Digital track circuit health state analysis method, device, equipment and medium

Abstract

The invention relates to the technical field of railway train operation control, and provides a digital track circuit health state analysis method, device, equipment and medium, and the method comprises the steps: carrying out the deviation standardization processing of the electrical parameters of a track section; performing parameter category labeling on the standardized electrical parameters, and generating a data instance of the track section according to the standardized electrical parameters and the electrical parameter categories; determining a training data set and a test data set of the track section according to the data instance, training the support vector machine model according to the training data set, and performing parameter optimization on the trained support vector machine model by using the test data set; using the support vector machine model after parameter tuning to classify preset electrical parameters to be analyzed to obtain category instances of the track section; and determining the risk state of the digital track circuit according to the distance between the category instance and the decision plane. According to the invention, the accuracy of health state analysis of the digital track circuit can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of railway train operation control, and in particular to a method, device, equipment and medium for analyzing the health status of a digital track circuit. Background Art

[0002] Digital track circuits achieve electrical insulation between adjacent sections through signals of different frequencies. They are used for train position occupancy checks and information transmission from trains to the ground. They are usually composed of indoor and outdoor equipment. Indoor electronic equipment adopts a redundant design, and a single-channel failure does not affect the system function. However, outdoor equipment usually does not have a redundant design because its working environment is complex and prone to failures, resulting in overall system failure. Therefore, the health risk assessment and measurement of track circuits is the key to operation and maintenance, and thus the health status of digital track circuits needs to be analyzed.

[0003] The existing digital track circuit health status analysis technology can only give a binary conclusion on whether each test parameter meets the requirements. The utilization rate of measurement data is insufficient, and it is difficult to measure the health risk situation of the entire equipment. The inaccurate analysis of the causes of abnormal equipment performance indicators leads to the adoption of incorrect response measures, resulting in low accuracy in the digital track circuit health status analysis. Summary of the invention

[0004] In view of the above problems, embodiments of the present invention provide a method, apparatus, device and medium for analyzing the health status of a digital track circuit.

[0005] In a first aspect, an embodiment of the present invention provides a method for analyzing the health status of a digital track circuit, comprising:

[0006] Acquiring electrical parameters of a track section in a digital track circuit, performing deviation standardization processing on the electrical parameters, and obtaining standardized electrical parameters;

[0007] Perform parameter category labeling on the standardized electrical parameters to obtain electrical parameter categories, and generate data instances of track sections according to the standardized electrical parameters and the electrical parameter categories;

[0008] Determine a training data set and a test data set of the track section according to the data example, train a pre-configured support vector machine model according to the training data set, and use the test data set to perform parameter tuning on the trained support vector machine model;

[0009] Using the support vector machine model after parameter tuning to classify the preset electrical parameters to be analyzed, and obtain the category instance of the track section corresponding to the electrical parameters to be analyzed;

[0010] The risk state of the digital track circuit is determined according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

[0011] According to an embodiment of the present invention, the electrical parameters include resonant frequency shift, receiving voltage, transmitting voltage, rail voltage and shunt resistance.

[0012] According to an embodiment of the present invention, the measurement of the resonance frequency shift includes the following steps:

[0013] Turn off the power supply of the tested section in the track section of the digital track circuit and the indoor sending units of the two sections adjacent to the tested section;

[0014] After turning off the power, connect the two terminals in the connection room of the preset tuning unit box to the signal amplifier and signal generator respectively, and connect the vehicle antenna of the digital track circuit to the spectrum analyzer;

[0015] Using the signal generator to generate a frequency sweep signal, and amplifying the frequency sweep signal through the signal amplifier;

[0016] After the amplified frequency sweep signal passes through the resonant circuit of the tuning unit, the frequency corresponding to the maximum amplitude signal is read by the spectrum analyzer as the resonant frequency of the resonant circuit;

[0017] The absolute difference between the resonant frequency and the design frequency corresponding to the measured section is used as the resonant frequency offset of the resonant circuit.

[0018] According to an embodiment of the present invention, performing deviation normalization processing on the electrical parameters to obtain standardized electrical parameters includes:

[0019] Convert the parameter value corresponding to each electrical parameter into a feature vector;

[0020] The deviation of each electrical parameter is normalized according to the characteristic vector by the following formula to obtain the normalized electrical parameter:

[0021]

[0022] Among them, x ′ is the standardized electrical parameter, x is the eigenvalue corresponding to the electrical parameter in the eigenvector, and x min is the minimum value in the eigenvector, x max is the maximum value in the eigenvector.

