Train Overtime Fault Judgment Method and Device

Abstract

The present invention discloses a method and device for judging train timeout faults, which relates to the field of artificial intelligence technology. The method includes: extracting the feature values corresponding to each timeout fault feature from the historical interface data of the GSM-R network; using a decision tree as a model and adopting the CART algorithm to train the model respectively with the training samples corresponding to the fault category, fault cause, and fault code, so as to obtain the timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively; inputting the interface data of the GSM-R network with unknown train timeout faults into the timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively to obtain the corresponding fault category judgment result, fault cause judgment result, and fault code judgment result; and obtaining the judgment result of the unknown train timeout fault according to the fault category judgment result, fault cause judgment result, and fault code judgment result. The efficiency of judging the fault attribution of train timeout faults is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to a method and device for judging train timeout faults. Background Art

[0002] This section aims to provide background or context for the embodiments of the present invention described in the claims. The description herein is not admitted to be prior art merely because it is included in this section.

[0003] The timeout fault of the Chinese Train Control System 3 (CTCS-3) means that when a train uses the CTCS-3 train control system to control the train operation, the train triggers the service brake due to the timeout of the GSM-R network wireless communication (also called the interruption of the vehicle-ground communication). The frequent deceleration and acceleration processes will affect the passenger riding comfort, and the resulting train delays will also reduce the railway operation efficiency. Therefore, the research on timeout faults is of great significance. How to determine the fault attribution and quickly locate the fault point is a key and also one of the difficulties in the CTCS-3 timeout fault handling process.

[0004] The wireless communication bearer network of the train control system is the Global System for Mobile Communications – Railway (GSM-R) network. According to the vehicle-ground communication rules of the train control system and the basic communication principle of the GSM-R network, the interface data of the Abis interface, A interface, and Primary Rate Interface (RPI) interface of the GSM-R network are monitored, analyzed and processed, and the features related to the CTCS-3 timeout fault are extracted, providing a basis for the CTCS-3 timeout fault category judgment and fault cause analysis.

[0005] Each interface provides various features or feature groups for timeout fault category judgment and fault cause analysis. The PRI interface is divided into the application layer, security layer, transport layer, network layer, link layer, and physical layer. The A interface undertakes the messages between the Base Station Controller (BSC) and the Mobile-service Switching Center (MSC), as well as between the Mobile Terminal (MT) and the MSC. The main features related to the Abis interface signaling are handover signaling and connection release signaling. Since there are many levels and processes in network data signaling, generally, a group of features corresponds to a fault cause. When the number of features and combinations is large, it will be very difficult to organize and combine these features into system logic. Especially when new feature combinations are continuously added during wireless optimization, it can no longer meet the requirements to complete system logic integration through manual methods. Summary of the Invention

[0006] An embodiment of the present invention provides a method for judging train timeout faults, which is used to improve the efficiency and accuracy of judging the fault attribution of CTCS-3 timeout faults. The method includes:

[0007] Extracting the feature values corresponding to each timeout fault feature from the historical interface data of the GSM-R network of the railway integrated digital mobile communication system;

[0008] Preprocessing each timeout fault feature and the feature values corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively;

[0009] Using a decision tree as a model and adopting the Classification and Regression Tree (CART) algorithm, training the model respectively with the training samples corresponding to the fault category, fault cause, and fault code to obtain timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively;

[0010] Inputting the interface data of the GSM-R network corresponding to an unknown train timeout fault into the timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively to obtain corresponding fault category judgment results, fault cause judgment results, and fault code judgment results;

[0011] Obtaining the judgment result of the unknown train timeout fault according to the fault category judgment result, the fault cause judgment result, and the fault code judgment result.

