Navigation positioning system data credibility decision-making method

Abstract

The invention belongs to the technical field of navigation and positioning of coal mine tunneling equipment, and provides a data credibility decision-making method for a navigation and positioning system in order to solve the technical problem that a navigation and positioning system data related credibility judgment and decision-making mechanism method does not exist in an underground coal mine at present. A corresponding all-parameter data model is established, dimension reduction processing is carried out, and principal components in the all-parameter data model are reserved; establishing a navigation positioning data decision mechanism, carrying out weighted K-means calculation on the principal component, judging the operation state of the to-be-tested equipment based on the full-parameter data model, fusing the two mechanisms to carry out credibility decision analysis, and adjusting the operation state of the equipment body according to a decision analysis result, so that the credibility of navigation positioning data is improved, and the reliability of the to-be-tested equipment is improved. And meanwhile, the equipment operation efficiency, the roadway forming quality and the mining safety can be further improved.

Description

Technical Field

[0001] The invention belongs to the technical field of navigation and positioning of coal mine excavation equipment, and in particular relates to a data credibility decision method for a navigation and positioning system. Background Art

[0002] At present, the evaluation criteria for intelligent tunneling equipment are mainly that the tunneling equipment has the functions of autonomous positioning, attitude determination, and orientation, and has the functions of precise autonomous positioning and navigation underground, precise attitude control, and automatic cutting.

[0003] Based on the above functional requirements, intelligent tunneling equipment is equipped with a navigation and positioning system for accurate navigation and positioning, attitude control and cutting control of underground coal mine equipment. In actual applications, due to the influence of the underground coal mine environment and the operation process of tunneling equipment, the navigation and positioning system has inaccurate heading angle guidance, untimely attitude information update, navigation information errors, etc. during operation, which leads to equipment deviation and even penetration of adjacent tunnels, seriously affecting the equipment cutting efficiency, tunnel forming quality and mining safety.

[0004] At present, as intelligent tunneling equipment is in the development stage, existing technologies are mainly used to solve problems such as navigation and positioning, combined algorithms, and cutting control. There is no technical content such as navigation and positioning system data-related credibility judgment and decision-making mechanism methods. Summary of the invention

[0005] In order to solve at least one of the above technical problems existing in the prior art, the present invention provides a method for determining the credibility of navigation and positioning system data.

[0006] The present invention is implemented by the following technical solution: a navigation and positioning system data credibility decision method, comprising the following steps:

[0007] Determine the navigation and positioning mode of the device under test, obtain parameter data of the corresponding navigation and positioning mode, and establish a full parameter data model based on the parameter data;

[0008] Performing dimensionality reduction processing on the full parameter data model to retain the principal components in the full parameter data model;

[0009] A navigation and positioning data decision mechanism is established, and a weighted K-means calculation is performed on the principal components. The first fault condition of the navigation and positioning of the equipment under test is predicted according to the calculation results. Meanwhile, the operating status of the equipment under test is judged based on the full-parameter data model. The second fault condition of the navigation and positioning of the equipment under test is obtained according to the judgment result. A credibility decision analysis is performed based on the first fault condition and the second fault condition, and the operating status of the equipment fuselage is adjusted according to the decision analysis result.

[0010] Preferably, the dimension reduction process is performed on the full parameter data model, including:

[0011] Calculating the covariance matrix of the full parameter data model;

[0012] Calculating the eigenvalues ​​and eigenvectors of the covariance matrix;

[0013] The principal components in the full parameter data model are determined according to the size of the eigenvalues, and the principal components are checked for dimensionality reduction based on the eigenvectors.

[0014] Preferably, the covariance matrix is:

[0015] Cov(m)=1 / (n-1)*(m-mean(m))′*(m-mean(m))

[0016] Where m is the full parameter data model, the covariance matrix Cov(m) is an f×f matrix, f is the number of features of the original data, and the element in the i-th row and j-th column of the covariance matrix represents the covariance between the i-th feature and the j-th feature; mean(m) is the average value of each feature, and n is the number of samples.