[0023] According to an embodiment of the present invention, the step of generating a data instance of a track segment according to the standardized electrical parameters and the electrical parameter categories includes:

[0024] Matching the standardized electrical parameters with the electrical parameter categories to obtain a corresponding relationship;

[0025] When the electrical parameter category in the corresponding relationship is a preset first category, dividing the track section into a positive instance;

[0026] When the electrical parameter category in the corresponding relationship is a preset second category, classifying the track section as a negative instance;

[0027] The positive instances and the negative instances are aggregated into data instances of a track segment.

[0028] According to an embodiment of the present invention, determining the risk state of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model includes:

[0029] extracting a first distance between a positive instance in the category instance and the decision plane;

[0030] extracting a second distance between a negative instance in the category instance and the decision plane;

[0031] When the first distance is greater than a preset distance threshold, determining that the risk status of the digital track circuit is low risk;

[0032] When the second distance is greater than a preset distance threshold, it is determined that the risk state of the digital track circuit is high risk.

[0033] According to an embodiment of the present invention, before training the pre-configured support vector machine model according to the training data set, the method further includes:

[0034] Determine the penalty parameter according to the tolerance of the classifier in the preset support vector machine model;

[0035] Select the Gaussian kernel function according to the preset classification attributes;

[0036] Determining model parameters of the support vector machine model according to the penalty parameter and the Gaussian kernel function;

[0037] The support vector machine model is configured by the model parameters.

[0038] In a second aspect, an embodiment of the present invention provides a digital track circuit health status analysis device, characterized in that it includes:

[0039] A deviation standardization processing module is used to obtain electrical parameters of a track section in a digital track circuit, and perform deviation standardization processing on the electrical parameters to obtain standardized electrical parameters;

[0040] A data instance generation module, used for labeling the standardized electrical parameters by parameter category to obtain the electrical parameter category, and generating a data instance of the track section according to the standardized electrical parameters and the electrical parameter category;

[0041] A parameter tuning module, used to determine a training data set and a test data set of a track section according to the data instance, train a pre-configured support vector machine model according to the training data set, and perform parameter tuning on the trained support vector machine model using the test data set;

[0042] A track section category analysis module is used to classify preset electrical parameters to be analyzed using the support vector machine model after parameter tuning, and obtain a category instance of the track section corresponding to the electrical parameters to be analyzed;

[0043] The risk status determination module is used to determine the risk status of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

[0044] In a third aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the multi-fuel cell system fan control method described in the above aspect.

[0045] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the digital track circuit health status analysis method described in the above aspects.

[0046] Compared with the prior art, the above technical solution of the present invention has the following beneficial effects:

[0047] The embodiments of the present invention can comprehensively measure the electrical performance indicators of digital track circuits by collecting electrical parameters, detect the resonant frequency of the circuit tuning unit, increase the detection range of fault causes, and improve the detection quality; by using the support vector machine algorithm, quantitative characterization of the health risk status of digital track circuits can be achieved, the utilization rate of data can be improved, and the health risk status of the entire equipment can be measured. Therefore, the digital track circuit health status analysis method, device, equipment and medium proposed in the present invention can solve the problem of low accuracy when performing digital track circuit health status analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] 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 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.

[0049] Figure 1 A flowchart showing a method for analyzing the health status of a digital track circuit according to a first embodiment of the present invention is shown;

[0050] Figure 2 A schematic diagram showing the measurement of the resonant frequency shift according to the first embodiment of the present invention is shown;

[0051] Figure 3 A schematic diagram showing the measurement of rail surface voltage according to the first embodiment of the present invention is shown;

[0052] Figure 4 A schematic diagram showing a method for measuring shunt resistance according to a first embodiment of the present invention is shown;

[0053] Figure 5 A functional module diagram of a digital track circuit health status analysis device according to a third embodiment of the present invention is shown;

[0054] Figure 6 A schematic diagram of the composition structure of an electronic device for implementing the digital track circuit health status analysis method according to the fourth embodiment of the present invention is shown. DETAILED DESCRIPTION

[0055] The present disclosure is further described below in conjunction with the embodiments shown in the accompanying drawings.

