Multi-system information fusion method and system for coal mine
By processing and fusion of coal mine multi-system data, and using lidar and SLAM technology to build a composite map, the problem of inability to share and interconnect information in coal mine system is solved, and efficient resource utilization and rapid location of faulty equipment are achieved.
Patent Information
- Application Number
- CN202411132956.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-12-06
AI Technical Summary
The systems in coal mines operate independently, and information cannot be shared and interconnected, resulting in low resource utilization, which seriously restricts the progress of the integration of "two-in-one" construction.
By acquiring multi-system data, data preprocessing, normalization processing, data fusion of the same type, feature extraction, and displaying fault information on the composite map. The map is built using lidar and SLAM technology to achieve rapid location of equipment failures.
It improves the resource utilization rate of coal mine system information, quickly locates faulty equipment, and improves the accuracy of equipment fault judgment.
Smart Images

Figure CN119089380A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of information technology, and in particular relates to a multi-system information fusion method and system for coal mines. Background Art
[0002] At present, the system integration of various manufacturers has not achieved true system integration. Each system runs independently. It is very difficult to achieve integration on the previous system and requires the addition of many devices.
[0003] Information systems such as mine monitoring and personnel positioning have long been operating independently, with no way to share information or interconnect them. This has seriously restricted the progress of the "two-in-one" integration of coal mines, resulting in a large amount of duplicate construction and low resource utilization. Summary of the invention
[0004] The purpose of the present invention is to provide a multi-system information fusion method and system for coal mines, which processes data uploaded by multiple systems, fuses data of the same type, analyzes the fused data and displays equipment fault information on a constructed composite map, thereby solving the problem that existing coal mine information cannot be shared and interconnected, resulting in low resource utilization.
[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:
[0006] The present invention is a multi-system information fusion method for coal mines, comprising the following steps:
[0007] Step S1: Obtain data uploaded by multiple systems of a coal mine;
[0008] Step S2: preprocessing the acquired data;
[0009] Step S3: normalizing the processed data;
[0010] Step S4: fusing data of the same type;
[0011] Step S5: extracting features from the fused data to reduce the classification correlation in the initial data state;
[0012] Step S6: construct a composite map and perform fault diagnosis on the data according to the classification association degree.
[0013] As a preferred technical solution, in step S2, data preprocessing includes data cleaning, data format conversion and spatial reference unification; the data cleaning is used to delete data information with missing information, misspellings and similar repeated records; the data format conversion uses ArcGIS to convert the data format; the spatial reference unification uses SphereMercator for projection processing, and the format is selected as shapefile format; after the sample data is preprocessed, data fusion work needs to be carried out, through which errors and redundant data are removed, and the data is strictly limited to a controllable range, thereby improving the accuracy of coal mine equipment fault judgment.
[0014] As a preferred technical solution, the operating status information of each part is collected and uploaded through a dedicated sensor device. In the data information processing, it is necessary to consider the influencing conditions such as the external environment and human factors. In order to ensure that the monitoring signal data meets the reference requirements, it is necessary to perform preliminary processing according to the actual operation of the equipment to effectively remove the signals containing interference and noise. When cleaning the data, the data observation sequence to be cleaned is {x1, x2, ..., x n}, the data of this sequence can be expressed as:
[0015] d i =2x j -(x j+1 -x j-1 ),j=2,3...,n-1;
[0016] Then the statistical mean of the sequence data can be obtained and mean square error σ d :
[0017]
[0018] As a preferred technical solution, in step S3, the specific formula for normalization processing is as follows:
[0019]
[0020] Where, X i (d) is the normalized data. The mapping interval of the differential data is [0.1, 0.9]. i The original data value provided by the system, maxX i and minX i Respectively represent the maximum and minimum values in the data samples, i = 1, 2, 3, ..., n.
[0021] As a preferred technical solution, in step S4, Kalman filtering is used for fusion, and the specific formula is as follows:
[0022] Xk+1 =g(X k ,U k )+W k
[0023] Y k+1 =h(X k+1 )+V k+1 ;
[0024] Where, X k+1 is the system state variable at the k+1th moment, g is the state transfer matrix function, h is the observed variable state function, U k is the input data at the kth moment, W k is the system noise at the kth moment, Y k+1 is the fused data at the k+1th moment, V k is the observation noise at the kth moment.
