compensates for the outlier threshold through median and error functions.
Enhanced clustering robustness and effectiveness of mine emission detection data, realize accurate detection of abnormal emissions, improve detection efficiency and convergence, can more accurately identify outlier detection points, and improve mine gas detection emissions Control efficiency.
Smart Images
Figure CN118963225B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of emission detection technology, and specifically to a mine gas detection and intelligent emission control system based on big data. Background Art
[0002] The mine gas detection and intelligent emission control system based on big data is an intelligent system that uses big data technology, the Internet of Things, sensors, artificial intelligence and other advanced technologies to monitor, analyze and control harmful gases in the mine environment in real time. Its main purpose is to ensure the safety of mine workers, reduce environmental pollution and improve production efficiency. However, the general mine gas detection and intelligent control system has the problem that the distance measurement cannot accurately identify similar detection point emission patterns under multi-dimensional features, and the initial clustering center cannot identify high-density abnormal emission clusters, which leads to the inability to identify abnormal transactions or malicious detection point emissions; the general mine gas detection and intelligent control system parameters are improperly set and the convergence of parameter optimization is poor, which cannot accurately identify abnormal detection points, resulting in low detection point detection efficiency. Summary of the invention
[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a mine gas detection and intelligent emission control system based on big data. In view of the problem that the distance measurement of general mine gas detection and intelligent control systems cannot accurately identify similar detection point emission patterns under multi-dimensional features, and the initial clustering center cannot identify abnormal emission clusters with high density, which leads to the inability to identify abnormal transactions or malicious detection point emissions, this scheme introduces adaptive weights and maximum value normalization to design an adaptive distance measurement function, designs a dynamic clustering center selection mechanism based on maximum local density, and a mechanism for initializing the remaining clustering centers based on multi-layer labeling, thereby enhancing the robustness and effectiveness of clustering mine emission detection data. It can identify abnormal emission clusters with high density; thus, it can realize accurate detection of abnormal emissions from e-commerce detection points; in view of the improper parameter settings of general mine gas detection and intelligent control systems and the poor convergence of parameter optimization, it is impossible to accurately identify abnormal detection points, which leads to low efficiency of detection point detection. This scheme is based on individual fitness drive, combines global and local optimization to design a position update strategy, balances accuracy and convergence speed, and improves the convergence of anomaly detection; based on the outlier detection mechanism of the inverse error function, it can effectively distinguish the detection points with abnormal emissions, and use the median and error function to compensate for the setting of the abnormal threshold, which can more accurately detect the outlier detection points; thus, the efficiency of mine gas detection emission control is improved.
[0004] The technical solution adopted by the present invention is as follows: The mine gas detection and intelligent emission control system based on big data provided by the present invention includes a data acquisition module, a data preprocessing module, a clustering processing module and a gas detection emission and intelligent control module;
[0005] The data acquisition module collects mine emission detection data;
[0006] The data preprocessing module performs data cleaning, data conversion and standardization on the collected data;
[0007] The cluster processing module designs an adaptive distance metric function by introducing adaptive weights and maximum value normalization; designs a dynamic cluster center selection mechanism based on maximum local density and a mechanism for initializing the remaining cluster centers based on multi-layer marking; based on individual fitness drive, combines global and local optimization to design a position update strategy to achieve cluster optimization; obtains anomaly sets based on an outlier detection mechanism based on an inverse error function; and then realizes cluster processing;
[0008] The gas detection, emission and intelligent control module realizes gas detection and emission control based on the clustering result.
[0009] Furthermore, in the data acquisition module, the mine emission detection data includes gas detection data, environmental monitoring data, equipment operation data and detection location data.
[0010] Furthermore, in the data preprocessing module, the data cleaning is to process missing values and duplicate values; the data conversion is to convert the data into a vector form; and the standardization is to standardize the data based on the maximum and minimum normalization method.