[0056] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0057] The present invention proposes a digital track circuit health status analysis method based on railway train operation control technology. Based on the support vector machine theory and combined with the multi-category classification method, a digital track health status detection model is built to achieve quantitative characterization of the health risk status of the digital track circuit. Compared with traditional methods, the digital track circuit health status analysis technology is more efficient and reduces human subjectivity, and has great potential and application prospects in railway train operation control interpretation applications.

[0058] Embodiment 1

[0059] like Figure 1 As shown, the present invention proposes a method for analyzing the health status of a digital track circuit, comprising the following steps:

[0060] S1. Acquire electrical parameters of a track section in a digital track circuit, perform deviation standardization processing on the electrical parameters, and obtain standardized electrical parameters.

[0061] In the embodiment of the present invention, the digital track power realizes the electrical insulation of adjacent sections through different frequencies, which is used for train position occupancy inspection and information transmission from the train to the ground. It is usually composed of indoor and outdoor equipment. The indoor equipment is mainly electronic equipment, including encoding, amplification, input, output and other units; the outdoor equipment is mainly electrical equipment, including tuning units, signal injection equipment and cables, etc. In order to more comprehensively measure the electrical performance indicators of the digital track circuit, the resonant frequency of the circuit tuning unit can be monitored to increase the detection range of the fault cause and improve the detection quality.

[0062] In the embodiment of the present invention, the electrical parameters include resonant frequency offset, receiving voltage, transmitting voltage, rail voltage and shunt resistance.

[0063] In detail, measuring electrical parameters, including resonant frequency shift, can more comprehensively measure the electrical performance indicators of the digital track circuit, and the resonant frequency shift refers to the change of the resonant frequency relative to its ideal value or initial value in the resonant circuit.

[0064] In an embodiment of the present invention, the measurement of the resonant frequency shift includes the following steps:

[0065] Turn off the power supply of the tested section in the track section of the digital track circuit and the indoor sending units of the two sections adjacent to the tested section;

[0066] After turning off the power, connect the two terminals in the connection room of the preset tuning unit box to the signal amplifier and signal generator respectively, and connect the vehicle antenna of the digital track circuit to the spectrum analyzer;

[0067] Using the signal generator to generate a frequency sweep signal, and amplifying the frequency sweep signal through the signal amplifier;

[0068] After the amplified frequency sweep signal passes through the resonant circuit of the tuning unit, the frequency corresponding to the maximum amplitude signal is read by the spectrum analyzer as the resonant frequency of the resonant circuit;

[0069] The absolute difference between the resonant frequency and the design frequency corresponding to the measured section is used as the resonant frequency offset of the resonant circuit.

[0070] Specifically, turn off the power of the indoor transmitting unit in the tested section and the two adjacent sections, connect the two terminals in the tuning unit box connected to the signal amplifier and signal generator respectively, set up the track circuit vehicle antenna above the track and connect it to the spectrum analyzer, use the signal generator to generate a sweep signal, the sweep signal frequency should cover all frequencies of the digital track circuit, after passing through the resonant circuit of the tuning unit, read the frequency of the maximum amplitude signal on the spectrum analyzer, which is the resonant frequency of the resonant circuit, and then consult the product design documents to get the design frequency of the section, the difference between the two is the absolute value of the resonant frequency offset of the sending / receiving resonant circuit, such as Figure 2 As shown in the figure, it is a schematic diagram for measuring the resonant frequency offset. The two terminals in the tuning unit box are connected to the signal amplifier and the signal generator respectively. The signal generator generates a swept frequency signal, which is amplified by the signal amplifier. Then, the frequency of the maximum amplitude signal read on the spectrum analyzer is the resonant frequency of the resonant circuit. The resonant frequency offset is obtained based on the difference between the absolute value of the resonant frequency and the design frequency.

[0071] Specifically, the receiving voltage refers to the voltage amplitude of the outdoor signal received by the indoor receiving unit, which is measured at the corresponding terminal of the indoor branch cabinet with a multimeter; the sending voltage refers to the voltage amplitude of the signal sent by the indoor sending unit to the outdoor equipment, which is measured at the corresponding terminal of the indoor branch cabinet with a multimeter; the rail surface voltage refers to the voltage amplitude between the two points where the tuning unit injects the signal into the rail, which is divided into the sending end and the receiving end, such as Figure 3 As shown in the figure, it is a schematic diagram of measuring the rail surface voltage. By tuning the voltage amplitude between two points of the unit, U1 is the rail surface voltage sent by a track section, and U2 is the rail surface voltage received by the adjacent section corresponding to the track section, so as to measure the opposite voltage; the shunt resistance is the resistance value that can just make the track relay fall, which can be measured by the volt-ampere method, such as Figure 4 As shown in the figure, it is a schematic diagram for measuring shunt resistance. First, set the rheostat to a larger resistance value. At the same time, confirm that the section is in an idle state on the indoor interlocking terminal. Gradually lower the resistance value of the rheostat. When the section just changes from idle to occupied, record the readings of the voltmeter and ammeter and calculate the resistance value of the rheostat. Add the resistance value of the wire to get the shunt resistance.

