A method for comparing railway train fuzzy data protocols

Abstract

The present application is a railway train fuzzy data protocol comparison method, comprising the following steps: obtaining the messages of each port of the present train protocol, and respectively aligning the message data of each port with the collection time stamp to obtain a time sequence set of each port message; performing data preprocessing on the original data of the time sequence set of each port message; dividing the preprocessed time sequence set into a determined time sequence sub-set Y and a fuzzy time sequence sub-set X; calculating the correlation of the determined time sequence set Y and the fuzzy time sequence set X; predicting the variable type and meaning of the fuzzy protocol variable through the correlation calculation result; verifying the correctness of the assumed result through short-term confirmation means, and determining whether the assumed meaning and type conform to the actual operation of the vehicle through long-term verification means. The method is convenient and fast to determine the meaning of the fuzzy protocol, reduces the working intensity of the technical personnel, and improves the working efficiency.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of rail transit vehicles, and relates to a railway train fuzzy data protocol comparison method. BACKGROUND

[0002] The intelligent operation and maintenance system cannot be separated from the support of massive train operation data, but old train models without train data acquisition and transmission functions have become a short board in the data acquisition process. In system design, the old train models can complete data acquisition by adding recorders with acquisition and wireless transmission functions. In the process of analyzing train data of old train models, due to train expansion, overhaul, addition of new systems and other technical reasons, the retained data protocol does not conform to the existing vehicles. The operation parties of early old train models and foreign manufacturers do not standardize the management of data protocols, resulting in the retained data protocol not conforming to the existing vehicles. The inaccuracy of the train data protocol seriously affects the accuracy of train data analysis and hinders the data integrity process of the intelligent operation and maintenance platform. SUMMARY

[0003] In order to solve the problem of fuzzy data protocol confirmation and update, the present application provides a train fuzzy data protocol comparison method to improve the accuracy of train data analysis and complete the data integrity of the intelligent operation and maintenance platform. The railway train fuzzy data protocol comparison method provided by the present application is a comparison method that meets the data analysis requirements of existing old train models, and is based on the retained protocol to analyze the correlation between the fuzzy protocol variables and the determined variables in the retained protocol.

[0004] A railway train fuzzy data protocol comparison method comprises the following steps:

[0005] Step 1, obtaining the messages of each port of the existing train protocol, and respectively aligning the message data of each port with the acquisition time stamp to obtain a time sequence set of each port message;

[0006] Step 2, data preprocessing is performed on the original data of the time sequence set of each port message;

[0007] Step 3, the preprocessed time sequence set is divided into a determined time sequence sub-set Y and a fuzzy time sequence sub-set X;

[0008] Step 4, calculating the correlation of the determined time sequence set Y and the fuzzy time sequence set X;

[0009] Step 5, predicting the variable type and meaning of the fuzzy protocol variable through the correlation calculation result;

[0010] Step 6: Verify the correctness of the assumed result by short-term confirmation means, and determine the meaning and type of the assumption by long-term verification means to meet the actual operation of the vehicle.

[0011] Further: the raw data of the time series set of each port message is preprocessed as follows:

[0012] All time series are preprocessed, and the preprocessing process includes normalization, time length clipping, and piecewise aggregation average;

[0013] S21: Normalization, by linear normalization, as shown in formula (1), the value of the time series is concentrated in the range [0, 1]:

[0014]

[0015] x represents the original data value of a certain time series, and x' represents the normalized time series value;

[0016] S22: The record file of the train operation data record is long in time, and the time series is clipped with the train single-direction running time T as the fixed clipping time, T represents the time length of the train running from the starting station to the terminal station; The time series set is clipped with T as the period, and the length of the time series is uniform, and the whole train running process traverses all the running scenes and line station conditions;

[0017] S23: Reduce the dimension of each time series by piecewise aggregation average method, the specific formula is as follows:

[0018]

[0019] Wherein, is the i-th element of the processed sequence set X, N is the length of the preprocessed set, and n is the original input sequence length;

[0020] Slide a fixed-size sliding window on the time series, and calculate the mean value of the data in the sliding window as the representation of the data in the whole window, and reduce the sampling points of the time series. Extract the features of the sequence.