[0012] An embodiment of the present invention also provides a device for judging train timeout faults, which is used to improve the efficiency and accuracy of judging the fault attribution of CTCS-3 timeout faults. The device includes:

[0013] An extraction module, configured to extract the feature values corresponding to each timeout fault feature from the historical interface data of the GSM-R network;

[0014] A first processing module, configured to preprocess each timeout fault feature and the feature values corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively;

[0015] A model training module, configured to use a decision tree as the model, adopt the CART algorithm, and respectively use the training samples corresponding to the fault category, fault cause, and fault code to train the model to obtain timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively;

[0016] A second processing module, configured to input the interface data of the GSM-R network corresponding to the unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively, and obtain corresponding fault category decision results, fault cause decision results, and fault code decision results;

[0017] A third processing module, configured to obtain the decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result.

[0018] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned train timeout fault decision method is implemented.

[0019] An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned train timeout fault decision method is implemented.

[0020] An embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the above-mentioned train timeout fault decision method is implemented.

[0021] In the embodiments of the present invention, eigenvalue corresponding to each timeout fault feature is extracted from the historical interface data of the GSM-R network; preprocessing is performed on each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively; using a decision tree as the model and adopting the CART algorithm, the model is trained respectively using the training samples corresponding to the fault category, fault cause, and fault code to obtain timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively; the interface data of the GSM-R network corresponding to the unknown train timeout fault is input into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively to obtain corresponding fault category decision results, fault cause decision results, and fault code decision results; the decision result of the unknown train timeout fault is obtained according to the fault category decision result, the fault cause decision result, and the fault code decision result. In this way, the fault attribution of the CTCS-3 timeout fault is judged by using the timeout fault decision models corresponding to the fault category, fault cause, and fault code, which improves the efficiency and accuracy of judging the fault attribution of the CTCS-3 timeout fault; and the timeout fault decision model uses a decision tree as the model, and the output logic has a high degree of visualization and is easy to understand. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings. In the drawings:

[0023] Figure 1 It is a flowchart of a method for judging train timeout faults provided in the embodiments of the present invention;

[0024] Figure 2 It is a flowchart of a method for preprocessing each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively;

[0025] Figure 3 It is a flowchart of a method for respectively configuring the association relationship between the fault category, fault cause, fault code and the timeout fault feature;

[0026] Figure 4 It is an example diagram of the relationship between the fault category and the fault cause provided in the embodiments of the present invention;

[0027] Figure 5A flowchart of a method for preprocessing timeout fault characteristics and corresponding characteristic values corresponding to fault categories, fault causes, and fault codes respectively provided in an embodiment of the present invention;

[0028] Figure 6 An example diagram of timeout fault characteristics, corresponding characteristic values, and a combined item of integrated timeout fault characteristics and characteristic values provided in an embodiment of the present invention;

[0029] Figure 7 A flowchart of a method for obtaining training samples corresponding to fault categories, fault causes, and fault codes respectively according to the preprocessing results corresponding to fault categories, fault causes, and fault codes provided in an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of the decision principle of a timeout fault decision model corresponding to a fault category provided in an embodiment of the present invention;

[0031] Figure 9 A schematic diagram of a train timeout fault decision device provided in an embodiment of the present invention. Detailed implementation manners

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Herein, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention. The term "and / or" herein merely describes an association relationship and means that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one or any combination of at least two of a plurality. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set composed of A, B, and C.

[0033] In the description of this specification, the terms "include", "comprise", "have", "contain", etc. are all open-ended terms, that is, they are meant to include but not limited to. The descriptions with reference to terms such as "an embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps therein is not limited and can be adjusted appropriately as needed.