[0072] Furthermore, the collected electrical parameters have different characteristic dimensions. The characteristic dimensions are different and the numerical differences are also large, which affects the convergence speed and classification accuracy of the model. Therefore, it is necessary to remove the influence of the dimensions.

[0073] In the embodiment of the present invention, the standardized electrical parameter refers to an electrical parameter that maps all features to the interval [0, 1] by using deviation standardization, and the deviation standardization process (also called normalization) is to scale the data to a specific range.

[0074] In the embodiment of the present invention, the step of performing deviation standardization processing on the electrical parameters to obtain standardized electrical parameters includes:

[0075] Convert the parameter value corresponding to each electrical parameter into a feature vector;

[0076] The deviation of each electrical parameter is normalized according to the characteristic vector by the following formula to obtain the normalized electrical parameter:

[0077]

[0078] Among them, x ′ is the standardized electrical parameter, x is the eigenvalue corresponding to the electrical parameter in the eigenvector, and x min is the minimum value in the eigenvector, x max is the maximum value in the eigenvector.

[0079] In detail, the parameter values ​​collected for each electrical parameter, such as the resonant frequency offset, receiving voltage, transmitting voltage, rail voltage, and shunt resistance in the electrical parameters, are converted into the characteristic vector corresponding to each electrical parameter. For example, the characteristic vector corresponding to the resonant frequency offset is {x1, x2, …, x n}, each electrical parameter corresponds to an n-dimensional eigenvector, and the deviation of the parameter value in each electrical parameter is normalized based on the eigenvector. For example, the eigenvector corresponding to the resonant frequency offset is selected as {x1, x2, …, x n The maximum and minimum values ​​in} are used to normalize the deviation of each eigenvalue in the eigenvector, so that each eigenvalue is mapped to the interval [0,1] to obtain standardized electrical parameters. Similarly, the receiving voltage, transmitting voltage, rail voltage, and shunt resistance are also mapped to each eigenvalue in the corresponding eigenvector to the interval [0,1] to obtain standardized electrical parameters.

[0080] Furthermore, after the electrical parameters of the collected track section are normalized according to their deviations, the electrical parameters are classified to analyze the working status of a track circuit.

[0081] S2. Label the standardized electrical parameters by parameter categories to obtain electrical parameter categories, and generate data instances of track segments according to the standardized electrical parameters and the electrical parameter categories.

[0082] In an embodiment of the present invention, the electrical parameter category includes a normal category and a fault category. The five electrical parameter characteristics of a track section plus its electrical parameter category are taken as an instance. The fault records of the operating unit are consulted, and the track section without faults within three months is taken as a positive instance, and vice versa, as a negative instance, thereby obtaining the electrical parameter category corresponding to the standardized electrical parameters. That is, according to the defined category rules, each standardized electrical parameter is marked as a corresponding category, and a threshold can be configured for classification, and the electrical parameters above the threshold are taken as fault categories, and the electrical parameters below the threshold are taken as normal categories, thereby obtaining the electrical parameter category corresponding to the standardized electrical parameters.

[0083] Furthermore, in order to analyze the working status of a track circuit, the track sections should be divided into positive instances and negative instances according to the collected electrical parameters to facilitate subsequent model training.

[0084] In the embodiment of the present invention, the data instance refers to five electrical parameter characteristics of a track section plus the category corresponding to the electrical parameter characteristics as an instance, so as to analyze the working status of the track circuit.

[0085] In the embodiment of the present invention, the data instance of generating the track segment according to the standardized electrical parameters and the electrical parameter categories includes:

[0086] Matching the standardized electrical parameters with the electrical parameter categories to obtain a corresponding relationship;

[0087] When the electrical parameter category in the corresponding relationship is a preset first category, dividing the track section into a positive instance;

[0088] When the electrical parameter category in the corresponding relationship is a preset second category, classifying the track section as a negative instance;

[0089] The positive instances and the negative instances are aggregated into data instances of a track segment.