[0021] Further: the correlation degree is calculated by Pearson coefficient to calculate the correlation value, and the calculation formula of the Pearson coefficient is as follows:

[0022]

[0023] Wherein, N is the length of the sequence x(iR and y(i), and τ is the time delay;

[0024] Then the correlation degree is determined by looking up the table according to the correlation value.

[0025] Further, the process of predicting the variable type and meaning of the fuzzy protocol variable according to the correlation calculation result is as follows:

[0026] First, assume X Byte Collect all the variable forms corresponding to the time series as byte variables, and traverse X Byte Collect all subsets in the set, calculate the Pearson coefficient between all time series, and obtain the Pearson coefficient matrix P of X Byte P is a symmetric matrix, and for X Byte The correlation between time series i and all time series in the set is the row vector P i of the P matrix.

[0027] If the elements of P i are all <0.6, it means that the time series i has weak correlation with other time series in the set, then the first four Boolean quantities form a high four-bit half-byte sequence, and the last four Boolean quantities form a low four-bit half-byte sequence. Calculate the Pearson coefficient of the two half-byte sequences and judge the Pearson coefficient correlation of the two half-byte sequences.

[0028] If any element of P i is ≥0.6, it means that the time series i has strong correlation with other time series in the set, so it can be determined that the variable form corresponding to the sequence is based on byte variable, and further combined with the value range and change trend of the two sequences to determine whether it is a byte variable or double-byte variable.

[0029] In summary, for X Byte The set is reorganized into X Bool , X Byte , X Word , and X Dword according to the presumed variable form.

[0030] Further, the process of judging the Pearson coefficient correlation of the two half-byte sequences is as follows:

[0031] If the Pearson coefficient of the two half-byte variables is greater than ≥0.6, it is determined that the variable form of the time series i is byte variable, and if the Pearson coefficient of the two half-byte variables is <0.6, it is determined that the variable form of the time series i is Boolean variable.

[0032] A railway train fuzzy data protocol comparison device, comprising:

[0033] The acquisition module is used for acquiring the messages of each port of the present train protocol, and aligning the message data of each port with the collection time stamp to obtain the time series set of each port message.

[0034] a preprocessing module for data preprocessing of raw data of the time series set of each port message;

[0035] a division module for dividing the preprocessed time series set into a determined time series sub-set Y and a fuzzy time series sub-set X;

[0036] a calculation module for calculating the correlation of the determined time series set Y and the fuzzy time series set X;

[0037] a prediction module for predicting the variable type and meaning of the fuzzy protocol variable through the correlation calculation result;

[0038] a determination module for verifying the correctness of the assumed result through short-term confirmation means and determining that the assumed meaning and type conform to the actual operation of the vehicle through long-term verification means.

[0039] The railway train fuzzy protocol comparison method provided by the present application can quickly determine the meaning of the fuzzy protocol, reduce the work intensity of the technical personnel, improve the work efficiency, improve the accuracy of train data analysis, and make up for the short board of data acquisition of the intelligent operation and maintenance platform. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor under the premise of not paying creative labor.

[0041] Figure 1 The method flowchart of the present application;

[0042] Figure 2 The data type prediction flowchart of the present application. DETAILED DESCRIPTION

[0043] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict, and the present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0044] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. The following description of at least one example embodiment is merely illustrative in nature and is in no way limiting on the application or its applications or uses. Based upon a review of the embodiments contained herein, those of ordinary skill in the art can make other embodiments that are within the scope of the application without undue experimentation. Persons of ordinary skill in the art can obtain all other embodiments from the embodiments in the present application without any creative work under the premise that no creative work is made.

[0045] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments in accordance with the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, steps, operations, elements, components, and / or groups thereof, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof.