[0011] Furthermore, the cluster processing module specifically includes the following contents:
[0012] Construct an emission detection feature matrix unit; each row i of the emission detection feature matrix U represents a detection point, and each column k represents a different emission detection feature;
[0013] Design an adaptive distance metric function unit; expressed as: ;
[0014] Where, d i,j is the adaptive distance between detection point i and detection point j; m is the number of features; is the weight of emission detection feature k; x i,k and x j,k are the values of the kth emission detection feature at the ith detection point and the jth detection point, respectively; d(·) is the Mahalanobis distance; max(·) is the maximum value;
[0015] Initialize the cluster center unit; select the first cluster center z1 based on the maximum local density, expressed as: ; Calculate the adaptive distance d between all samples and the first cluster center i,1 ; Initialize the labels of all samples to 1; Find the sample with the largest adaptive distance to z1 and use it as the second cluster center z2; Calculate the adaptive distance d between all samples and z2 i,2 ; When d is satisfied i,2 <θ·d i,1 , update their marks to 2; ;in, is the local density of the i-th detection point; is a step function; is the distance threshold; is the threshold of adaptive distance; is the initial scaling factor, which controls the compactness of the clustering process; median(·) is the median; D is the sample set; is the smoothing parameter;
[0016] Initialize the decision unit; calculate the minimum adaptive distance D between the sample and all determined cluster centers i If satisfied ; then continue to generate new cluster centers; and so on, until no more ; Where n is the total number of detection points; D i is the minimum adaptive distance between the i-th detection point and all determined cluster centers;
[0017] Data allocation unit; based on the minimum adaptive distance, the detection points of non-cluster centers are allocated to the clusters where the cluster centers are located; the positions of the cluster centers are selected and updated as the new generation cluster centers; iterate repeatedly until the maximum number of iterations is reached or the clustering converges; if the clustering converges, the clustering is completed; if only the maximum number of iterations is reached, it goes to the clustering optimization unit; otherwise, it continues to iterate;
[0018] Cluster optimization unit; specifically includes:
[0019] Initialization; constructing a parameter search space based on the weights of each emission detection feature, distance threshold, adaptive distance threshold, and initial scaling factor; randomly initializing individual positions, and taking the average silhouette coefficient of the clustering results after K iterations based on individual positions as the individual fitness value;
[0020] Location update; location update method uses ;in, and are the updated and previous positions of the i1th individual respectively; w t is the movement weight; F ctis the fitness of the current individual; F ps is the fitness of the individual in the previous iteration; F max is the maximum fitness of the population in the previous iteration; is the mean of the individual positions in the neighborhood; is the local optimal enhancement factor;
[0021] Optimization judgment; a fitness threshold is set in advance. When there is an individual fitness value higher than the fitness threshold, the mine emission detection data is clustered based on the individual position, and the clustering result is obtained when the clustering converges; if the maximum number of iterations is reached, it returns to initialization; otherwise, it returns to position update;
[0022] Outlier detection unit; Outlier set Defined as ; ;in, is the i-th detection point; x is the detection point; is the complementary form of the inverse error function; c is the detection threshold.
[0023] Furthermore, the gas detection and emission control module detects the detection points in the outlier set as emission anomalies based on the clustering results, and the system sends alarm information to mine managers to achieve emission control.
[0024] The beneficial effects achieved by the present invention using the above scheme are as follows:
[0025] (1) In view of the fact that the distance measurement of general mine gas detection and intelligent control systems cannot accurately identify similar emission patterns of detection points under multi-dimensional features, and the initial cluster center cannot identify abnormal emission clusters with high density, which leads to the inability to identify abnormal emissions. This scheme introduces adaptive weights and maximum value normalization to design an adaptive distance measurement function, designs a dynamic cluster center selection mechanism based on the maximum local density, and initializes the remaining cluster centers based on multi-layer labeling. This enhances the robustness and effectiveness of clustering mine emission detection data, and can identify abnormal emission clusters with high density; thus, accurate detection of abnormal emissions at detection points can be achieved.