[0017] Preferably, after determining the principal component in the full parameter data model according to the size of the eigenvalue, the method further includes:

[0018] The sum of the eigenvalues ​​of the principal components is obtained, the proportion of the eigenvalue corresponding to each principal component to the sum of the eigenvalues ​​of the principal components is calculated, and the principal components are selected according to the calculation results to obtain the principal components in the final retained full-parameter data model.

[0019] Preferably, performing weighted K-means calculation on the principal component includes:

[0020] Select k data in the principal components as center points for data distribution, wherein the center points at least include data types corresponding to normal operation of equipment and problems with equipment;

[0021] The weight is calculated according to the allocation result, and iterative update is performed according to a preset maximum number of iterations to obtain a first fault condition of the navigation and positioning of the device under test.

[0022] Preferably, judging the operating state of the device under test based on the full parameter data model includes:

[0023] Determine the corresponding fault type according to the navigation and positioning method of the device under test;

[0024] It is determined whether there is data corresponding to the fault type in the full parameter data model according to the fault type, and a warning is issued when corresponding data exists.

[0025] Preferably, a credibility evaluation index is set to judge the validity of the decision analysis result, and the credibility evaluation index at least includes: false alarm rate, missed alarm rate, accuracy rate, recall rate and precision rate.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The present invention establishes a full-parameter data model for navigation and positioning of mine equipment, obtains characteristic values ​​through dimensionality reduction processing, and uses a weighted K-means algorithm to discriminate, warn and make decisions on navigation and positioning data. At the same time, the accuracy of the results is judged according to evaluation indicators to improve the credibility of navigation and positioning data. The present invention prominently solves the technical problem that the authenticity and reliability of navigation data cannot be determined when navigating mine equipment in the prior art. At the same time, the equipment body operation status can be adjusted in time according to the decision analysis results to improve the equipment operation efficiency, tunnel forming quality and mining safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 The present invention provides a flowchart of a method for determining the credibility of navigation and positioning system data. DETAILED DESCRIPTION

[0030] In conjunction with the drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention are clearly and completely described. 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 implementation methods obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size should fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention. It should be noted that in this specification, relational terms such as first and second are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.

[0032] At present, the existing technical inventions are mainly concentrated on different implementation methods of navigation and positioning of equipment in mines, such as inertial navigation, visual navigation, lidar positioning, etc., as well as multi-sensor combined positioning methods, combined positioning algorithms, precision improvement measures, etc. Through the combination of different positioning methods, the posture acquisition and positioning of equipment in mines can be realized.

[0033] The prior art inventions are only about the realization of attitude acquisition and navigation positioning, and mainly solve the problem of whether the positioning function exists, but do not evaluate and confirm the accuracy and credibility of the navigation positioning data, as well as the impact of the data on the subsequent operation of the equipment. If the navigation positioning data is not analyzed and tested in a timely manner, there may be inaccurate navigation positioning data, long-term non-updates, or even data errors that cause the equipment to follow the navigation positioning data for mining, resulting in deviation or penetration of adjacent tunnels, which is of great significance to the safety of equipment mining, the quality and stability of tunnel forming, and the safety of coal mine production.

[0034] Therefore, in response to the above problems, this patent proposes a navigation and positioning system data credibility decision method to judge, detect and analyze the navigation and positioning data in the equipment mining process, identify invalid data and erroneous data, and issue early warnings on the equipment operation status.

[0035] Each embodiment of the present invention takes the inertial navigation positioning method as an example, and other navigation positioning methods can also be used as the core for research. The present invention is not limited to this, and the present invention is applicable to any mine equipment navigation positioning method.

[0036] like Figure 1 As shown, an embodiment of the present invention provides a flowchart of a navigation and positioning system data credibility decision method, comprising the following steps:

[0037] S1: Determine the navigation and positioning mode of the device under test, obtain parameter data of the corresponding navigation and positioning mode, and establish a full parameter data model based on the parameter data.