[0034] It has been found through research that the CTCS-3 timeout fault means that when the train uses the CTCS-3 train control system to control the train operation, the train triggers the service brake due to the timeout of the wireless communication in the GSM-R network (also called the interruption of vehicle-ground communication). The frequent deceleration and acceleration processes will affect the passenger riding comfort, and the resulting train delays will also reduce the railway operation efficiency. Therefore, the research on timeout faults is of great significance. How to determine the fault attribution and quickly locate the fault point is the key and also one of the difficulties in the CTCS-3 timeout fault handling process. The wireless communication bearer network of the train control system is the GSM-R network. According to the vehicle-ground communication rules of the train control system and the basic communication principles of the GSM-R network, the interface data of the Abis interface, A interface, and PRI interface of the GSM-R network are monitored, analyzed and processed, and the features related to the CTCS-3 timeout fault are extracted to provide a basis for the judgment of the CTCS-3 timeout fault category and the analysis of the fault cause. Each interface provides various features or feature groups for the judgment of the timeout fault category and the analysis of the fault cause. The PRI interface is divided into the application layer, security layer, transport layer, network layer, link layer, and physical layer. The A interface undertakes the messages between the BSC and the MSC, and between the MT and the MSC. The main features in the Abis interface signaling are the handover signaling and the connection release signaling. Since there are many levels and processes of network data signaling data, generally a group of features corresponds to a fault cause. When the number of features and combinations is large, it will be very difficult to organize and combine these features into a system logic. Especially when new feature combinations are continuously added during the wireless optimization, it can no longer meet the requirements to complete the system logic integration through manual methods.

[0035] In view of the above research, an embodiment of the present invention provides a method for judging train timeout faults, as Figure 1 shown, including:

[0036] S101: Extract the feature values corresponding to each timeout fault feature from the historical interface data of the railway integrated digital mobile communication system GSM-R network;

[0037] S102: Preprocess each timeout fault feature and the feature values corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively;

[0038] S103: Using the decision tree as the model and adopting the classification and regression tree CART algorithm, train the model respectively with the training samples corresponding to the fault category, fault cause, and fault code to obtain the timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively;

[0039] S104: Input the interface data of the GSM-R network corresponding to the unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively, to obtain the corresponding fault category decision result, fault cause decision result, and fault code decision result;

[0040] S105: Obtain the decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result.

[0041] In the embodiment of the present invention, the eigenvalue corresponding to each timeout fault feature is extracted from the historical interface data of the GSM-R network; preprocess each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain the training samples corresponding to the fault category, fault cause, and fault code respectively; use the decision tree as the model and adopt the CART algorithm to train the model respectively with the training samples corresponding to the fault category, fault cause, and fault code to obtain the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively; input the interface data of the GSM-R network corresponding to the unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively to obtain the corresponding fault category decision result, fault cause decision result, and fault code decision result; obtain the decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result. In this way, the timeout fault decision models corresponding to the fault category, fault cause, and fault code are used to judge the fault attribution of the CTCS-3 timeout fault, which improves the efficiency and accuracy of judging the fault attribution of the CTCS-3 timeout fault; and the timeout fault decision model uses the decision tree as the model, and the output logic has a high degree of visualization and is easy to understand.

[0042] The above S101 to S105 will be described in detail below.

[0043] Regarding the above S101, the historical interface data of the GSM-R network includes, for example: the historical interface data of the Abis interface, the historical interface data of the A interface, the historical interface data of the PRI interface, etc.

[0044] The timeout fault features include, for example, but are not limited to at least one of the following: ABNORMAL_UPQUAL (whether the uplink quality deteriorates before the link is disconnected), CURRENT_CALL_SUCCESS (the current call is successful or failed), HOEND_CELL_MT2_NORMAL (the measurement report of the target cell communicating with the standby radio is normal or abnormal), LAST_NEXT_CALL_ABNORMAL (whether there is a call with a very short service time near the abnormal event), etc.

[0045] It should be noted here that the timeout fault characteristics can be customized according to the actual application scenarios, and can also be continuously updated and added in combination with the actual application scenarios. This is just an example for illustration and does not mean that the timeout fault characteristics described in this application only include the above several types.

