Catenary multi-parameter diagnostic analysis method based on standard state

Abstract

The invention belongs to the field of electrified rail transit overhead contact systems, and provides an overhead contact system diagnosis method and system based on a standard state, a storage medium and electronic equipment. The method comprises the following steps: constructing standard state data based on contact network detection parameters; obtaining any parameter data in the pantograph-catenary comprehensive detection period of the catenary, carrying out single parameter analysis based on the standard state data, and judging a stationary span and an abnormal span; performing multi-parameter association analysis on the abnormal span to obtain all abnormal parameter association data in the abnormal span; and substituting the abnormal parameter associated data into an interaction item regression model to obtain a dependent variable value of the interaction item regression model, and speculating the cause of the abnormality of the overhead line system. According to the method, diagnostic analysis and anomaly recognition of the parameters can be automatically completed, the possibility of manual intervention and misjudgment is reduced, and the operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of electrified rail transit overhead contact network, and in particular relates to an overhead contact network diagnosis method, system, storage medium and electronic equipment based on a standard state. Background Art

[0002] In recent years, with the application of technologies such as big data, cloud computing, and artificial intelligence, the detection, monitoring, and analysis technology of electrified railways has entered the intelligent stage. The defect diagnosis methods for various detection parameters of electrified railway contact networks currently mainly use threshold analysis, image-assisted analysis, repeated comparison analysis, graphical analysis, data comprehensive analysis, and data fusion analysis.

[0003] 1. The threshold analysis method finds "local" or "point" problems by judging the over-limit data; 2. Image-assisted analysis and data fusion analysis: After the defects are analyzed at the threshold, the defect type, location, and degree of danger are further confirmed using images, videos, and data from other detection devices of the 6C system; 3. Repeated comparative analysis and graphical analysis are to observe the distribution, repeatability and change trend of detection parameters in time series; 4. Comprehensive data analysis is to construct the contact network dynamic performance index (CDI) through multiple parameters such as pull-out value, contact line height (conductor height), bow-net contact force, hard points, arcing, etc., to evaluate the current collection quality of the bow-net as a whole.

[0004] None of the above methods use statistical methods to perform diagnostic analysis on the parameter data itself, and are unable to predict the mathematical logic changes and anomalies of the detection parameters, and further unable to predict the failure trend of the equipment, and achieve early warning of failures and preventive maintenance.

[0005] CN114660378A discloses a comprehensive contact network diagnosis method based on multi-source detection parameters, which uses multiple multi-dimensional detection data to conduct a comprehensive diagnosis and evaluation of the contact network. However, there is no further analysis of abnormal data, and the source of the abnormality cannot be determined, let alone predicted based on existing data. Summary of the invention

[0006] The present invention aims at the deficiencies in the prior art and provides a contact network diagnosis method based on a standard state.

[0007] The present invention is implemented by the following technical solution: comprising: Based on the catenary detection parameters, standard state data is constructed; Obtain any parameter data during the comprehensive detection of the catenary and bow network, perform single parameter analysis based on the standard state data, and determine the stable span and abnormal span; Performing multi-parameter correlation analysis on the abnormal span to obtain all abnormal parameter correlation data within the abnormal span; Substitute the abnormal parameter association data into the interactive regression model to obtain the dependent variable value of the interactive regression model, and infer the cause of the abnormality of the contact network.

[0008] Further, acquiring contact network detection data, preprocessing the contact network detection data, and obtaining the arbitrary parameter data; A single-sample KS test is performed on the arbitrary parameter data to obtain the stable span and the abnormal span.

[0009] Furthermore, the influence of positioning error on the contact network detection parameters is eliminated.

[0010] Further, based on the standard state data, the proportion of the arbitrary parameter data not distributed in the area where the standard state is located is counted, and the data of the arbitrary parameter data not distributed in the area where the standard state is located is regarded as an abnormal data point; The span where the abnormal data points are located is the abnormal span, and the other spans are stable spans.

[0011] Furthermore, the arbitrary parameter data are analyzed by multivariate analysis of variance, paired sample T test and K related sample test method, and the arbitrary parameter detection data are compared in the abnormal span.

[0012] Furthermore, the two independent variables of the interaction regression model were determined; The interaction between the two independent variables is calculated, the two independent variables in the detection parameters are substituted into the interaction term regression model, and the change amount of the dependent variable is calculated.

[0013] Furthermore, the larger the absolute value of the dependent variable is, the stronger the interaction between the two independent variables is.

[0014] In a second aspect, an embodiment of the present invention provides a contact network diagnosis system based on a standard state, characterized in that it includes: Data acquisition module, used to collect contact network detection parameters; A data processing module is used to pre-process the contact network detection parameters to obtain arbitrary parameter data; perform a single sample KS test on the arbitrary parameter data to obtain a stable span and an abnormal span; analyze and test multiple parameters in the abnormal span to obtain all abnormal parameter data in the abnormal span; The modeling and analysis module is used to substitute the abnormal parameter data into the interaction term regression model to infer the cause of the abnormality of the contact network.