[0038] In this embodiment, the inertial navigation parameter data is:

[0039] nav=[α β γ lon lat alt]

[0040] Where α is the equipment heading angle, β is the equipment pitch angle, γ is the equipment roll angle, lon is the longitude measured by inertial navigation, lat is the latitude measured by inertial navigation, and alt is the altitude measured by inertial navigation.

[0041] The positional parameter data is:

[0042] loc=[XYZ rld hd L]

[0043] Where X, Y, and Z are the three-dimensional coordinates of the equipment, rld is the left and right deviation of the fuselage, hd is the height deviation of the fuselage, and L is the fuselage footage.

[0044] The basic parameter data of tunnel and equipment are:

[0045] bas=[WH w I]

[0046] Where W is the tunnel width, H is the tunnel height, w is the equipment width, and I is the equipment cutting current.

[0047] Therefore, the full parameter data model under inertial navigation positioning is:

[0048] m=[nav loc bas]=[α β γ lon lat alt XYZ rld hd LWH w I]

[0049] S2: Perform dimensionality reduction processing on the full parameter data model to retain the principal components in the full parameter data model.

[0050] Optionally, the full-parameter data model is subjected to dimensionality reduction processing, including: calculating the covariance matrix of the full-parameter data model; calculating the eigenvalues ​​and eigenvectors of the covariance matrix; determining the principal components in the full-parameter data model according to the sizes of the eigenvalues, and performing dimensionality reduction verification on the principal components based on the eigenvectors.

[0051] In this embodiment, a principal component analysis method is used to retain the principal components of the full parameter data model, so as to retain as much original data information as possible while maintaining the effect of dimensionality reduction.

[0052] In this embodiment, the covariance matrix Cov(m) of the full parameter data model m is calculated as:

[0053] Cov(m)=1 / (n-1)*(m-mean(m))′*(m-mean(m))

[0054] Among them, the covariance matrix Cov(m) is an f×f matrix, f is the number of features of the original data, and the element in the i-th row and j-th column of the covariance matrix represents the covariance between the i-th feature and the j-th feature; mean(m) is the average value of each feature, and n is the number of samples.

[0055] Perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors. The eigenvalue indicates the importance of the eigenvector, and the eigenvector indicates the projection direction of the data in the new feature space. The calculation formula is as follows:

[0056] Cov(m)*v=λ*v

[0057] In the formula, v is the eigenvector and λ is the eigenvalue.

[0058] In this embodiment, according to the size of the eigenvalues, 90% of the eigenvalues ​​with the largest values ​​are selected as principal components, so that the selected principal components can retain the original data information as much as possible.

[0059] In this embodiment, the original data is projected into a new feature space, and the dimension of the new feature space is equal to the number of principal components. The projected data is converted back to the original space for data reconstruction to check the degree of information loss in dimensionality reduction. The steps are as follows:

[0060] Recalculate the projection matrix: P = U(:,1:K)

[0061] Reconstructed data: Z = m*P*P'

[0062] Among them, P is the projection matrix, U is the eigenvector matrix, K is the number of selected principal components, and Z represents the reconstructed data.

[0063] Optionally, after determining the principal components in the full parameter data model according to the size of the eigenvalues, it also includes: obtaining the sum of the eigenvalues ​​of the principal components, calculating the proportion of the eigenvalue corresponding to each principal component to the sum of the eigenvalues ​​of the principal components, and filtering the principal components according to the calculation results to obtain the principal components in the full parameter data model that are finally retained.

[0064] In this embodiment, the amount of information retained after dimensionality reduction is evaluated by calculating the proportion of the variance of each principal component to the total variance. The steps are as follows:

[0065] var=(value i / sum(values))*100%

[0066] Among them, value i is the eigenvalue of the i-th principal component, and sum(values) is the sum of the eigenvalues ​​of all principal components.

[0067] According to the calculation results of this formula, the number of principal components to be retained is selected so that the principal components in the final retained full-parameter data model retain as much original data information as possible while maintaining the effect of dimensionality reduction.