[0025] As a preferred technical solution, in step S5, the specific steps of extracting features from data are as follows:
[0026] Step S51: Standardize the original matrix data and convert the state signal into standard application data; record the monitoring parts as I1, I2, ..., I n , the data collected by the sensor is recorded as x 1j ,x 2j ,...,x nj ; where x ij is the sample value of the i-th monitoring part of the j-th monitoring object; after all sample values are standardized, they are put into X=(x ij ) mxn And obtain the new matrix Z = (z ij ) mxn , the specific formula is as follows:
[0027]
[0028] In the formula, m is the total number of sample values;
[0029] Step S52: Calculate the covariance matrix of the matrix Z. The specific expression is as follows:
[0030]
[0031] In the formula, r ik is the correlation coefficient value between random numbers;
[0032] Step S53: perform eigendecomposition on the covariance matrix, obtain eigenvalues and arrange them in order from large to small;
[0033] Step S54: using the cumulative contribution rate method to obtain the number of principal components of the system monitoring signal corresponding to the coal mine;
[0034] Step S55: determine K principal components, and calculate and obtain the first K eigenvalues and unit eigenvectors.
[0035] As a preferred technical solution, in step S54, the calculation formula of the main component contribution rate is: Then the cumulative contribution rate of the first K principal components is In the formula, λ i and λ j They are represented as the i-th and j-th eigenvalues respectively.
[0036] As a preferred technical solution, in step S55, for the first K principal elements, the projection w of the matrix Z on the principal element matrix in the direction of the eigenvector is i , then the i-th principal component sample value of the l-th data sample can be obtained as:
[0037]
[0038] In the formula, is the eigenvalue λ i The jth component of the corresponding eigenvector, x lj is the initial sample value of the lth monitored object regarding the jth feature.
[0039] As a preferred technical solution, in step S6, the composite map is constructed using a laser radar as the main sensor. The laser radar performs a 360° two-dimensional or three-dimensional scan within the monitoring range to generate point cloud data information. Then, the surrounding environment map is constructed based on the data information provided by the SLAM laser radar, and its own position is calculated. The conversion of the laser radar to the SLAM map data requires an intermediate conversion, and the formula is:
[0040]
[0041] In the formula, ζ x , y They represent the horizontal and vertical coordinates of the laser radar, θ is the observation angle of the laser radar, η is the probability of the existence of obstacles, T α is the position coordinate of the laser point;
[0042] The SLAM composite map construction technology for complex industrial environments mainly studies two aspects: map optimization SLAM technology and topological map establishment technology. Among them, the common technology of multimodal sensor data processing provides a basis for solving the front-end problems of SLAM and the loop detection problems of SLAM; the back-end uses the back-end optimization algorithm based on SLAM and the loop detection algorithm based on key frame feature matching to improve the accuracy and robustness of the sub-environment reconstruction map. A hierarchical map structure based on topological structure is adopted to reduce the computing and storage resource occupation of the full-scale scene map. The topological layer of the hierarchical topological map uses the GICP algorithm based on three-dimensional laser for global loop detection to connect each sub-environment of the node layer, realize the unification of the sub-environment map in the spatial dimension, and correct the global map error at the same time. The advantage of lidar is that it can measure the angle and distance of obstacle points with higher accuracy, which is convenient for positioning navigation and map construction.
[0043] The present invention is a multi-system information fusion system for coal mines, comprising a multi-system unit, a data processing unit, a feature extraction unit, a composite map construction unit and a fault location unit;
[0044] The multi-system unit is installed in multiple locations and on multiple devices in the coal mine, and is uploaded to the data processing unit via a wireless communication module;
[0045] The data processing unit includes a data cleaning module, a format conversion module, a space reference unification module and a data fusion module; the data cleaning module is used to clean the data uploaded by multiple system units; the format conversion module is used to unify the data format uploaded by multiple system units; the space reference unification module is used to build a surrounding environment map according to the location information collected by the laser radar and calculate its own location; the data fusion module is used to fuse the data of multiple system units;
[0046] The feature extraction unit is used to reduce the dimension and remove noise of the original data according to the second-order statistical characteristics of the data, and the extracted features reduce the classification association degree under the initial data state so as to obtain the corresponding principal component value;
[0047] The composite map construction unit is used to generate point cloud data information based on the information collected by the laser radar, construct a surrounding environment map, and calculate its own position;
[0048] The fault location unit is used to determine whether the equipment is faulty based on the collected data and to display the location of the faulty equipment on the generated composite map.