[0015] Based on the same inventive concept, an embodiment of the present invention further provides a computer storage medium, wherein the storage medium stores computer executable instructions, and the computer executable instructions, when executed, can implement the aforementioned contact network diagnosis method.

[0016] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, including: Processor and memory; The processor is coupled to a memory; The processor is used to read and execute the program or instruction stored in the memory, so that the device executes the aforementioned method.

[0017] Compared with the prior art, the present invention has the following advantages: 1. Participate in comprehensive diagnosis of abnormalities: The method of the present invention can not only discover the "local" or "point" problems of abnormalities, but also reveal the mathematical logic changes and abnormalities of parameter data, and realize comprehensive and in-depth diagnostic analysis.

[0018] 2. Accurate defect warning: By comparing the difference between actual test data and standard state, abnormal parameters are automatically identified and warning information is given, which improves the accuracy and timeliness of warning.

[0019] 3. Significantly improve operation and maintenance efficiency: The method of the present invention can automatically complete parameter diagnosis analysis and abnormality identification, reduce the possibility of manual intervention and misjudgment, and improve operation and maintenance efficiency.

[0020] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] 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 or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are 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.

[0022] Figure 1 A flow chart of a method for diagnosing a contact network based on a standard state is shown; Figure 2 An example graph of abnormal data points in a single parameter diagnostic analysis is shown; Figure 3 An example graph of abnormal data points in multi-parameter association analysis is shown. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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.

[0024] The scheme of the present invention uses the single-sample KS test in statistics to perform single-parameter diagnostic analysis on the basis of the constructed standard state data of the contact network detection parameters to determine whether the contact network is in an abnormal state; and uses multivariate analysis of variance, paired sample T test and K-related sample test method to perform multi-parameter diagnostic analysis to infer the causes of contact network defects or abnormal states.

[0025] Specifically, Figure 1 As shown, the present invention discloses a method for multi-parameter diagnosis and analysis of a contact network based on a standard state, comprising the following steps: S1: Obtain the test data, pre-process the test data, and perform a single-sample KS test, i.e., Kolmogorov-Smirnov test, on the obtained arbitrary parameter data to obtain the stable span and abnormal span of the overhead contact network; The specific steps of step S1 are: S11: Acquire detection data, construct standard state data, eliminate the influence of positioning error from the detection data, and obtain arbitrary parameter data; S12: Perform a single-sample KS test on any parameter data, count the proportion of the arbitrary parameter data not distributed in the area where the standard state is located, and regard the data of the arbitrary parameter data not distributed in the area where the standard state is located as an abnormal data point.

[0026] The test steps of the one-sample KS test are as follows: (1) Propose the hypothesis that the test data follows the standard data distribution; (2) Set the significance level to 0.05. The significance level is the probability of making an error when estimating the population parameter to fall within a certain span. (3) Calculate the empirical distribution function Fn(χ), the distribution function of a discrete random variable with values ​​X1, X2, …, Xn with equal probability 1 / n; (4) Calculate the theoretical distribution function F(χ) based on the assumed distribution; (5) Calculate the KS test statistic Dn. Dn represents the statistic in the KS test, which measures the maximum difference between the cumulative distribution function of the sample data and the theoretical distribution. (6) Find the critical value Dα. In the KS test, by comparing Dn and Dα, it can be determined whether to reject the null hypothesis, that is, whether the sample data comes from the assumed theoretical distribution; (7) If Dn>Dα, then reject the null hypothesis that there are abnormal data points; otherwise accept the null hypothesis that there are no abnormal data points.

[0027] S13: According to the proportion of abnormal data points, determine whether each span of the contact network is an abnormal span, a risk span or a normal span. If the abnormal data accounts for more than 50%, it is regarded as an abnormal span; if the abnormal data is between 30% and 50%, it is regarded as a risk span; if the abnormal data is less than 30%, delete the abnormal data points and determine it as a normal span. The span with abnormal data points has been diagnosed as an abnormal span, and the rest of the spans are stable spans.

[0028] For example, Figure 2 As shown, the detection data 4 between span 1 and span 2 has data points that are not distributed in the standard state area, and abnormal data points appear. The span where the abnormal data points are located is the abnormal span, and the other spans are stable spans.

[0029] S2: Calculate the data correlation degree with multiple detection parameters within the abnormal span.

[0030] The specific steps of step S2 are: S21: Obtain the abnormal cross-data or contact network parameter defects diagnosed in step S1; S22: In the abnormal span, multivariate analysis of variance, paired sample T test and K related sample test method are used to calculate the difference, variance, range, change rate, etc. of multiple detection parameter data compared with the standard data at the defect position in S21 to obtain the abnormal parameter data in the abnormal span.

[0031] The paired sample T test is a statistical method used to test whether there is a significant difference between the means of two related samples. The K-related sample test method, namely the chi-square test, is to measure the degree of deviation between the actual observed value and the theoretical inference value of the statistical sample. The degree of deviation between the actual observed value and the theoretical inference value determines the size of the chi-square value.

[0032] S3: Establish and use the interactive regression model to infer the main reasons for abnormal span or defects.