[0068] S3: Establish a navigation and positioning data decision mechanism, perform weighted K-means calculation on the principal components, predict the first fault condition of the navigation and positioning of the device under test based on the calculation results, and judge the operating status of the device under test based on the full-parameter data model, obtain the second fault condition of the navigation and positioning of the device under test based on the judgment result, perform credibility decision analysis based on the first fault condition and the second fault condition, and adjust the operating status of the equipment body according to the decision analysis results.

[0069] Optionally, performing weighted K-means calculation on the principal component includes: selecting k data in the principal component as center points for data distribution, and the center points include at least data types corresponding to normal operation of the equipment and equipment problems; calculating the weight according to the distribution result, and iteratively updating according to a preset maximum number of iterations to obtain the first fault condition of the navigation positioning of the equipment under test.

[0070] In this embodiment, the weighted K-means algorithm is implemented as follows:

[0071] 1) Initialization. Select K center points as the initial cluster center points, and assign a weight value to each center point. In the present invention, the K center points are respectively types such as normal operation of equipment, various problems and failures.

[0072] 2) Assign data to the center: Assign the data points to the nearest cluster based on the distance from each data point to the center of each cluster.

[0073] 3) Calculate new cluster centers: Calculate new cluster centers based on the weights and coordinates of the data points in each cluster. Data points with larger weights contribute more to the calculation of cluster centers.

[0074] 4) Update weights: Update the weights of data points based on the distance from each data point to the newly calculated cluster center.

[0075] 5) Repeat steps 2)-4) until a stop condition is met. The stop condition in the embodiment of the present invention is that a preset maximum number of iterations is reached.

[0076] 6) Output results: The final K cluster centers and the cluster labels of each data point are the output results of the first fault condition of the weighted K-means algorithm. Through this output result, the classification of the data after dimensionality reduction is realized, and the possible faults and fault types of the equipment are predicted.

[0077] Optionally, judging the operating status of the device under test based on the full-parameter data model includes: determining a corresponding fault type according to a navigation and positioning method of the device under test; judging whether data of a corresponding fault type exists in the full-parameter data model according to the fault type, and issuing a warning when corresponding data exists.

[0078] In this embodiment, the judgment of the equipment operation state is mainly based on the data in the full parameter data model to judge whether the equipment has a second fault condition in its current operation state. The main judgment basis is:

[0079] 1) Heading angle deviation: |α-α0|=ε α Among them, α0 is the target heading angle, ε α is the heading angle deviation threshold.α ≤3°, a yellow warning is issued; when 3°≤ε α ≤5°, an orange warning is issued, and the operator is prompted to adjust the machine body to the set mining direction; when ε α ≥5°, a red warning is issued and the equipment is triggered to stop operating.

[0080] 2) Left-right deviation of the machine body: rld. When rld≤(W / 2-w / 2) / 2, a warning is issued to remind the operator to adjust the position of the machine body in the tunnel and restart mining at the center line of the tunnel.

[0081] In this embodiment, the navigation data credibility analysis decision is performed based on the comprehensive weighted K-means calculation and the operating status of the device under test. When a fault occurs in the weighted K-means calculation result or the judgment result corresponding to any part of the operating status of the device under test, it indicates that the current navigation data is inaccurate and the equipment body operation needs to be adjusted in time.

[0082] Optionally, a credibility evaluation index is set to judge the validity of the decision analysis result, and the credibility evaluation index at least includes: false alarm rate, missed alarm rate, accuracy rate, recall rate and precision rate.

[0083] In this embodiment, the credibility of the decision result is judged by the following indicators.

[0084] 1) False alarm rate: E1 = FP / (FP+TN)

[0085] 2) Missing reporting rate: E2 = FN / (FN+TN)

[0086] 3) Accuracy: Pre = TP / (TP+FP)

[0087] 4) Recall rate: Rec = TP / (TP+FN)

[0088] 5) Accuracy: Acc = (TP + TN) / (TP + TN + FN + FP)

[0089] Among them, TP is the number of positive samples identified as positive samples, TN is the number of negative samples identified as negative samples, FN is the number of positive samples identified as negative samples, and FP is the number of negative samples identified as positive samples. In this embodiment, positive samples, that is, the equipment navigation and positioning system data is operating normally; negative samples, that is, the equipment navigation and positioning system data has faults such as yaw, deviation, information error, etc.