[0090] In detail, the standardized electrical parameters are matched with their electrical parameter categories, such as the electrical category corresponding to the resonant frequency offset is normal, the electrical category corresponding to the receiving voltage is normal, the electrical category corresponding to the transmitting voltage is faulty, the electrical category corresponding to the track surface voltage is faulty, and the electrical category corresponding to the shunt resistance is normal, thereby obtaining a corresponding relationship, and then when the electrical parameter categories in the corresponding relationship are all the preset first categories, the track section is divided into a positive instance, that is, when the electrical parameter categories are all normal, the track section is divided into a positive instance; when any of the electrical parameter categories is faulty, the track section is divided into a negative instance, that is, the track section with a working status of "normal" is a positive instance, recorded as 1, and the track section with a working status of "faulty" is a negative instance, recorded as -1, and the fault records of the operating unit are consulted, and the track section that has not had a fault within 3 months is listed as a positive instance, otherwise it is listed as a negative instance, so that the positive instance and the negative instance are used as data instances of the track section, wherein the preset first category is the normal category, and the preset second category is the fault category.

[0091] Furthermore, in order to adjust the model parameters and verify the reliability of the algorithm, it is necessary to divide the training set and the test set. Therefore, it is necessary to divide the training set and the test set of the model according to the positive instances and negative instances in the data instances.

[0092] S3. Determine a training data set and a test data set of the track segment according to the data instance, train a pre-configured support vector machine model according to the training data set, and use the test data set to perform parameter tuning on the trained support vector machine model.

[0093] In the embodiment of the present invention, the training data set is data used to train the model and adjust the parameters and structure of the model, and the test data set is used to evaluate the performance of the trained model and check the generalization ability of the model on unseen data.

[0094] In an embodiment of the present invention, the step of determining a training data set and a test data set of a track segment according to the data instance includes:

[0095] Selecting a first data set of the track segment from the data instance according to a preset first ratio value by using a preset stratified sampling, and using the selected first data set as a training data set of the track segment;

[0096] A second data set of the track segment is selected from the data instance according to a preset second ratio value, and the selected second data set is used as a test data set of the track segment.

[0097] In detail, in order to adjust the model parameters and verify the reliability of the algorithm, it is necessary to divide the training set and the test set, with 70% of all samples as the training set and 30% as the test set. Since the overall failure rate of digital track circuits is low and the proportion of negative instances is small, a stratified sampling method is used to extract 70% of the positive and negative samples as the training set, and the rest as the test set, where the first ratio value is 70%, the second ratio value is 30%, the first data set refers to the positive and negative instances selected according to 70%, and the second data set refers to the positive and negative instances selected according to 30%, so that 70% of the positive and negative samples in the positive and negative instances are extracted as the training set, and the rest are used as the test set.

[0098] Specifically, the data of the training data set is input into a pre-configured support vector machine model for training. After the training is completed, the classification accuracy is tested with the data of the test data set, and the accuracy of the model training is detected. If the classification accuracy is not high, the penalty parameter C can be increased, the kernel function parameters can be adjusted, or other kernel functions can be replaced to achieve higher classification accuracy. When the model is trained to a relatively satisfactory accuracy, the support vector machine model with optimized parameters is output.

[0099] Furthermore, when the model is trained to a relatively satisfactory accuracy rate, it can be used to analyze historical and new measurement data, and new data samples can be input into the model for classification to obtain the corresponding category status.

[0100] S4. Classify the preset electrical parameters to be analyzed by using the support vector machine model after parameter tuning, and obtain the category instance of the track section corresponding to the electrical parameters to be analyzed.

[0101] In an embodiment of the present invention, the electrical parameters to be analyzed in the track section are input into the support vector machine model after parameter tuning for classification, so as to obtain the category instance of the track section corresponding to the electrical parameters to be analyzed, that is, the support vector machine model finds an optimal hyperplane to separate the data points of different categories as much as possible. This hyperplane has the largest margin and the distance between the nearest data point (that is, the support vector). The track section whose category instance is a positive instance represents that the working state is good at this time, and the track section whose category instance is a negative instance represents that the working state is poor at this time and needs to be maintained as soon as possible.

[0102] Furthermore, the support vector machine model can output the distance between the classified instance and the decision plane. The larger the distance, the higher the confidence of the classification. Therefore, the risk status of the digital track circuit can be accurately determined based on the distance between the classified instance and the decision plane.