[0026] (2) In order to solve the problem that the parameters of general mine gas detection and intelligent control systems are improperly set and the convergence of parameter optimization is poor, which makes it impossible to accurately identify abnormal detection points and thus leads to low detection efficiency of detection points, this scheme is driven by individual fitness and combines global and local optimization to design a position update strategy, balance accuracy and convergence speed, and improve the convergence of anomaly detection; based on the outlier detection mechanism of the inverse error function, it can effectively distinguish the detection points with abnormal emissions, and use the median and error function to compensate for the setting of the abnormal threshold, which can more accurately detect outlier detection points, thereby improving the efficiency of mine gas detection and emission control. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 A schematic diagram of a mine gas detection and intelligent emission control system based on big data provided by the present invention;
[0028] Figure 2 Schematic diagram of the clustering processing module.
[0029] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION
[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0031] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.
[0032] Example 1, see Figure 1 , the mine gas detection and intelligent emission control system based on big data provided by the present invention includes a data acquisition module, a data preprocessing module, a clustering processing module and a gas detection emission and intelligent control module;
[0033] The data acquisition module collects mine emission detection data and sends the data to the data preprocessing module;
[0034] The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; and sends the data to the clustering processing module;
[0035] The cluster processing module designs an adaptive distance measurement function by introducing adaptive weights and maximum value normalization; designs a dynamic cluster center selection mechanism based on maximum local density and a mechanism for initializing the remaining cluster centers based on multi-layer marking; based on individual fitness drive, combines global and local optimization to design a position update strategy to achieve cluster optimization; obtains anomaly sets based on an outlier detection mechanism based on an inverse error function; and then implements cluster processing; and sends data to the gas detection emission and intelligent control module;
[0036] The gas detection, emission and intelligent control module realizes gas detection and emission control based on the clustering result.
[0037] Example 2, see Figure 1 This embodiment is based on the above embodiment. In the data acquisition module, the mine emission detection data includes gas detection data, environmental monitoring data, equipment operation data and detection location data.
[0038] Example 3, see Figure 1 This embodiment is based on the above embodiment. In the data preprocessing module, data cleaning is to process missing values and duplicate values; data conversion is to convert data into vector form; and standardization is to standardize data based on the maximum and minimum normalization method.
[0039] Example 4, see Figure 1 and Figure 2 This embodiment is based on the above embodiment, and the cluster processing module specifically includes the following contents:
[0040] Construct an emission detection feature matrix unit; each row i of the emission detection feature matrix U represents a detection point, and each column k represents a different emission detection feature;
[0041] Design an adaptive distance metric function unit; expressed as: ;
[0042] Where, d i,j is the adaptive distance between detection point i and detection point j; m is the number of features; is the weight of emission detection feature k; x i,k and x j,k are the values of the kth emission detection feature at the ith detection point and the jth detection point, respectively; d(·) is the Mahalanobis distance; max(·) is the maximum value;
[0043] Initialize the cluster center unit; select the first cluster center z1 based on the maximum local density, expressed as: ; Calculate the adaptive distance d between all samples and the first cluster center i,1; Initialize the labels of all samples to 1; Find the sample with the largest adaptive distance to z1 and use it as the second cluster center z2; Calculate the adaptive distance d between all samples and z2 i,2 ; When d is satisfied i,2 <θ·d i,1 , update their marks to 2; ;in, is the local density of the i-th detection point; is a step function; is the distance threshold; is the threshold of adaptive distance; is the initial scaling factor, which controls the compactness of the clustering process; median(·) is the median; D is the sample set; is the smoothing parameter;
[0044] Initialize the decision unit; calculate the minimum adaptive distance D between the sample and all determined cluster centers i If satisfied ; then continue to generate new cluster centers; and so on, until no more ; Where n is the total number of detection points; D i is the minimum adaptive distance between the i-th detection point and all determined cluster centers;
[0045] Data allocation unit; based on the minimum adaptive distance, the detection points of non-cluster centers are allocated to the clusters where the cluster centers are located; the positions of the cluster centers are selected and updated as the new generation of cluster centers; iterate repeatedly until the maximum number of iterations is reached or the clustering converges; if the clustering converges, the clustering is completed; if only the maximum number of iterations is reached, it goes to the clustering optimization unit; otherwise, it continues to iterate.
