Multi-dimensional risk analysis method and device for motor train unit
Through the multi-dimensional risk analysis method, EMU data is obtained from multiple data sources, and advanced repair cycles are associated, and analysis is carried out based on year and advanced repair dimensions. The problems of insufficient data integration and single analysis methods in the existing technology are solved, and the problem of Comprehensive risk analysis of EMU failures and accidents improves analysis efficiency and accuracy.
Patent Information
- Application Number
- CN202410849798.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-08
AI Technical Summary
In the analysis of EMU faults and accidents, data integration and specifications are insufficient, and the analysis methods are single, which makes it difficult to reflect the fault laws during the entire life cycle of the vehicle. The traditional year dimension calculation is not enough to analyze the fault evolution laws from the essential level of mobile equipment.
Using multi-dimensional risk analysis method, multi-source data of EMUs is obtained from multiple data sources, sub-dimensional labels are determined, advanced repair cycles are associated, risk analysis is carried out based on year and advanced repair dimensions, detailed fault and accident analysis results are generated, and detailed fault and accident analysis results are provided. Visualize to show risk trends.
It realizes automated analysis of EMU risks, improves data processing and analysis efficiency, provides massive data aggregation and integration across systems and spaces, and multi-dimensional risk analysis, revealing the failures and accidents of vehicles in different years and advanced repairs. The rules provide data support for the safety and maintenance strategies of EMUs.
Smart Images

Figure CN118916801A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of EMU risk analysis, and more specifically, to a multi-dimensional risk analysis method and device for EMUs. Background Art
[0002] In order to meet the safety needs and risk prevention of modern railways, all high-speed EMU failures and accidents must be recorded and converted into data. The National Railway Group, railway bureaus, EMU sections and other users analyze these data to provide reliable technical support for EMU operation safety.
[0003] However, when analyzing these data, it is necessary to obtain different data from multiple sources in a scattered manner, and there is a lack of professional integration and standardization; and the analysis method is relatively simple, and the analysis results are difficult to reflect the failure patterns throughout the life cycle of the vehicle. Summary of the invention
[0004] In view of this, the purpose of the present application is to provide a multi-dimensional risk analysis method and device for EMUs, which can comprehensively analyze EMU risks from multiple dimensions based on massive data and reflect the failure patterns of the vehicle throughout its life cycle.
[0005] The present application provides a multi-dimensional risk analysis method for a train set, the method comprising the following steps:
[0006] Acquire multi-source EMU data for the EMU from pre-built EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU;
[0007] Determine a sub-dimension label for each of the multi-source EMU data, and determine a plurality of sub-dimensions according to the sub-dimension label;
[0008] Associating the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determining the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle;
[0009] Based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, the risk analysis results of the EMU in the multiple sub-dimensions are determined from the year dimension and the advanced maintenance dimension respectively; the risk analysis results include fault analysis results and accident analysis results.
[0010] In some embodiments, in the multi-dimensional risk analysis method for EMUs, determining a sub-dimensional label for each of the multi-source EMU data, and determining a plurality of sub-dimensions according to the sub-dimensional label, includes:
[0011] Determine a vehicle model platform label, a function system label and at least one target category label as sub-dimension labels; the target category label is customized according to business requirements;
[0012] According to each sub-dimension label, determine the first sub-dimension corresponding to each sub-dimension label;
[0013] According to the first sub-dimension corresponding to each sub-dimension label and the superior-subordinate relationship between different first sub-dimensions, the first sub-dimensions with the superior-subordinate relationship are combined to obtain the second sub-dimension;
[0014] The first sub-dimension and the second sub-dimension are determined as the multiple sub-dimensions; wherein the multiple sub-dimensions include a vehicle platform dimension, a target category dimension, a functional system dimension, a functional system dimension under the vehicle platform dimension, and a functional system dimension under the target category dimension.
[0015] In some embodiments, in the multi-dimensional risk analysis method for EMUs, based on the time attribute of the multi-source EMU data, the risk analysis results of the EMUs under the multiple sub-dimensions are determined from the year dimension, including:
[0016] Based on the time attribute of the multi-source EMU data, annual failure statistics and annual accident statistics of multiple sub-dimensions under the year dimension are determined; the annual failure statistics of the sub-dimensions include the number of failures and annual mileage of the sub-dimensions in multiple years; the annual accident statistics of the sub-dimensions include the number of accidents and accident mileage of the sub-dimensions in multiple years;
[0017] Based on the annual failure statistics results and annual accident statistics results of the multiple sub-dimensions under the year dimension, determine the annual failure analysis results and the annual accident analysis results of the multiple sub-dimensions under the year dimension;
[0018] The annual failure analysis result summarizes the failure rates per million kilometers of multiple years; the failure rates per million kilometers of the year are determined based on the number of failures in the year, the annual mileage and the first conversion coefficient;
[0019] The accident analysis results for the year summarize the accident rates per million kilometers for multiple years; the accident rates per million kilometers for the year are determined based on the number of accidents in that year, the annual mileage and the second conversion factor.