[0046] The relative arrangement of parts and steps, numerical expressions, and numerical values set forth in the examples are not intended to limit the scope of the application unless otherwise specifically stated. It should be apparent that the dimensions of the various parts shown in the drawings are not to scale and that the dimensions of the parts as disclosed herein are intended to be exemplary only. Well-known technologies, methods and equipment may not be discussed in detail because that technology, methods and equipment should be considered as part of the disclosure where appropriate. In all examples shown and discussed herein, any specific value should be interpreted as illustrative only and not as a limitation. Thus, other examples of example embodiments can have different values. It should be noted that like reference numerals and letters refer to like items in the following drawings and, as such, further discussion of such items across figures need not be discussed further.

[0047] In the description of the present application, it should be understood that the orientation words such as "front, back, up, down, left, right", "transverse, vertical, perpendicular, horizontal" and "top, bottom" and the like indicated orientation or position relationship are generally based on the orientation or position relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, without the opposite indication, these orientation words do not indicate and imply that the indicated device or element must have a specific orientation or be constructed and operated in a specific orientation, therefore, it cannot be understood as a limitation on the scope of protection of the present application: the orientation words "inner, outer" refer to the inner and outer relative to the contour of each component.

[0048] For the purposes of the description, reference can be made to relative terms such as "on", "above", "upper surface", "upper", etc. to describe the spatial relationship of one device or feature to another device or feature as shown in the drawings. It is to be understood that the relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, then a device described as "above" or "on top of" another device would then be oriented "below" or "on bottom of" the other device. Accordingly, the exemplary term "above" can encompass both an orientation of above and below. The devices can be oriented in other ways (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein interpreted accordingly.

[0049] In addition, it should be noted that the use of "first", "second", etc. words to define parts, only for the convenience of the corresponding parts are distinguished, such as no other declaration, the above words have no special meaning, therefore can not be understood as limiting the scope of protection of the present application.

[0050] The embodiment records the data messages of each data port of the train in real time through the vehicle-mounted data acquisition recorder. The complete data messages are distinguished by port number, and the time stamp of the message record is also recorded.

[0051] The retention agreement of the old vehicle model does not conform to the agreement of the existing vehicle, and there are the following situations:

[0052] The existing vehicle agreement is extended on the basis of the retention agreement, most of the retention agreement is consistent, and the increased part is predicted based on the retention agreement.

[0053] The existing vehicle agreement is reduced on the basis of the retention agreement, and the comparison of the existing vehicle agreement is completed after the comparison of each.

[0054] The existing vehicle agreement is spliced on the basis of the retention agreement, and the screen information of the original vehicle, the vehicle-mounted record file and the standardized data agreement template of each subsystem of the train are predicted.

[0055] In view of the above situations of the old vehicle model train fuzzy agreement, the present application provides a fuzzy agreement comparison method to solve the above problems. The steps are as follows:

[0056] Step 1, obtaining the message of each port of the existing vehicle agreement, and aligning the message data of each port with the acquisition time stamp, obtaining the time sequence set of each port message;

[0057] Step 2, data preprocessing is performed on the original data of the time sequence set of each port message;

[0058] Step 3, after the vehicle agreement and artificial confirmation, the pre-processed time series set is divided into a determined time series sub-set Y and a fuzzy time series sub-set X;

[0059] Step 4, calculate the correlation of the determined time series set Y and the fuzzy time series set X;

[0060] Step 5, predict the variable type and meaning of the fuzzy protocol variable through the correlation calculation result;

[0061] Step 6, verify the correctness of the assumed result through short-term confirmation means, and determine that the assumed meaning and type meet the actual operation of the vehicle through long-term verification means.

[0062] Step 1, Step 2, Step 3, Step 4, Step 5 and Step 6 are sequentially executed;

[0063] The network data of the train can be roughly divided into two categories: running data and fault data. For the running data of the train, the data is obtained by monitoring the train network through the train data recorder, and the comparison of the fuzzy running data protocol provides data support. For the fault data of the train, it is sporadic and needs to be triggered by specific conditions to obtain fault data of each port. The acquisition of train fault data is divided into two ways. Trigger high-frequency faults in the form of artificial fault injection to compare and confirm the fuzzy protocol. Low-frequency faults or faults that are difficult to trigger through artificial simulation need to accumulate long-term fault data to achieve the purpose of final comparison and confirmation.