[0033] The specific steps of step S3 are: S31: Establish and use the interactive regression model to calculate the dependent variable values ​​of the two independent variables through the interactive regression model; The construction and calculation steps of the interaction regression model are as follows: (1) Determine the two independent variables of the interaction regression model, such as hard point and guide height, and the interaction term is X1*X2; (2) Establish the interactive regression model Y=β0+β1*X1+β2*X2+β3*X1*X2+ε; Among them, Y is the dependent variable, X1 and X2 are independent variables, here we take the hard point and derivative height; β0 is the intercept term, β1 and β2 are called net regression coefficients, β3 is the interaction term coefficient, which indicates the influence of the interaction between independent variables X1 and X2 on Y, and ε is the error term.

[0034] (3) Calculate the interaction between hard point X1 and guide height X2, substitute hard point X1 and guide height X2 in the detection parameters into the interaction regression model, and then calculate the change of the dependent variable Y.

[0035] S32: Based on the size of the dependent variable Y, infer the main cause of the abnormal span or defect. The larger the absolute value of Y, the stronger the interaction between the two independent variables.

[0036] For example, Figure 3 As shown, the detection data in the marked area diagnoses hard point anomalies. Through multi-parameter correlation analysis, it is associated with the excessive height of the guide at the same position. It can be inferred that the main cause of the hard point anomaly is the excessive height of the guide.

[0037] In another embodiment of the present invention, a computer storage medium is provided, wherein the storage medium stores computer executable instructions, and the computer executable instructions implement the aforementioned contact network diagnosis method when executed.

[0038] In another embodiment, the present invention provides an electronic device, comprising: Processor and memory; The processor is coupled to a memory; The processor is used to read and execute the program or instruction stored in the memory, so that the device executes the aforementioned method.

[0039] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for diagnosing a contact network based on a standard state, characterized in that: include: Based on the catenary detection parameters, standard state data is constructed; Obtain any parameter data during the comprehensive detection of the catenary and bow-net, perform single parameter analysis based on the standard state data, and determine the stable span and abnormal span; Performing multi-parameter correlation analysis on the abnormal span to obtain all abnormal parameter correlation data within the abnormal span; Substitute the abnormal parameter association data into the interactive regression model to obtain the dependent variable value of the interactive regression model, and infer the cause of the abnormality of the contact network.

2. The diagnostic method according to claim 1, characterized in that Obtain any parameter data during the comprehensive detection of the contact network bow and catenary, perform single parameter analysis based on the standard state data, and judge the stable span and abnormal span, including: Acquire contact network detection data, pre-process the contact network detection data, and obtain the arbitrary parameter data; A single-sample KS test is performed on the arbitrary parameter data to obtain the stable span and the abnormal span.

3. The diagnostic method according to claim 2, characterized in that Obtain contact network detection parameters and preprocess the contact network detection parameters. include, Eliminate the influence of positioning error on the contact network detection parameters.

4. The diagnostic method according to claim 2, characterized in that: Performing a single sample KS test on the arbitrary parameter data to obtain the stable span and the abnormal span includes: Based on the standard state data, the proportion of the arbitrary parameter data not distributed in the area where the standard state is located is counted, and the data where the arbitrary parameter data is not distributed in the area where the standard state is located is regarded as an abnormal data point; The span where the abnormal data points are located is the abnormal span, and the other spans are stable spans.

5. The diagnostic method according to claim 1, characterized in that: Perform multi-parameter correlation analysis on the abnormal span to obtain all abnormal parameter correlation data in the abnormal span, including: The arbitrary parameter data are analyzed by multivariate analysis of variance, paired sample T test and K related sample test method, and the difference between the arbitrary parameter detection data and the data in the standard state in the abnormal span is calculated. According to the difference, the abnormal parameter data in the abnormal span is obtained.

6. The diagnostic method according to claim 1, characterized in that: Substituting the abnormal parameter association data into the interactive regression model to obtain the dependent variable value of the interactive regression model, including: Determine the two independent variables of the interaction regression model; The interaction between the two independent variables is calculated, the two independent variables in the detection parameters are substituted into the interaction term regression model, and the change amount of the dependent variable is calculated.

7. The diagnostic method according to claim 6, characterized in that The dependent variable value of the interactive regression model is obtained, and the causes of abnormal contact network are speculated, including: The larger the absolute value of the dependent variable is, the stronger the interaction between the two independent variables is.

8. A contact network diagnostic system based on standard state, characterized in that: include, Data acquisition module, used to collect contact network detection parameters; A data processing module is used to pre-process the contact network detection parameters to obtain arbitrary parameter data; perform a single sample KS test on the arbitrary parameter data to obtain a stable span and an abnormal span; analyze and test multiple parameters in the abnormal span to obtain all abnormal parameter data in the abnormal span; The modeling and analysis module is used to substitute the abnormal parameter data into the interaction term regression model to infer the cause of the abnormality of the contact network.

9. A computer storage medium, characterized in that: The storage medium stores computer executable instructions, which, when executed, implement a contact network diagnosis method based on a standard state as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: Processor and memory; The processor is coupled to a memory; The processor is used to read and execute the program or instruction stored in the memory, so that the device executes a contact network diagnostic method based on a standard state as described in any one of claims 1-7.

Images