[0103] S5. Determine a risk state of a digital track circuit according to a distance between the category instance and a decision plane corresponding to the support vector machine model.

[0104] In the embodiment of the present invention, the risk state refers to a high risk state or a low risk state corresponding to the digital track circuit.

[0105] In an embodiment of the present invention, determining the risk state of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model includes:

[0106] extracting a first distance between a positive instance in the category instance and the decision plane;

[0107] extracting a second distance between a negative instance in the category instance and the decision plane;

[0108] When the first distance is greater than a preset distance threshold, determining that the risk status of the digital track circuit is low risk;

[0109] When the second distance is greater than a preset distance threshold, it is determined that the risk state of the digital track circuit is high risk.

[0110] In detail, in the support vector machine model, the decision plane (or hyperplane) is the boundary that separates different categories. For each data point, the distance between the decision plane and the data point can be used to evaluate the classification confidence of the point. The equation of the decision plane can be expressed as w T x+b=0, where w is the weight vector, b is the bias term, x is the feature vector, and the distance from the data point to the decision plane is Then for the positive instance x pos The first distance from the data point belonging to the positive instance to the decision plane is For negative instances x neg The second distance from the data point (belonging to the negative instance) to the decision plane is Furthermore, if the distance of the positive instance is greater than the preset distance threshold, it means that the model has a high confidence in the classification of these positive instances, and the risk status is determined to be low risk. If the distance of the negative instance is greater than the preset distance threshold, it means that the model has a high confidence in the classification of these negative instances, and the risk status is determined to be high risk, where the distance threshold is custom configured.

[0111] Specifically, the support vector machine model can output the distance between the classified instance and the decision plane. The larger the distance, the higher the confidence of the classification. Therefore, for positive instances with a large distance from the decision plane, it means that the risk level of the track circuit at that moment is very low, and for negative instances with a large distance, it means that the risk level of the track circuit at that moment is very high, which can help operators adjust maintenance strategies.

[0112] Embodiment 2

[0113] In order to understand the present invention more clearly, a second embodiment is used below to further explain the configuration of the support vector machine model in the embodiment of the present invention.

[0114] In an embodiment of the present invention, before training the support vector machine model, it is necessary to configure the model parameters of the support vector machine model. By adjusting the penalty parameters and kernel function parameters, the classification accuracy of the model can be optimized and misclassification can be reduced. Appropriate parameter settings help the support vector machine model achieve a better balance between training data and data to be analyzed, reducing overfitting or underfitting.

[0115] In the embodiment of the present invention, before training the pre-configured support vector machine model according to the training data set, the method further includes:

[0116] Determine the penalty parameter according to the tolerance of the classifier in the preset support vector machine model;

[0117] Select the Gaussian kernel function according to the preset classification attributes;

[0118] Determining model parameters of the support vector machine model according to the penalty parameter and the Gaussian kernel function;

[0119] The support vector machine model is configured by the model parameters.

[0120] In detail, through the support vector machine of soft margin classification, the penalty parameter C represents the tolerance of the classifier to boundary noise points, and the value is between [0,1], and the default value is 0. The larger the C is, the smaller the tolerance of the classifier to noise is, so a smaller C value should be appropriately selected at the beginning of training so that the algorithm can converge as soon as possible; commonly used kernel functions include linear kernel function, polynomial kernel function, Gaussian kernel function, etc. The linear kernel function has no special parameters to be set; the polynomial kernel function has 3 parameters; the Gaussian kernel function has 1 kernel function, and the increase in the number of parameters will increase the difficulty of model training. The Gaussian kernel function has good support for linear inseparability and has only 1 parameter, so the Gaussian kernel function is preferred, where the classification attribute refers to the attribute of linear inseparability.

[0121] Specifically, the penalty parameter C value and the kernel parameter of the Gaussian kernel are adjusted through cross-validation. The kernel parameter controls the influence range of the Gaussian kernel function. The grid search method and other methods can be used to find the optimal parameter combination, and then the selected C value and gamma value are used to configure the SVM model for training and evaluation of the model effect.