[0090] The present invention establishes a full-parameter data model for navigation and positioning of mine equipment, obtains characteristic values ​​through dimensionality reduction processing, and uses a weighted K-means algorithm to distinguish, warn and make decisions on navigation and positioning data. At the same time, the accuracy of the results is judged according to evaluation indicators to improve the credibility of navigation and positioning data. The present invention prominently solves the technical problem that the authenticity and reliability of navigation data cannot be determined in the prior art, and can further improve equipment operation efficiency, tunnel forming quality and mining safety.

[0091] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A navigation and positioning system data credibility decision method, characterized in that: The steps include: Determine the navigation and positioning mode of the device under test, obtain parameter data of the corresponding navigation and positioning mode, and establish a full parameter data model based on the parameter data; Performing dimensionality reduction processing on the full parameter data model to retain the principal components in the full parameter data model; A navigation and positioning data decision mechanism is established, and a weighted K-means calculation is performed on the principal components. The first fault condition of the navigation and positioning of the equipment under test is predicted according to the calculation results. Meanwhile, the operating status of the equipment under test is judged based on the full-parameter data model. The second fault condition of the navigation and positioning of the equipment under test is obtained according to the judgment result. A credibility decision analysis is performed based on the first fault condition and the second fault condition, and the operating status of the equipment fuselage is adjusted according to the decision analysis result.

2. The navigation and positioning system data credibility decision method according to claim 1, characterized in that: The dimensionality reduction processing of the full parameter data model includes: Calculating the covariance matrix of the full parameter data model; Calculating the eigenvalues ​​and eigenvectors of the covariance matrix; The principal components in the full parameter data model are determined according to the size of the eigenvalues, and the principal components are checked for dimensionality reduction based on the eigenvectors.

3. The navigation and positioning system data credibility decision method according to claim 2, characterized in that: The covariance matrix is: Cov(m)=1 / (n-1)*(m-mean(m))′*(m-mean(m)) Where m is the full parameter data model, the covariance matrix Cov(m) is an f×f matrix, f is the number of features of the original data, and the element in the i-th row and j-th column of the covariance matrix represents the covariance between the i-th feature and the j-th feature; mean(m) is the average value of each feature, and n is the number of samples.

4. The navigation and positioning system data credibility decision method according to claim 2, characterized in that: After determining the principal component in the full parameter data model according to the size of the eigenvalue, the method further includes: The sum of the eigenvalues ​​of the principal components is obtained, the proportion of the eigenvalue corresponding to each principal component to the sum of the eigenvalues ​​of the principal components is calculated, and the principal components are selected according to the calculation results to obtain the principal components in the final retained full-parameter data model.

5. The method for determining the credibility of navigation and positioning system data according to claim 1, characterized in that: The weighted K-means calculation of the principal component includes: Select k data in the principal components as center points for data distribution, wherein the center points at least include data types corresponding to normal operation of equipment and problems with equipment; The weight is calculated according to the allocation result, and iterative update is performed according to a preset maximum number of iterations to obtain a first fault condition of the navigation and positioning of the device under test.

6. The navigation and positioning system data credibility decision method according to claim 1, characterized in that: Determining the operating state of the device under test based on the full parameter data model includes: Determine the corresponding fault type according to the navigation and positioning method of the device under test; It is determined whether there is data corresponding to the fault type in the full parameter data model according to the fault type, and a warning is issued when corresponding data exists.

7. The navigation and positioning system data credibility decision method according to claim 1, characterized in that: Credibility evaluation indicators are set to judge the validity of the decision analysis results, and the credibility evaluation indicators at least include: false alarm rate, missed alarm rate, accuracy rate, recall rate and precision rate.

Images