[0020] In some embodiments, in the multi-dimensional risk analysis method for EMUs, based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions are determined from the advanced maintenance dimension, including:
[0021] Determine the multi-source EMU data corresponding to the advanced maintenance cycle based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle;
[0022] Based on the advanced maintenance data, determine whether the EMU has actually completed advanced maintenance within a given period after the fault occurs;
[0023] If it is not repaired, the fault data or accident data is removed from the multi-source EMU data, and the multi-source EMU data is updated. The updated multi-source EMU data includes fault samples and accident samples;
[0024] Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions are determined from the advanced maintenance dimension.
[0025] In some embodiments, in the multi-dimensional risk analysis method for an EMU, based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions are determined from the advanced maintenance dimension, including:
[0026] Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the advanced maintenance failure statistics and advanced maintenance accident statistics of multiple sub-dimensions under the advanced maintenance dimension are respectively determined; the advanced maintenance failure statistics of the sub-dimensions include the number of advanced maintenance failure samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension; the advanced maintenance accident statistics of the sub-dimensions include the number of advanced maintenance accident samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension;
[0027] Based on the advanced repair failure statistics results and advanced repair accident statistics results of multiple sub-dimensions under the advanced repair dimension, determine the advanced repair failure analysis results and advanced repair accident analysis results of multiple sub-dimensions under the advanced repair dimension;
[0028] The high-level maintenance failure analysis results are summarized with failure rates per million kilometers of multiple speed levels; the failure rates per million kilometers of the speed level are determined based on the number of high-level maintenance failure samples of the speed level, the annual mileage and the third conversion factor;
[0029] The high-level maintenance accident analysis results summarize accident rates per million kilometers at multiple speed levels; the accident rates per million kilometers at each speed level are determined based on the number of high-level maintenance accident samples at that speed level, the annual mileage, and a fourth conversion factor.
[0030] In some embodiments, in the multi-dimensional risk analysis method for EMUs, the annual mileage of the advanced maintenance dimension is calculated by the following method:
[0031] Determine the number of standard groups of the vehicle model platform according to the number of groups of specific vehicle models under the vehicle model platform and the preset standard group conversion coefficient corresponding to the group;
[0032] When the advanced maintenance cycle intervals corresponding to the target category dimensions are different, determining the estimated coefficient of annual mileage based on the advanced maintenance cycle intervals corresponding to the target category dimensions;
[0033] The product of the estimated coefficient and the number of standard groups of the vehicle platform is taken as the annual mileage of the advanced repair.
[0034] In some embodiments, in the multi-dimensional risk analysis method for a train set, after obtaining the risk analysis result for the train set, the method further includes:
[0035] Visualize the risk analysis results of multiple sub-dimensions under the year dimension and advanced repair dimension;
[0036] Taking the year dimension or the advanced repair dimension as the X-axis, taking the failure rate per million kilometers in the risk analysis result corresponding to the dimension as the Y-axis, respectively determining the broken line marks corresponding to different sub-dimensions, and generating a two-dimensional chart of the risk analysis result of the year dimension or the advanced repair dimension;
[0037] and / or,
[0038] The year dimension or advanced repair dimension is used as the X-axis, the failure rate per million kilometers in the risk analysis result corresponding to the dimension is used as the Y-axis, and different sub-dimensions are corresponded to the Z-axis to generate a three-dimensional chart of the risk analysis results of the year dimension or advanced repair dimension.
[0039] In some embodiments, the multi-dimensional risk analysis method for EMUs further includes:
[0040] In response to a selection operation on a year dimension or an advanced repair dimension in a graphical user interface, displaying the fault analysis results and / or accident analysis results of the EMU under the multiple sub-dimensions from the year dimension or the advanced repair dimension selected by the user;
[0041] In response to an expansion operation on the displayed fault analysis result and / or accident analysis result, the fault statistical result corresponding to the fault analysis result and / or the accident statistical result corresponding to the accident analysis result are displayed.
[0042] In some embodiments, in the multi-dimensional risk analysis method for EMUs, obtaining multi-source EMU data for EMUs from pre-constructed EMU multi-type data sources includes:
[0043] Acquire original EMU data from a pre-built EMU multi-type data source, and aggregate the operation data, fault data, and accident data of each scenario in the original EMU data as original multi-source EMU data;
[0044] Deduplication is performed on the original multi-source EMU data of the EMU to obtain deduplication-free multi-source EMU data;
[0045] The multi-source EMU data after deduplication is reviewed and located, and the multi-source EMU data is updated according to the functional system in the standard dictionary table to obtain updated multi-source EMU data.
[0046] In some embodiments, a multi-dimensional risk analysis device for a train set is also provided, the device comprising:
[0047] An acquisition module, used to acquire multi-source EMU data for an EMU from pre-built EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU;
[0048] A first determination module is used to determine a sub-dimension label of each of the multi-source EMU data, and determine a plurality of sub-dimensions according to the sub-dimension label;
[0049] A second determination module is used to associate the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determine the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle;
[0050] The third determination module is used to determine the risk analysis results of the EMU in the multiple sub-dimensions from the year dimension and the advanced maintenance dimension respectively based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle; the risk analysis results include fault analysis results and accident analysis results.