[0046] By performing the above operations, the distance measurement of general mine gas detection and intelligent control systems cannot accurately identify similar detection point emission patterns under multi-dimensional features, and the initial clustering center cannot identify abnormal emission clusters with high density, which leads to the inability to identify abnormal emissions. This scheme introduces adaptive weights and maximum value normalization to design an adaptive distance measurement function, designs a dynamic clustering center selection mechanism based on the maximum local density, and a mechanism for initializing the remaining clustering centers based on multi-layer labeling, which enhances the robustness and effectiveness of clustering mine emission detection data, and can identify abnormal emission clusters with high density; thereby achieving accurate detection of abnormal emissions at detection points.
[0047] Example 5, see Figure 1 and Figure 2 This embodiment is based on the above embodiment, and the cluster processing module further includes the following contents:
[0048] Cluster optimization unit; specifically includes:
[0049] Initialization; constructing a parameter search space based on the weights of each emission detection feature, distance threshold, adaptive distance threshold, and initial scaling factor; randomly initializing individual positions, and taking the average silhouette coefficient of the clustering results after K iterations based on individual positions as the individual fitness value;
[0050] Location update; location update method uses ;in, and are the updated and previous positions of the i1th individual respectively; w t is the movement weight; F ct is the fitness of the current individual; F ps is the fitness of the individual in the previous iteration; F max is the maximum fitness of the population in the previous iteration; is the mean of the individual positions in the neighborhood; is the local optimal enhancement factor;
[0051] Optimization judgment; a fitness threshold is set in advance. When there is an individual fitness value higher than the fitness threshold, the mine emission detection data is clustered based on the individual position, and the clustering result is obtained when the clustering converges; if the maximum number of iterations is reached, it returns to initialization; otherwise, it returns to position update;
[0052] Outlier detection unit; Outlier set Defined as ; ;in, is the i-th detection point; x is the detection point; is the complementary form of the inverse error function; c is the detection threshold.
[0053] By performing the above operations, the problem of improper parameter setting and poor convergence of parameter optimization of general mine gas detection and intelligent control systems, which makes it impossible to accurately identify abnormal detection points and leads to low efficiency of detection point detection, is solved. This scheme is driven by individual fitness and combines global and local optimization to design a position update strategy to balance accuracy and convergence speed and improve the convergence of anomaly detection. The outlier detection mechanism based on the inverse error function can effectively distinguish the detection points with abnormal emissions, and the setting of the abnormal threshold is compensated by the median and error function, so as to more accurately detect outlier detection points, thereby improving the efficiency of mine gas detection and emission control.
[0054] Example 6, see Figure 1 ,This embodiment is based on the above embodiment, and the gas detection and emission control module detects the detection points in the outlier value set as emission anomalies based on the clustering results, and the system sends alarm information to the mine management personnel to achieve emission control.
[0055] Embodiment 7, see Figure 2 , this embodiment is based on the above embodiment, and the mine gas detection and intelligent emission control system based on big data provided by the present invention includes a data acquisition module, a data preprocessing module, a clustering processing module and a gas detection emission and intelligent control module;
[0056] The data acquisition module collects mine emission detection data and sends the data to the data preprocessing module;
[0057] The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; and sends the data to the clustering processing module;
[0058] The cluster processing module designs an adaptive distance measurement function by introducing adaptive weights and maximum value normalization; designs a dynamic cluster center selection mechanism based on maximum local density and a mechanism for initializing the remaining cluster centers based on multi-layer marking; based on individual fitness drive, combines global and local optimization to design a position update strategy to achieve cluster optimization; obtains anomaly sets based on an outlier detection mechanism based on an inverse error function; and then implements cluster processing; and sends data to the gas detection emission and intelligent control module;
[0059] The gas detection, emission and intelligent control module realizes gas detection and emission control based on the clustering result.
[0060] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0061] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.
[0062] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.