[0046] For the above S102, the fault category, fault cause, and fault code can be pre-customized according to the actual application scenarios, which respectively show the reasons for the train timeout fault from different dimensions, such as Figure 2 As shown, it is a method flowchart for preprocessing each timeout fault characteristic and the corresponding characteristic value to obtain training samples corresponding to the fault category, fault cause, and fault code provided by an embodiment of the present invention, including:

[0047] S201: Configure the association relationships between the fault category, fault cause, fault code, and timeout fault characteristics respectively.

[0048] Specifically, as Figure 3 As shown, it is a method flowchart for configuring the association relationships between the fault category, fault cause, fault code, and timeout fault characteristics respectively provided by an embodiment of the present invention, including:

[0049] S301: Configure multiple fault categories and the corresponding fault causes for each fault category; wherein, each fault category corresponds to at least one fault cause.

[0050] Exemplarily, the fault categories include, for example: on-vehicle faults, GSM-R radio network faults, GSM-R core network faults, wireless interference faults, wireless network optimization faults, Radio Block Centre (RBC) faults, other communication faults, unknown cause faults, etc.

[0051] Among them, for example Figure 4As shown in the figure, the causes corresponding to vehicle-mounted faults include, for example: MT hardware problems, single MT problems, MT software problems, other problems with the MT module, loose Subscriber Identity Module (SIM) cards, other vehicle-mounted problems, etc.; the causes corresponding to GSM-R radio network faults include, for example: combiner faults, carrier frequency board faults, cross-BSC handover failures, Abis interface time slot faults, base station hibernation, etc.; the causes corresponding to GSM-R core network faults include, for example: MSC disconnection, MSC faults, other core network faults, etc.; the causes corresponding to radio interference faults include, for example: burst interference, operator base station wideband - blocking - interference, etc.; the causes corresponding to radio network optimization faults include, for example: multipath interference, handover problems, etc.; the causes corresponding to RBC faults include, for example: RBC-RBC communication timeout during handover, failure to send data to the vehicle, RBC not sending application layer data, etc.; the causes corresponding to other communication faults include, for example: missing interface monitoring data, etc.; the causes corresponding to unknown cause faults include, for example: incorrect packets of Movement Authority (MA) due to cell handover, the vehicle judges incorrect High-Level Data Link Control (HDLC) frame type or timing, and sends Frame Reject (FRMR), RBC judges incorrect HDLC frame type or timing, and sends FRMR, RBC judges incorrect Transport Protocol Data Unit (TPDU) frame type, and sends ER (Error), the vehicle judges incorrect TPDU frame type, and sends ER, and the vehicle's secure transport layer sends Disconnect Indication / Disconnect Request (DI / DR), etc.

[0052] S302: Configure the timeout fault characteristics corresponding to each fault cause; where each fault cause corresponds to multiple timeout fault characteristics, and the same timeout fault characteristic corresponds to multiple fault causes.

[0053] Among them, the fault category and fault cause are caused by abnormal characteristic values corresponding to the timeout fault characteristics. The timeout fault characteristics affecting different fault causes are not completely the same, that is, one fault cause corresponds to multiple fault characteristics, and one fault characteristic can also affect multiple fault causes, and specific settings can be made in combination with the actual scenario.

[0054] Exemplarily, for the fault cause: MT software problem, it corresponds to timeout fault characteristics such as ABNORMAL_UPQUAL (whether the uplink quality deteriorates before chain disconnection), CURRENT_CALL_SUCCESS (current call success or failure), HOEND_CELL_MT2_NORMAL (normal or abnormal measurement report of the target cell communicating with the standby radio), LAST_NEXT_CALL_ABNORMAL (whether there is a call with a very short service time near the abnormal event), etc. However, the timeout fault characteristic ABNORMAL_UPQUAL also corresponds to the fault cause: MSC disconnection.

[0055] S303: Configure multiple fault codes and the timeout fault characteristics corresponding to each fault code; wherein, each fault code corresponds to multiple timeout fault characteristics, and the same timeout fault characteristic corresponds to multiple fault codes.