[0064] The process of obtaining the messages of each port of the existing vehicle protocol and aligning the message data of each port with the collection timestamp is as follows:

[0065] The message includes data variables confirmed according to the retained protocol and data variables to be confirmed;

[0066] According to the form of the data variables in the retained protocol, the confirmed data variables are split;

[0067] The basic data variable form includes: Boolean variable Bool, byte variable Byte, word variable Word, double word variable DWord, etc.

[0068] The message data of each port is respectively aligned with the collection timestamp, and the time series set of each port is obtained.

[0069] According to the retained protocol, the port message is split into a time series set as follows:

[0070] Y = {Y Bool , YByte , Y Word , Y Dword} (1)

[0071] where Y Bool is a subset of Boolean variable time series, Y Byte is a subset of byte variable time series, Y Word is a subset of word variable time series, Y Dword is a subset of double word variable time series;

[0072] For the fuzzy data variable to be determined, all fuzzy variables are first split by default as byte variables, and after time alignment, a time series set is formed:

[0073] X Byte ={X1(x0, x1, x2), X2(x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 )…}(2)

[0074] wherein: X1 is a subset composed of eight Boolean variable time series from x0 to x7, and similarly, eight Boolean variable time series of the same byte form a subset.

[0075] The original data of each port obtained by the vehicle-mounted data acquisition recorder will have some data quality problems, and the time series set formed after time alignment will also have data quality problems, which need to be preprocessed. The process of data preprocessing of the original data of the time series set of each port message is as follows:

[0076] Data preprocessing removes noise data and irrelevant data in the data set, processes missing data, missing values, identifies and deletes isolated points, etc.

[0077] All time series are preprocessed, and the preprocessing process includes normalization, time length clipping, and piecewise aggregation averaging three processes.

[0078] S21: Normalize each time series, and through linear normalization, as shown in formula (3), the value of the time series is concentrated in the range of [0, 1]:

[0079]

[0080] x represents the original data value of a certain time series, and x' represents the normalized time series value;

[0081] S22: The recording file time of the train operation data record is long. In order to facilitate comparison, we use the train one-way running time T as the fixed time length to clip the time sequence, T represents the time length of the train running from the starting station to the terminal station; the time sequence set is clipped with T as the period, and the length of the time sequence is uniform, and T is taken as the period,

[0082] All running scenarios and line station conditions are traversed. The port sampling period of the vehicle-mounted network is millisecond level, and the collection value of a variable can be repeatedly sampled in the sampling period; the sequence is sampled and valued in a very short time, the value range of the sequence changes little, and even repeated values appear;

[0083] S23: The specific formula is as follows by using the method of segment-by-segment aggregation average:

[0084]

[0085] Among them, is the i-th element of the processed sequence set X, N is the length of the preprocessed set, and n is the length of the original input sequence;

[0086] A fixed size sliding window is slid on the time sequence, and the mean value of the data in the sliding window is calculated as the representation of the data in the whole window. The sampling points of the time sequence are reduced, the features of the sequence are extracted, the data amount of the subsequent correlation calculation is reduced, and the efficiency is improved.

[0087] Further, based on the determined time sequence set Y = {Y Bool , Y Byte , Y Word , Y Dword}, the correlation of each time and N in the fuzzy time sequence set X Byte is analyzed, and the specific process is as follows:

[0088] The correlation level of two time sequences is divided:

[0089] (1) There is no correlation between the two time sequences at the same time scale;

[0090] (2) There is a fuzzy correlation between the two time sequences at the same time scale;

[0091] (3) There is a strong correlation between the two time sequences at the same time scale, but there is no clear functional relationship between the two time sequences;

[0092] (4) There is a clear functional relationship between the two time sequences at the same time scale.

[0093] The present application does not consider the function relationship of time series, only analyzes and compares the correlation of two time series to predict and determine the specific meaning of the fuzzy protocol message.

[0094] The Pearson coefficient is used to measure the correlation of two time series, and the formula (3) is as follows.