[0122] The embodiments of the present invention can comprehensively measure the electrical performance indicators of digital track circuits by collecting electrical parameters, detect the resonant frequency of the circuit tuning unit, increase the detection range of fault causes, and improve the detection quality; by using the support vector machine algorithm, quantitative characterization of the health risk status of digital track circuits can be achieved, the utilization rate of data can be improved, and the health risk status of the entire equipment can be measured. Therefore, the digital track circuit health status analysis method, device, equipment and medium proposed in the present invention can solve the problem of low accuracy when performing digital track circuit health status analysis.

[0123] Embodiment 3

[0124] like Figure 5 As shown, this embodiment also provides a functional module diagram of a digital track circuit health status analysis device.

[0125] The digital track circuit health status analysis device 100 described in this embodiment can be installed in an electronic device. According to the functions implemented, the digital track circuit health status analysis device 100 may include a deviation standardization processing module 101, a data instance generation module 102, a parameter tuning module 103, a track section category analysis module 104 and a risk status determination module 105. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0126] In this embodiment, the functions of each module / unit are as follows:

[0127] The deviation standardization processing module 101 is used to obtain electrical parameters of a track section in a digital track circuit, and perform deviation standardization processing on the electrical parameters to obtain standardized electrical parameters;

[0128] The data instance generation module 102 is used to label the standardized electrical parameters by parameter category to obtain the electrical parameter category, and generate the data instance of the track section according to the standardized electrical parameters and the electrical parameter category;

[0129] The parameter tuning module 103 is used to determine a training data set and a test data set of the track section according to the data example, train a pre-configured support vector machine model according to the training data set, and perform parameter tuning on the trained support vector machine model using the test data set;

[0130] The track section category analysis module 104 is used to classify the preset electrical parameters to be analyzed by using the support vector machine model after parameter tuning, and obtain the category instance of the track section corresponding to the electrical parameters to be analyzed;

[0131] The risk status determination module 105 is used to determine the risk status of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

[0132] In detail, each module described in the digital track circuit health status analysis device 100 described in the embodiment of the present invention adopts the same technical means as the digital track circuit health status analysis method described in Example 1 and Example 2 when used, and can produce the same technical effects, which will not be repeated here.

[0133] Embodiment 4

[0134] like Figure 6 As shown, this embodiment also provides a computer electronic device, which may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a digital track circuit health status analysis program.

[0135] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is the control core (ControlUnit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (for example, executing a digital track circuit health status analysis program, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device and process data.

[0136] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 11 may also include both an internal storage unit of the electronic device and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device, such as the code of the digital track circuit health status analysis program, but also can be used to temporarily store data that has been output or is to be output.

[0137] The communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0138] The communication interface 13 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0139] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0140] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0141] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0142] The digital track circuit health status analysis program stored in the memory 11 in the electronic device is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0143] Acquiring electrical parameters of a track section in a digital track circuit, performing deviation standardization processing on the electrical parameters, and obtaining standardized electrical parameters;

[0144] Perform parameter category labeling on the standardized electrical parameters to obtain electrical parameter categories, and generate data instances of track sections according to the standardized electrical parameters and the electrical parameter categories;

[0145] Determine a training data set and a test data set of the track section according to the data example, train a pre-configured support vector machine model according to the training data set, and use the test data set to perform parameter tuning on the trained support vector machine model;

[0146] Using the support vector machine model after parameter tuning to classify the preset electrical parameters to be analyzed, and obtain the category instance of the track section corresponding to the electrical parameters to be analyzed;

[0147] The risk state of the digital track circuit is determined according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

[0148] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0149] Furthermore, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0150] Embodiment 5

[0151] This embodiment provides a storage medium storing a computer program. When the computer program is executed by a processor, the steps of the digital track circuit health status analysis method described above are implemented.

[0152] These program codes can also be loaded onto a computer or other programmable data processing device so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions executed on the computer or other programmable device for implementing the process. Figure 1 The steps of a specified function in a process or multiple processes.

[0153] Storage media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, modules of programs or other data. Examples of storage media can include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0154] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0155] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0157] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0158] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is not limited only according to the above description, and it is intended that all changes within the meaning and scope of equivalent elements within the scope of protection are included in the present invention.