[0056] Specifically, the fault code can also be used to analyze the timeout fault characteristic. Different fault codes can represent different meanings. The fault code can be custom-configured in combination with the actual application scenario. There are multiple timeout fault characteristics affecting each fault code, and the same timeout fault characteristic can affect multiple fault codes.

[0057] S202: Respectively obtain the timeout fault characteristics and corresponding characteristic values corresponding to the fault category, fault cause, and fault code according to the association relationships among the fault category, fault cause, fault code, and timeout fault characteristics, as well as each timeout fault characteristic and the characteristic value corresponding to each timeout fault characteristic.

[0058] S203: Respectively preprocess the timeout fault characteristics and corresponding characteristic values corresponding to the fault category, fault cause, and fault code.

[0059] Specifically, as Figure 5 shown, it is a method flowchart for respectively preprocessing the timeout fault characteristics and corresponding characteristic values corresponding to the fault category, fault cause, and fault code provided by an embodiment of the present invention, including:

[0060] S501: Respectively encode the fault category, fault cause, and fault code in a one-hot manner.

[0061] Exemplarily, encoding is performed in a one-hot manner. Each fault cause corresponds to one encoding, each fault category corresponds to one encoding, and each fault code also corresponds to one encoding.

[0062] S502: Respectively integrate the timeout fault characteristics and corresponding characteristic values corresponding to the fault category, fault cause, and fault code to obtain a combination item of the timeout fault characteristic and the characteristic value.

[0063] Exemplarily, as Figure 6As shown in the figure, it is an example diagram of a timeout fault feature, corresponding feature value, and the combined item of the integrated timeout fault feature and feature value provided by the embodiment of the present invention. The timeout fault feature CONFIG_NAME includes: A_ABNORMAL_REASON_CODE (the interruption reason for the MSC to initiate a release as shown in the A interface signaling), ABIS_HOFAIL_THEN_BACK (abnormal backhaul due to handover failure), A_CLEAR_REQUEST_CAUSE (the reason for the explicit request on the A interface), and their corresponding feature values CONFIG_VALUE are '3', 'yes', 'RESOURCE UNAVAILABLE / EQUIPMENT FAILURE (resource unavailable or equipment failure)' respectively. Then, the timeout fault feature and the corresponding feature value are integrated to obtain the corresponding combined item of the timeout fault feature and feature value CONFIG_NAME_VALUE, including: A_ABNORMAL_REASON_CODE = '3', ABIS_HOFAIL_THEN_BACK = 'yes', A_CLEAR_REQUEST_CAUSE = 'RESOURCE UNAVAILABLE / EQUIPMENT FAILURE'.

[0064] S204: Respectively obtain the training samples corresponding to the fault category, fault cause, and fault code according to the preprocessing results corresponding to the fault category, fault cause, and fault code.

[0065] Specifically, as Figure 7 shown in the figure, it is a method flowchart for respectively obtaining the training samples corresponding to the fault category, fault cause, and fault code according to the preprocessing results corresponding to the fault category, fault cause, and fault code provided by the embodiment of the present invention, including:

[0066] S701: Obtain the possibility level of each code causing the train timeout fault according to the number of combined items of the timeout fault feature and feature value corresponding to each fault category, fault cause, and fault code that cause the train timeout fault.

[0067] S702: Obtain the training samples corresponding to the fault category, fault cause, and fault code according to the combined items of the timeout fault feature and feature value corresponding to each code of the fault category, fault cause, and fault code, and the possibility level.

[0068] For the above S103, after obtaining the training samples, train the model respectively using the training samples corresponding to the fault category, fault cause, and fault code to obtain the timeout fault decision models corresponding to the fault category, fault cause, and fault code.