[0049] The present invention has the following beneficial effects:
[0050] The present invention processes data uploaded by multiple systems, fuses data of the same type, analyzes the fused data and displays equipment fault information on a constructed composite map, thereby improving resource utilization of coal mine system information and quickly locating faulty coal mine equipment.
[0051] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0053] Figure 1 A flow chart of a multi-system information fusion method for coal mines according to the present invention;
[0054] Figure 2 The present invention is a schematic diagram of the structure of a multi-system information fusion system for coal mines. DETAILED DESCRIPTION
[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0056] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0058] Embodiment 1
[0059] See also Figure 1 As shown, the present invention is a multi-system information fusion method for coal mines, comprising the following steps:
[0060] Step S1: Obtain data uploaded by multiple systems of a coal mine;
[0061] Step S2: preprocessing the acquired data;
[0062] Step S3: normalizing the processed data;
[0063] Step S4: fusing data of the same type;
[0064] Step S5: extracting features from the fused data to reduce the classification correlation in the initial data state;
[0065] Step S6: construct a composite map and perform fault diagnosis on the data according to the classification association degree.
[0066] In step S2, data preprocessing includes data cleaning, data format conversion and spatial reference unification; data cleaning is used to delete data information with missing information, misspellings and similar duplicate records; data format conversion uses ArcGIS to convert the data format; spatial reference unification uses SphereMercator for projection processing, and the format is selected as shapefile format; after sample data preprocessing, data fusion work needs to be carried out to remove errors and redundant data, and strictly limit the data within a controllable range, thereby improving the accuracy of coal mine equipment fault judgment.
[0067] The operating status information of each part is collected and uploaded through dedicated sensor equipment. In data information processing, it is necessary to consider the influencing conditions such as external environment and human factors. In order to ensure that the monitoring signal data meets the reference requirements, it is necessary to perform preliminary processing according to the actual operation of the equipment to effectively remove the signals containing interference and noise. When cleaning data, the data observation sequence to be cleaned is {x1,x2,...,x n}, the data of this sequence can be expressed as:
[0068] d i =2x j -(x j+1 -x j-1 ),j=2,3...,n-1;
[0069] Then the statistical mean of the sequence data can be obtained and mean square error σ d :
[0070]
[0071] In step S3, the specific formula for normalization processing is as follows:
[0072]
[0073] Where, X i (d) is the normalized data. The mapping interval of the differential data is [0.1, 0.9].i The original data value provided by the system, maxX i and minX i Respectively represent the maximum and minimum values in the data samples, i = 1, 2, 3, ..., n.
[0074] In step S4, Kalman filtering is used for fusion, and the specific formula is as follows:
[0075] X k+1 =g(X k ,U k )+W k
[0076] Y k+1 =h(X k+1 )+V k+1 ;
[0077] Where, X k+1 is the system state variable at the k+1th moment, g is the state transfer matrix function, h is the observed variable state function, U k is the input data at the kth moment, W k is the system noise at the kth moment, Y k+1 is the fused data at the k+1th moment, V k is the observation noise at the kth moment.
[0078] In step S5, the specific steps of extracting features from data are as follows:
[0079] Step S51: Standardize the original matrix data and convert the state signal into standard application data; record the monitoring parts as I1, I2, ..., I n , the data collected by the sensor is recorded as x 1j ,x 2j ,...,x nj ; where x ij is the sample value of the i-th monitoring part of the j-th monitoring object; after all sample values are standardized, they are put into X=(x ij ) mxn And obtain the new matrix Z = (z ij ) mxn , the specific formula is as follows:
[0080]
[0081] In the formula, m is the total number of sample values;
[0082] Step S52: Calculate the covariance matrix of the matrix Z. The specific expression is as follows:
[0083]
[0084] In the formula, r ik is the correlation coefficient value between random numbers;
[0085] Step S53: perform eigendecomposition on the covariance matrix, obtain eigenvalues and arrange them in order from large to small;
[0086] Step S54: using the cumulative contribution rate method to obtain the number of principal components of the system monitoring signal corresponding to the coal mine;
[0087] Step S55: determine K principal components, and calculate and obtain the first K eigenvalues and unit eigenvectors.
[0088] In step S54, the calculation formula of the principal component contribution rate is: Then the cumulative contribution rate of the first K principal components is In the formula, λ i and λ j They are represented as the i-th and j-th eigenvalues respectively.