[0095]

[0096] Wherein, N is the length of the sequence x(iR and y(i), τ is the time delay, and the correlation between the two time series is not necessarily strictly time synchronized.

[0097] Because the bus transmission delay of the train system and the system delay of the mechanical and electrical subsystems are considered, the total delay can be considered according to the characteristics of the subsystem corresponding to the current port, and the multiple of the port sampling period is taken as the value of τ. E(x) and E(y) are the mean of the two sequences, and σ(x) and σ(y) are the standard deviation of the two sequences.

[0098] The absolute value of the Pearson coefficient can be measured by the following table to measure the correlation degree of the two sequences.

[0099] Degree of correlation Very strong correlation Strong correlation Moderate correlation Weak correlation No correlation |γ xy |]]> (0.8,1] (0.6,0.8] (0.4,0.6] (0.2,0.4] (0,0.2]

[0100] The Pearson coefficient between Y and the fuzzy sequence set X in the sequence set is calculated and determined according to the following method. bool

[0101] Further, the process of predicting the variable type and meaning of the fuzzy protocol variable through the correlation calculation result is as follows:

[0102] First, assume that the variable form corresponding to all time series in the X Byte set is a byte variable, and all subsets in the X Byte set are traversed, the Pearson coefficient between all time series is calculated, and the Pearson coefficient matrix P of X Byte is obtained, P is a symmetric matrix, and for the time series i in the X Byte set, the correlation with all time series is the row vector P i of the P matrix.

[0103] If the elements of P i are all <0.6, it means that the time series i has weak correlation with other time series in the set, then the first four Boolean quantities form a high four-bit half-byte sequence, the last four Boolean quantities form a low four-bit half-byte sequence, and the Pearson coefficient of the two half-byte sequences is calculated and the Pearson coefficient correlation of the two half-byte sequences is judged.

[0104] If the elements of P i ​If any element of the set is greater than or equal to 0.6, it indicates that the time series i has a strong correlation with other time series in the set, and it can be determined that the variable form corresponding to the series is based on byte variables. Further, the value range and the change trend of the two series are combined to determine whether it is a word variable or a double word variable.

[0105] In summary, X Byte The set is reorganized into X Bool , X Byte , X Word , X Dword} according to the presumed variable form.

[0106] The process of determining the Pearson correlation coefficient of the two half-byte sequences is as follows:

[0107] If the Pearson correlation coefficient of the two half-byte variables is greater than 0.6, it is determined that the variable form of the time series i is a byte variable. If the Pearson correlation coefficient of the two half-byte variables is less than 0.6, it is determined that the variable form of the time series i is a Boolean variable.

[0108] Further, the process of calculating the correlation between the determined time series set Y and the fuzzy time series set X is as follows:

[0109] The absolute value of the Pearson correlation coefficient of each sequence x(i) in the set X and each sequence y(j) in the determined time set Y is calculated, and thus the correlation matrix P ij is obtained.

[0110] The elements greater than 0.6 in the correlation matrix are retained, and only the strongly correlated and strongly correlated elements are used as the subsequent comparison research objects.

[0111] After it is determined that the element y(i) in the Y set has a strong correlation with x(i) in the X set, the meaning of the variable corresponding to y(i) has a strong correlation with the meaning of the variable corresponding to x(i), the corresponding system components, the associated state variables, and the fault feedback. The determined variables with strong correlation with y(i) are sorted according to the correlation coefficient, and the meanings of the variables with larger correlation coefficients and the corresponding components are analyzed to determine the range of the variable corresponding to y(i). Within the range, the meaning of the variable corresponding to y(i) is assumed based on data protocol files, system schematic diagrams, expert experience, etc.

[0112] Further, after assuming the variable type and meaning of y(i), the correctness of the assumed result is verified by short-term confirmation means such as system restart, fault signal injection, communication interference, etc. Long-term verification means are used to determine whether the assumed meaning and type meet the actual operation of the vehicle. A fault, a signal feedback, or a cumulative variable is the result of the interaction between systems.