[0069] Exemplarily, taking the timeout fault judgment model corresponding to the fault category as an example, as Figure 8 shown, it is a schematic diagram of the judgment principle of the timeout fault judgment model corresponding to a fault category provided by an embodiment of the present invention. Among them, at each node of the model, discrimination is performed according to the corresponding timeout fault feature and the feature value combination item to determine the fault cause corresponding to the train timeout fault. Each node will make a judgment according to a timeout fault feature, and then determine whether to enter the next node for continuous judgment or directly obtain the corresponding fault category according to the judgment result. For example, the first node makes a judgment on the timeout fault feature of "ALM_NMS (the alarm information contains feature values such as a combiner or an MTP3 link, etc.)". If it is determined that there is no problem with this timeout fault feature, it enters the second node for continuous judgment. If there is a problem with the first timeout fault feature, the corresponding fault category is directly obtained as a GSM-R wireless network fault.

[0070] For the above S104 - S105, after obtaining the timeout fault judgment models corresponding to the fault category, fault cause, and fault code respectively, the obtained timeout fault judgment models can be used to analyze the interface data of the GSM-R network corresponding to the unknown train timeout fault. The interface data of the GSM-R network corresponding to the unknown train timeout fault is respectively input into the timeout fault judgment models corresponding to the fault category, fault cause, and fault code to obtain the corresponding fault category judgment result, fault cause judgment result, and fault code judgment result. The judgment result of the unknown train timeout fault is obtained according to the fault category judgment result, the fault cause judgment result, and the fault code judgment result.

[0071] In an embodiment of the present invention, when applying the trained timeout fault judgment model, corresponding judgment conditions can be generated according to the trained timeout fault judgment model, and the judgment conditions are output to the judgment process of the database to complete the production of judgment fault classification data.

[0072] Here, for example, the SQL method can be used to generate corresponding judgment conditions, which is the process of loading the trained timeout fault model into a specific application.

[0073] An embodiment of the present invention also provides a train timeout fault judgment device as described in the following embodiment. Since the principle of the device to solve the problem is similar to the train timeout fault judgment method, the implementation of the device can refer to the implementation of the train timeout fault judgment method, and the repeated parts will not be described again.

[0074] As Figure 9 shown, it is a schematic diagram of a train timeout fault judgment device provided by an embodiment of the present invention, including: an extraction module 901, a first processing module 902, a model training module 903, a second processing module 904, and a third processing module 905; among them,

[0075] An extraction module 901, configured to extract the feature values corresponding to each timeout fault feature from the historical interface data of the GSM-R network;

[0076] A first processing module 902, configured to preprocess each timeout fault feature and the feature values corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively;

[0077] A model training module 903, configured to use a decision tree as the model, adopt the CART algorithm, and respectively use the training samples corresponding to the fault category, fault cause, and fault code to train the model to obtain timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively;

[0078] A second processing module 904, configured to input the interface data of the GSM-R network corresponding to an unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively, to obtain corresponding fault category decision results, fault cause decision results, and fault code decision results;

[0079] A third processing module 905, configured to obtain a decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result.

[0080] In a possible implementation manner, the first processing module is specifically configured to respectively configure the association relationships between the fault category, the fault cause, the fault code, and the timeout fault feature; respectively obtain the timeout fault features and the corresponding feature values corresponding to the fault category, the fault cause, and the fault code according to the association relationships between the fault category, the fault cause, the fault code, and the timeout fault feature, and each timeout fault feature and the feature values corresponding to each timeout fault feature; respectively preprocess the timeout fault features and the corresponding feature values corresponding to the fault category, the fault cause, and the fault code; and respectively obtain the training samples corresponding to the fault category, the fault cause, and the fault code according to the preprocessing results corresponding to the fault category, the fault cause, and the fault code.

[0081] In a possible implementation manner, the first processing module is specifically configured to configure multiple fault categories and the fault causes corresponding to each fault category; wherein, each fault category corresponds to at least one fault cause; configure the timeout fault features corresponding to each fault cause; wherein, each fault cause corresponds to multiple timeout fault features, and the same timeout fault feature corresponds to multiple fault causes; configure multiple fault codes and the timeout fault features corresponding to each fault code; wherein, each fault code corresponds to multiple timeout fault features, and the same timeout fault feature corresponds to multiple fault codes.