[0089] In step S55, for the first K principal components, the projection w of the matrix Z on the principal component matrix in the direction of the eigenvector is i , then the i-th principal component sample value of the l-th data sample can be obtained as:
[0090]
[0091] In the formula, is the eigenvalue λ i The jth component of the corresponding eigenvector, x lj is the initial sample value of the lth monitored object regarding the jth feature.
[0092] In step S6, the composite map is constructed using laser radar as the main sensor. The laser radar performs 360° two-dimensional or three-dimensional scanning within the monitoring range to generate point cloud data information. Then, the surrounding environment map is constructed based on the data information provided by the SLAM laser radar, and its own position is calculated. The conversion of the laser radar to the SLAM map data requires intermediate conversion, and the formula is:
[0093]
[0094] In the formula, ζ x , y They represent the horizontal and vertical coordinates of the laser radar, θ is the observation angle of the laser radar, η is the probability of the existence of obstacles, T α is the position coordinate of the laser point;
[0095] The SLAM composite map construction technology for complex industrial environments mainly studies two aspects: map optimization SLAM technology and topological map establishment technology. Among them, the common technology of multimodal sensor data processing provides a basis for solving the front-end problems of SLAM and the loop detection problems of SLAM; the back-end uses the back-end optimization algorithm based on SLAM and the loop detection algorithm based on key frame feature matching to improve the accuracy and robustness of the sub-environment reconstruction map. A hierarchical map structure based on topological structure is adopted to reduce the computing and storage resource occupation of the full-scale scene map. The topological layer of the hierarchical topological map uses the GICP algorithm based on three-dimensional laser for global loop detection to connect each sub-environment of the node layer, realize the unification of the sub-environment map in the spatial dimension, and correct the global map error at the same time. The advantage of lidar is that it can measure the angle and distance of obstacle points with higher accuracy, which is convenient for positioning navigation and map construction.
[0096] Embodiment 2
[0097] See also Figure 2 As shown, the present invention is a multi-system information fusion system for coal mines, which can be used to execute the method content of Example 1 of the present invention, including: a multi-system unit, a data processing unit, a feature extraction unit, a composite map construction unit and a fault location unit;
[0098] Multiple system units are installed in multiple locations and on a variety of equipment in the coal mine, and are uploaded to the data processing unit via wireless communication modules;
[0099] The data processing unit includes a data cleaning module, a format conversion module, a space reference unification module and a data fusion module; the data cleaning module is used to clean the data uploaded by multiple system units; the format conversion module is used to unify the data format uploaded by multiple system units; the space reference unification module is used to build a surrounding environment map based on the location information collected by the laser radar and calculate its own location; the data fusion module is used to fuse the data of multiple system units;
[0100] The feature extraction unit is used to reduce the dimension and denoise the original data according to the second-order statistical characteristics of the data. The extracted features reduce the classification association degree under the initial data state so as to obtain the corresponding principal component value;
[0101] The composite map construction unit is used to generate point cloud data information based on the information collected by the laser radar, construct a map of the surrounding environment, and calculate its own position;
[0102] The fault location unit is used to determine whether the equipment is faulty based on the collected data and to display the location of the faulty equipment on the generated composite map.
[0103] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0104] In addition, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.
[0105] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A multi-system information fusion method for coal mines, It is characterized in that The steps include: Step S1: Obtain data uploaded by multiple systems of a coal mine; Step S2: preprocessing the acquired data; Step S3: normalizing the processed data; Step S4: fusing data of the same type; Step S5: extracting features from the fused data to reduce the classification correlation in the initial data state; Step S6: construct a composite map and perform fault diagnosis on the data according to the classification association degree.
2. A multi-system information fusion method for coal mines according to claim 1, It is characterized in that In step S2, data preprocessing includes data cleaning, data format conversion and spatial reference unification; the data cleaning is used to delete data information with missing information, misspellings and similar duplicate records; the data format conversion uses ArcGIS to convert the data format; The spatial reference is uniformly projected using SphereMercator, and the shapefile format is selected.