Claims
1. Mine gas detection and intelligent emission control system based on big data, characterized by: It includes data acquisition module, data preprocessing module, cluster processing module and gas detection emission and intelligent control module; The data acquisition module collects mine emission detection data; The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; The cluster processing module designs an adaptive distance metric function by introducing adaptive weights and maximum value normalization; designs a dynamic cluster center selection mechanism based on maximum local density and a mechanism for initializing the remaining cluster centers based on multi-layer marking; Based on individual fitness drive, the global and local optimizations are combined to design the position update strategy to achieve clustering optimization; the outlier detection mechanism based on the inverse error function is used to obtain the outlier set; and then clustering processing is achieved; The gas detection, emission and intelligent control module realizes gas detection and emission control based on the clustering results; The cluster processing module includes a design adaptive distance metric function unit; it is expressed as: ; Where, d i,j is the adaptive distance between user i and user j; m is the number of features; is the weight of emission feature k; x i,k and x j,k are the values of the kth emission feature of the i-th user and the j-th user, respectively; d(·) is the Mahalanobis distance; max(·) is the maximum value; The cluster processing module also includes the following contents: Construct emission detection feature matrix unit; Each row i of the emission detection feature matrix U represents a detection point, and each column k represents a different emission detection feature; Initialize the cluster center unit; select the first cluster center z1 based on the maximum local density, expressed as: ; Calculate the adaptive distance d between all samples and the first cluster center i,1 ; Initialize the labels of all samples to 1; Find the sample with the largest adaptive distance to z1 and use it as the second cluster center z2; Calculate the adaptive distance d between all samples and z2 i,2 ; When d is satisfied i,2 <θ·d i,1 , update their marks to 2; ;in, is the local density of the i-th detection point; is a step function; is the distance threshold; is the threshold of adaptive distance; is the initial scaling factor, which controls the compactness of the clustering process; median(·) is the median; D is the sample set; is the smoothing parameter; Initialize the decision unit; calculate the minimum adaptive distance D between the sample and all determined cluster centers i If satisfied ; then continue to generate new cluster centers; and so on, until no more ; Where n is the total number of detection points; D i is the minimum adaptive distance between the i-th detection point and all determined cluster centers; Data allocation unit; based on the minimum adaptive distance, the detection points of non-cluster centers are allocated to the clusters where the cluster centers are located; the positions of the cluster centers are selected and updated as the new generation cluster centers; iterate repeatedly until the maximum number of iterations is reached or the clustering converges; if the clustering converges, the clustering is completed; if only the maximum number of iterations is reached, it goes to the clustering optimization unit; otherwise, it continues to iterate; Cluster optimization unit; Outlier detection unit; Outlier set Defined as ; ;in, is the i-th detection point; x is the detection point; is the complementary form of the inverse error function; c is the detection threshold; The cluster optimization unit specifically includes: Initialization; constructing a parameter search space based on the weights of each emission detection feature, distance threshold, adaptive distance threshold, and initial scaling factor; randomly initializing individual positions, and taking the average silhouette coefficient of the clustering results after K iterations based on individual positions as the individual fitness value; Location update; location update method uses ;in, and are the updated and previous positions of the i1th individual respectively; w t is the movement weight; F ct is the fitness of the current individual; F ps is the fitness of the individual in the previous iteration; F max is the maximum fitness of the population in the previous iteration; is the mean of the individual positions in the neighborhood; is the local optimal enhancement factor; Optimization judgment; a fitness threshold is set in advance. When there is an individual fitness value higher than the fitness threshold, the mine emission detection data is clustered based on the individual position, and the clustering result is obtained when the clustering converges; if the maximum number of iterations is reached, it returns to initialization; otherwise, it returns to position update.
2. The mine gas detection and intelligent emission control system based on big data according to claim 1 is characterized by: In the data acquisition module, the mine emission detection data includes gas detection data, environmental monitoring data, equipment operation data and detection location data.
3. The mine gas detection and intelligent emission control system based on big data according to claim 1 is characterized by: The gas detection emission and intelligent control module detects the detection points in the abnormal value set as emission anomalies based on the clustering results, and the system sends alarm information to the mine management personnel to achieve emission control.
4. The mine gas detection and intelligent emission control system based on big data according to claim 1 is characterized by: In the data preprocessing module, the data cleaning is to process missing values and duplicate values; the data conversion is to convert the data into a vector form; and the standardization is to standardize the data based on the maximum and minimum normalization method.