[0082] In a possible implementation, the first processing module is specifically configured to encode the fault category, fault cause, and fault code in a one-hot manner respectively; integrate the timeout fault features and corresponding feature values respectively corresponding to the fault category, fault cause, and fault code to obtain a combination item of timeout fault features and feature values.

[0083] In a possible implementation, the first processing module is specifically configured to obtain the possibility level of each encoding causing a train timeout fault according to the number of combination items of timeout fault features and feature values respectively corresponding to each fault category, fault cause, and fault code that cause the train timeout fault; obtain the training samples respectively corresponding to the fault category, fault cause, and fault code according to the combination items of timeout fault features and feature values corresponding to each encoding of the fault category, fault cause, and fault code, and the possibility level.

[0084] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned train timeout fault judgment method is implemented.

[0085] An embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the above-mentioned train timeout fault judgment method is implemented.

[0086] An embodiment of the present invention further provides a computer program product. The computer program product includes a computer program. When the computer program is executed by a processor, the above-mentioned train timeout fault judgment method is implemented.

[0087] In the embodiments of the present invention, eigenvalue corresponding to each timeout fault feature is extracted from the historical interface data of the GSM-R network; preprocessing is performed on each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, the fault cause, and the fault code respectively; using a decision tree as the model and adopting the CART algorithm, the model is trained respectively by using the training samples corresponding to the fault category, the fault cause, and the fault code to obtain timeout fault decision models corresponding to the fault category, the fault cause, and the fault code respectively; the interface data of the GSM-R network corresponding to the unknown train timeout fault is respectively input into the timeout fault decision models corresponding to the fault category, the fault cause, and the fault code to obtain corresponding fault category decision results, fault cause decision results, and fault code decision results; the decision result of the unknown train timeout fault is obtained according to the fault category decision result, the fault cause decision result, and the fault code decision result. In this way, the timeout fault decision models corresponding to the fault category, the fault cause, and the fault code are used to judge the fault attribution of the CTCS-3 timeout fault, which improves the efficiency and accuracy of judging the fault attribution of the CTCS-3 timeout fault; and the timeout fault decision model uses a decision tree as the model, and the output logic has a high degree of visualization and is easy to understand.

[0088] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0090] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes and / or blocks Figure 1 in one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes and / or blocks Figure 1 in one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0092] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for judging train overtime faults, characterized in that, Including: Extracting the eigenvalue corresponding to each timeout fault feature from the historical interface data of the GSM-R network of the railway integrated digital mobile communication system; Preprocessing each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively; Using a decision tree as the model and adopting the Classification and Regression Tree (CART) algorithm, training the model respectively with the training samples corresponding to the fault category, fault cause, and fault code to obtain timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively; Inputting the interface data of the GSM-R network corresponding to the unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively to obtain the corresponding fault category decision result, fault cause decision result, and fault code decision result; Obtaining the decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result; Preprocessing each timeout fault feature and the eigenvalue corresponding to each timeout fault feature to obtain training samples corresponding to the fault category, fault cause, and fault code respectively, including: Configuring the association relationships between the fault category, fault cause, fault code and the timeout fault features respectively; Respectively obtaining the timeout fault features and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code according to the association relationships between the fault category, fault cause, fault code and the timeout fault features, as well as each timeout fault feature and the eigenvalue corresponding to each timeout fault feature; Preprocessing the timeout fault features and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code respectively; Respectively obtaining the training samples corresponding to the fault category, fault cause, and fault code according to the preprocessing results corresponding to the fault category, fault cause, and fault code.