[0113] A railway train fuzzy data protocol comparison device, comprising:

[0114] An acquisition module is configured to acquire messages of each port of a current train protocol, and perform time alignment on message data of each port with a collection timestamp respectively to obtain a time sequence set of each port message;

[0115] A preprocessing module is configured to perform data preprocessing on original data of the time sequence set of each port message;

[0116] A division module is configured to divide the preprocessed time sequence set into a determined time sequence sub-set Y and a fuzzy time sequence sub-set X;

[0117] A calculation module is configured to calculate the correlation of the determined time sequence set Y and the fuzzy time sequence set X;

[0118] A prediction module is configured to predict the variable type and meaning of a fuzzy protocol variable through the correlation calculation result;

[0119] A determination module is configured to verify the correctness of the assumed result through a short-term confirmation means, and determine that the assumed meaning and type conform to the actual operation condition of the vehicle through a long-term verification means.

[0120] A railway train fuzzy data protocol comparison device, comprising:

[0121] An acquisition module is configured to acquire messages of each port of a current train protocol, and perform time alignment on message data of each port with a collection timestamp respectively to obtain a time sequence set of each port message;

[0122] A preprocessing module is configured to perform data preprocessing on original data of the time sequence set of each port message;

[0123] A division module is configured to divide the preprocessed time sequence set into a determined time sequence sub-set Y and a fuzzy time sequence sub-set X;

[0124] A calculation module is configured to calculate the correlation of the determined time sequence set Y and the fuzzy time sequence set X;

[0125] A prediction module is configured to predict the variable type and meaning of a fuzzy protocol variable through the correlation calculation result;

[0126] A determination module is configured to verify the correctness of the assumed result through a short-term confirmation means, and determine that the assumed meaning and type conform to the actual operation condition of the vehicle through a long-term verification means.

[0127] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for comparing railroad train fuzzy data protocols, the method comprising: The method comprises the following steps: Step 1, obtaining the messages of each port of the existing vehicle protocol, and time-aligning the message data of each port with the collected time stamp respectively to obtain a time sequence set of each port message; Step 2, data preprocessing is performed on the original data of the time sequence set of each port message; Step 3, divided into two subsets: a sub-set Y of retained protocol confirmation and a sub-set X of fuzzy confirmation to be confirmed, Step 4, for each sequence x in the sub-set X of fuzzy confirmation to be confirmed, the correlation of x with any sequence y in the sub-set Y of retained protocol confirmation is calculated; Step 5, through the correlation calculation result, the variable type and meaning corresponding to each sequence x in the sub-set X of fuzzy confirmation to be confirmed are predicted; The process of predicting the variable type and meaning corresponding to each sequence x in the sub-set X of fuzzy confirmation to be confirmed through the correlation calculation result is as follows: First, assume that the variable form corresponding to all time sequences in the sub-set X of fuzzy confirmation to be confirmed is a byte variable, for each sequence x in the sub-set X of fuzzy confirmation to be confirmed, the Pearson coefficient of x and any sequence y in the sub-set Y of retained protocol confirmation is calculated, The Pearson coefficient matrix P of the sub-set X of fuzzy confirmation to be confirmed is obtained, P is a symmetric matrix, for each sequence x in the sub-set X of fuzzy confirmation to be confirmed, a row vector p in the P matrix is used; If all elements of p are <0.6, it means that the corresponding sequence x has weak correlation with all time sequences in the sub-set Y of retained protocol confirmation, then the first four Boolean quantities form a high four-bit half-byte sequence, the last four Boolean quantities form a low four-bit half-byte sequence, and the Pearson coefficients of the two half-byte sequences are calculated and the Pearson coefficient correlation of the two half-byte sequences is judged; If any element of p is ≥0.6, it means that the corresponding sequence x has strong correlation with the time sequences in the sub-set Y of retained protocol confirmation, it can be determined that the variable form corresponding to the sequence is based on byte variable, and further combined with the value range and change trend of the two sequences to determine whether it is a byte variable or a double byte variable; The fuzzy subset X is reorganized into X = {X Bool , X Byte , X Word , X Dword} according to the presumed variable form. Step 6, the correctness of the assumed result is verified through short-term confirmation means, and the assumed meaning and type are determined to be consistent with the actual operation of the vehicle through long-term verification means.