[0159] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0160] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0161] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for analyzing the health status of a digital track circuit, characterized in that: The method comprises: Acquiring electrical parameters of a track section in a digital track circuit, performing deviation standardization processing on the electrical parameters, and obtaining standardized electrical parameters; Perform parameter category labeling on the standardized electrical parameters to obtain electrical parameter categories, and generate data instances of track sections according to the standardized electrical parameters and the electrical parameter categories; Determine a training data set and a test data set of the track section according to the data example, train a pre-configured support vector machine model according to the training data set, and use the test data set to perform parameter tuning on the trained support vector machine model; Using the support vector machine model after parameter tuning to classify the preset electrical parameters to be analyzed, and obtain the category instance of the track section corresponding to the electrical parameters to be analyzed; The risk state of the digital track circuit is determined according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

2. The digital track circuit health status analysis method according to claim 1, characterized in that: The electrical parameters include resonant frequency shift, receiving voltage, transmitting voltage, rail voltage and shunt resistance.

3. The digital track circuit health status analysis method according to claim 1, characterized in that: The measurement of the resonant frequency shift comprises the following steps: Turn off the power supply of the tested section in the track section of the digital track circuit and the indoor sending units of the two sections adjacent to the tested section; After turning off the power, connect the two terminals in the connection room of the preset tuning unit box to the signal amplifier and signal generator respectively, and connect the vehicle antenna of the digital track circuit to the spectrum analyzer; Using the signal generator to generate a frequency sweep signal, and amplifying the frequency sweep signal through the signal amplifier; After the amplified frequency sweep signal passes through the resonant circuit of the tuning unit, the frequency corresponding to the maximum amplitude signal is read by the spectrum analyzer as the resonant frequency of the resonant circuit; The absolute difference between the resonant frequency and the design frequency corresponding to the measured section is used as the resonant frequency offset of the resonant circuit.

4. The digital track circuit health status analysis method according to claim 1, characterized in that: The step of performing deviation standardization processing on the electrical parameters to obtain standardized electrical parameters includes: Convert the parameter value corresponding to each electrical parameter into a feature vector; The deviation of each electrical parameter is normalized according to the characteristic vector by the following formula to obtain the normalized electrical parameter: Among them, x ′ is the standardized electrical parameter, x is the eigenvalue corresponding to the electrical parameter in the eigenvector, and x min is the minimum value in the eigenvector, x max is the maximum value in the eigenvector.

5. The digital track circuit health status analysis method according to claim 1, characterized in that: The data example of generating the track segment according to the standardized electrical parameters and the electrical parameter categories includes: Matching the standardized electrical parameters with the electrical parameter categories to obtain a corresponding relationship; When the electrical parameter category in the corresponding relationship is a preset first category, dividing the track section into a positive instance; When the electrical parameter category in the corresponding relationship is a preset second category, classifying the track section as a negative instance; The positive instances and the negative instances are aggregated into data instances of a track segment.

6. The digital track circuit health status analysis method according to claim 1, characterized in that: The step of determining the risk state of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model includes: extracting a first distance between a positive instance in the category instance and the decision plane; extracting a second distance between a negative instance in the category instance and the decision plane; When the first distance is greater than a preset distance threshold, determining that the risk status of the digital track circuit is low risk; When the second distance is greater than a preset distance threshold, it is determined that the risk state of the digital track circuit is high risk.

7. The digital track circuit health status analysis method according to claim 1, characterized in that: Before training the pre-configured support vector machine model according to the training data set, the method further includes: Determine the penalty parameter according to the tolerance of the classifier in the preset support vector machine model; Select the Gaussian kernel function according to the preset classification attributes; Determining model parameters of the support vector machine model according to the penalty parameter and the Gaussian kernel function; The support vector machine model is configured by the model parameters.

8. A digital track circuit health status analysis device, characterized in that: The device comprises: A deviation standardization processing module is used to obtain electrical parameters of a track section in a digital track circuit, perform deviation standardization processing on the electrical parameters, and obtain standardized electrical parameters; A data instance generation module, used for labeling the standardized electrical parameters by parameter category to obtain the electrical parameter category, and generating a data instance of the track section according to the standardized electrical parameters and the electrical parameter category; A parameter tuning module, used to determine a training data set and a test data set of a track section according to the data instance, train a pre-configured support vector machine model according to the training data set, and perform parameter tuning on the trained support vector machine model using the test data set; A track section category analysis module is used to classify preset electrical parameters to be analyzed using the support vector machine model after parameter tuning, and obtain a category instance of the track section corresponding to the electrical parameters to be analyzed; The risk status determination module is used to determine the risk status of the digital track circuit according to the distance between the category instance and the decision plane corresponding to the support vector machine model.

9. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the digital track circuit health status analysis method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the digital track circuit health status analysis method described in any one of claims 1 to 7 are implemented.

Images