2. The train timeout fault judgment method according to claim 1, wherein Configuring the association relationships between the fault category, fault cause, fault code and the timeout fault features respectively, including: Configuring multiple fault categories and the fault cause corresponding to each fault category; wherein, each fault category corresponds to at least one fault cause; Configuring the timeout fault feature corresponding to each fault cause; wherein, each fault cause corresponds to multiple timeout fault features, and the same timeout fault feature corresponds to one or more fault causes; Configuring multiple fault codes and the timeout fault feature corresponding to each fault code; wherein, each fault code corresponds to multiple timeout fault features, and the same timeout fault feature corresponds to one or more fault codes.

3. The train timeout fault judgment method according to claim 1, characterized in that Preprocessing the timeout fault features and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code respectively, including: Encoding the fault category, fault cause, and fault code respectively in a one-hot manner; Integrating the timeout fault features and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code to obtain a timeout fault feature and eigenvalue combination item.

4. The train timeout fault judgment method according to claim 3, characterized in that Respectively obtaining the training samples corresponding to the fault category, fault cause, and fault code according to the preprocessing results corresponding to the fault category, fault cause, and fault code, including: According to the number of timeout fault characteristics and eigenvalue combination items corresponding to each fault category, fault cause, and fault code that respectively lead to train timeout faults, obtain the likelihood level of each code leading to train timeout faults; According to the timeout fault characteristics and eigenvalue combination items corresponding to each code of the fault category, fault cause, and fault code, as well as the likelihood level, obtain the training samples corresponding to the fault category, fault cause, and fault code respectively.

5. A train timeout fault judgment device, characterized in that, Including: An extraction module, configured to extract the eigenvalues corresponding to each timeout fault characteristic from the historical interface data of the GSM-R network; A first processing module, configured to preprocess each timeout fault characteristic and the eigenvalues corresponding to each timeout fault characteristic to obtain the training samples corresponding to the fault category, fault cause, and fault code respectively; A model training module, configured to use a decision tree as the model, adopt the CART algorithm, and respectively use the training samples corresponding to the fault category, fault cause, and fault code to train the model to obtain the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively; A second processing module, configured to input the interface data of the GSM-R network corresponding to an unknown train timeout fault into the timeout fault decision models corresponding to the fault category, fault cause, and fault code respectively to obtain the corresponding fault category decision result, fault cause decision result, and fault code decision result; A third processing module, configured to obtain the decision result of the unknown train timeout fault according to the fault category decision result, the fault cause decision result, and the fault code decision result; The first processing module is specifically configured to respectively configure the association relationships between the fault category, fault cause, and fault code and the timeout fault characteristics; According to the association relationships between the fault category, fault cause, and fault code and the timeout fault characteristics, as well as each timeout fault characteristic and the eigenvalues corresponding to each timeout fault characteristic, obtain the timeout fault characteristics and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code respectively; Respectively preprocess the timeout fault characteristics and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code; According to the preprocessing results corresponding to the fault category, fault cause, and fault code respectively, obtain the training samples corresponding to the fault category, fault cause, and fault code respectively.

6. The train timeout fault judgment device according to claim 5, wherein, The first processing module is specifically configured to configure multiple fault categories and the fault causes corresponding to each fault category; wherein, each fault category corresponds to at least one fault cause; Configure the timeout fault characteristics corresponding to each fault cause; wherein, each fault cause corresponds to multiple timeout fault characteristics, and the same timeout fault characteristic corresponds to multiple fault causes; Configure multiple fault codes and the timeout fault characteristics corresponding to each fault code; wherein, each fault code corresponds to multiple timeout fault characteristics, and the same timeout fault characteristic corresponds to multiple fault codes.

7. The train timeout fault judgment device according to claim 5, characterized in that, The first processing module is specifically configured to respectively encode the fault category, fault cause, and fault code in a one-hot manner; Integrate the timeout fault characteristics and the corresponding eigenvalues corresponding to the fault category, fault cause, and fault code respectively to obtain the timeout fault characteristic and eigenvalue combination items.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images