2. A method of fuzzy data protocol comparison for a railway train according to claim 1, characterized in that: The process of data preprocessing on the original data of the time sequence set of each port message is as follows: All time sequences are preprocessed, and the preprocessing process includes normalization, time length clipping, and piecewise aggregation averaging; S21: normalization processing, through linear normalization, as shown in formula (1), the values of the time sequence are concentrated in the range of [0, 1]: (1) a raw data value representative of a time series, a normalized time series value representative; S22: the record file time of the train operation data record is long, and the time sequence is clipped with the train single-direction running time T as the fixed time length, T represents the time length of the train running from the starting station to the terminal station; The time sequence set is clipped with T as the period, and the length of the obtained time sequence is uniform, and the train running process traverses all running scenarios and line station conditions with T as the period; S23: through the method of piecewise aggregation averaging, the dimension of each time sequence is reduced, and the specific formula is as follows: （2） wherein, for the i-th element of the processed ambiguous to be confirmed sub-set X, N is the pre-processed set length, n is the original input sequence length; Sliding a fixed size sliding window on the time series, and calculating the mean value of the data in the sliding window as the representation of the data in the whole window, reducing the sampling points of the time series, and extracting the features of the sequence.

3. A method for comparing railroad train fuzzy data protocols according to claim 1, wherein: The process of judging the Pearson coefficient correlation of the two half-byte sequences is as follows: If the Pearson coefficient of the two half-byte variables is greater than 0.6, it is determined that the variable form of the time series x is a byte variable, and if the Pearson coefficient of the two half-byte variables is less than 0.6, it is determined that the variable form of the time series x is a Boolean variable.

4. A railroad train fuzzy data protocol comparison apparatus, characterized by: It comprises: An acquisition module for acquiring messages of each port of the existing vehicle protocol, and time stamping each port message data with the collection time stamp to obtain a time series set of each port message; A preprocessing module for data preprocessing of the original data of the time series set of each port message; A division module for dividing into two subsets: a subset Y of retained protocol confirmations and a subset X of fuzzy confirmations to be confirmed, A calculation module for calculating the correlation of each sequence x in the subset X of fuzzy confirmations to be confirmed with any sequence y in the subset Y of retained protocol confirmations; A prediction module for predicting the variable type and meaning corresponding to each sequence x in the subset X of fuzzy confirmations to be confirmed through the correlation calculation result; The process of predicting the variable type and meaning corresponding to each sequence x in the subset X of fuzzy confirmations to be confirmed through the correlation calculation result is as follows: First, assume that all time series in the subset X of fuzzy confirmations to be confirmed correspond to byte variables, and calculate the Pearson coefficient of each sequence x in the subset X of fuzzy confirmations to be confirmed with any sequence y in the subset Y of retained protocol confirmations, Obtain the Pearson coefficient matrix P of the subset X of fuzzy confirmations to be confirmed, P is a symmetric matrix, and for each sequence x in the subset X of fuzzy confirmations to be confirmed, use a row vector p in the P matrix; If all elements of p are less than 0.6, it means that the corresponding sequence x has weak correlation with all time series in the subset Y of retained protocol confirmations, then the first four Boolean quantities form a high four-bit half-byte sequence, the last four Boolean quantities form a low four-bit half-byte sequence, and the Pearson coefficient of the two half-byte sequences is calculated and the Pearson coefficient correlation of the two half-byte sequences is judged; If any element of p is greater than or equal to 0.6, it means that the corresponding sequence x has strong correlation with the time series in the subset Y of retained protocol confirmations, and it can be determined that the variable form corresponding to the sequence is based on byte variables, and further combined with the value range and change trend of the two sequences to determine whether it is a byte variable or a double-byte variable. The fuzzy subset X is reorganized into X = {X Bool , X Byte , X Word , X Dword } according to the presumed variable form. A determination module for verifying the correctness of the assumed result through short-term confirmation means, and determining that the assumed meaning and type conform to the actual operation of the vehicle through long-term verification means.

Images