3. A multi-system information fusion method for coal mines according to claim 2, It is characterized in that When cleaning the data, suppose the data observation sequence to be cleaned is {x 1 ,x 2 ,...,x n }, the data of this sequence can be expressed as: d i =2x j -(x j+1 -x j-1 ),j=2,3...,n-1; Then the statistical mean of the sequence data can be obtained and mean square error σ d :
4. A multi-system information fusion method for coal mines according to claim 1, It is characterized in that In step S3, the specific formula for normalization processing is as follows: Where, X i (d) is the normalized data. The mapping interval of the differential data is [0.1, 0.9]. i The original data value provided by the system, maxX i and minX i Respectively represent the maximum and minimum values in the data samples, i = 1, 2, 3, ..., n.
5. A multi-system information fusion method for coal mines according to claim 1, It is characterized in that In step S4, Kalman filtering is used for fusion, and the specific formula is as follows: Where, X k+1 For the k+1 The system state variables at the moment, g is the state transfer matrix function, h is the observed variable state function, U k is the input data at the kth moment, W k is the system noise at the kth moment, Y k+1 is the fused data at the k+1th moment, V k is the observation noise at the kth moment.
6. A multi-system information fusion method for coal mines according to claim 1, It is characterized in that In step S5, the specific steps of extracting features from data are as follows: Step S51: Standardize the original matrix data and convert the state signal into standard application data; record the monitoring part as I 1 ,I 2 ,...,I n , the data collected by the sensor is recorded as x 1j ,x 2j ,...,x nj ; where x ij is the sample value of the i-th monitoring part of the j-th monitoring object; after all sample values are standardized, they are put into X=(x ij ) mxn And obtain the new matrix Z = (z ij ) mxn , the specific formula is as follows: In the formula, m is the total number of sample values; Step S52: Calculate the covariance matrix of the matrix Z. The specific expression is as follows: In the formula, r ik is the correlation coefficient value between random numbers; Step S53: perform eigendecomposition on the covariance matrix, obtain eigenvalues and arrange them in order from large to small; Step S54: using the cumulative contribution rate method to obtain the number of principal components of the system monitoring signal corresponding to the coal mine; Step S55: determine K principal components, and calculate and obtain the first K eigenvalues and unit eigenvectors.
7. A multi-system information fusion method for coal mines according to claim 6, It is characterized in that In step S54, the calculation formula of the main component contribution rate is: Then the cumulative contribution rate of the first K principal components is In the formula, λ i and λ j They are represented as the i-th and j-th eigenvalues respectively.
8. A multi-system information fusion method for coal mines according to claim 6, It is characterized in that In step S55, for the first K principal components, the projection w of the matrix Z on the principal component matrix in the direction of the eigenvector is i , then the i-th principal component sample value of the l-th data sample can be obtained as: In the formula, is the eigenvalue λ i The jth component of the corresponding eigenvector, x lj is the initial sample value of the lth monitored object regarding the jth feature.
9. A multi-system information fusion method for coal mines according to claim 1, It is characterized in that In step S6, the composite map is constructed using a laser radar as the main sensor. The laser radar performs a 360° two-dimensional or three-dimensional scan within the monitoring range to generate point cloud data information. Then, a surrounding environment map is constructed based on the data information provided by the SLAM laser radar, and its own position is calculated. The conversion of the laser radar to the SLAM map data requires an intermediate conversion, and the formula is: In the formula, ζ x , y They represent the horizontal and vertical coordinates of the laser radar, θ is the observation angle of the laser radar, η is the probability of the existence of obstacles, T α is the position coordinate of the laser point.
10. A multi-system information fusion system for coal mines according to claim 1, comprising a multi-system unit, a data processing unit, a feature extraction unit, a composite map construction unit and a fault location unit; Features: The multi-system unit is installed in multiple locations and on multiple devices in the coal mine, and is uploaded to the data processing unit via a wireless communication module; The data processing unit includes a data cleaning module, a format conversion module, a space reference unification module and a data fusion module; the data cleaning module is used to clean the data uploaded by multiple system units; the format conversion module is used to unify the data format uploaded by multiple system units; the space reference unification module is used to build a surrounding environment map based on the location information collected by the laser radar and calculate its own location; the data fusion module is used to fuse the data of multiple system units; The feature extraction unit is used to reduce the dimension and remove noise of the original data according to the second-order statistical characteristics of the data, and the extracted features reduce the classification association degree under the initial data state so as to obtain the corresponding principal component value; The composite map construction unit is used to generate point cloud data information based on the information collected by the laser radar, construct a surrounding environment map, and calculate its own position; The fault location unit is used to determine whether the equipment is faulty based on the collected data and to display the location of the faulty equipment on the generated composite map.