[0051] In an embodiment of the present application, a multi-dimensional risk analysis method and device for an EMU are provided; the multi-dimensional risk analysis method for an EMU obtains multi-source EMU data for the EMU from pre-constructed EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU; a sub-dimension label of each multi-source EMU data is determined, and a plurality of sub-dimensions are determined according to the sub-dimension label; the multi-source EMU data of the EMU is associated with the advanced maintenance data of the EMU, and the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle is determined; based on the time attribute of the multi-source EMU data and the association relationship between the multi-source EMU data and the advanced maintenance cycle, the multi-source EMU data is associated with the advanced maintenance cycle from the year dimension and the The advanced maintenance dimension determines the EMU under the multiple sub-dimensions to obtain risk analysis results for the EMU, and the risk analysis results include fault analysis results and accident analysis results. Based on this, the present application realizes the automated analysis of EMU risks in the year dimension and the advanced maintenance process dimension, greatly improves the efficiency of data processing and analysis, and provides cross-system and cross-space massive data convergence and fusion, statistical analysis, and risk analysis in more dimensions. From the two perspectives of the EMU year dimension and the advanced maintenance dimension, the accident and failure patterns of each subsystem of EMUs of different models in each year and advanced maintenance process are studied, providing data verification support for the extension of the maintenance process of the EMU and its subsystems, optimizing and strengthening the advanced maintenance projects for subsystems that have not been paid attention to, and providing a basis for "saving costs and reducing consumption". BRIEF DESCRIPTION OF THE DRAWINGS
[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0053] Figure 1 A flow chart of a multi-dimensional risk analysis method for a motor vehicle group according to an embodiment of the present application is shown;
[0054] Figure 2 A flow chart of a method for obtaining multi-source EMU data for an EMU according to an embodiment of the present application is shown;
[0055] Figure 3 A flow chart of a method for determining multiple sub-dimensions according to an embodiment of the present application is shown;
[0056] Figure 4 A flow chart of a method for determining the risk analysis results of the EMU under the multiple sub-dimensions from the year dimension in an embodiment of the present application is shown;
[0057] Figure 5 A line graph showing the failure rate per million kilometers in the year dimension-target category dimension described in the embodiment of the present application is shown;
[0058] Figure 6 A line graph showing the failure rate per million kilometers of the functional system in the year dimension-target category one according to the embodiment of the present application is shown;
[0059] Figure 7 A statistical line chart of the year dimension-vehicle model platform-annual million kilometers failure rate described in the embodiment of the present application is shown;
[0060] Figure 8 A three-dimensional bar chart of the year dimension-model platform-annual million-kilometer failure rate described in the embodiment of the present application is shown;
[0061] Fig. 9 A flow chart of a method for determining the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions from a high-level maintenance dimension according to an embodiment of the present application is shown;
[0062] Fig.10 A line chart showing the failure rate statistics per million kilometers in the advanced maintenance dimension-target category dimension described in an embodiment of the present application is shown;
[0063] Fig.11 A line chart showing the failure rate statistics of one million kilometers in the advanced repair dimension-vehicle platform dimension described in an embodiment of the present application is shown;
[0064] Fig.12 A structural schematic diagram of a multi-dimensional risk analysis device for EMUs described in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0065] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.
[0066] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.
[0067] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.
[0068] In order to meet the safety needs and risk prevention of modern railways, all high-speed EMU failures and accidents must be recorded and converted into data. The National Railway Group, railway bureaus, EMU sections and other users analyze these data to provide reliable technical support for EMU operation safety.
[0069] However, when analyzing these data, it is necessary to obtain different data from multiple sources in a scattered manner, and there is a lack of professional integration and standardization; and the analysis method is relatively simple, and the analysis results are difficult to reflect the failure patterns throughout the life cycle of the vehicle.
[0070] Specifically, in the process of analyzing high-speed EMU faults and accidents, the main data sources are the "operating faults" and "maintenance faults" modules in the EMU Management Information System (EMIS), information on train operation, routes, advanced repairs, and assignments; and the accident identification documents and investigation reports of the "Safety Supervision Report 3". These data are stored in a dispersed manner in the system, requiring users to download and integrate them separately, which is very troublesome; and some of the fault numbers have many uncertain factors in the process of data entry and manual classification and labeling, and there are many duplicate entries and irregular records, which also makes it impossible to fully utilize the data that could have provided great value.
[0071] The existing failure rate calculations in different dimensions often have problems such as poor timeliness, large workload, low adaptability, and inconsistent calculation process. Once the data from a certain source changes due to the statistical dimension, the calculation results of each dimension will undergo drastic changes.
[0072] Traditional analysis methods are relatively simple, and they mostly use the accident / failure rate per million kilometers in natural years to analyze the safety laws of high-speed EMUs. The traditional year dimension calculation method is not enough to analyze the evolution of its failures from the essential level of mobile equipment. Different frequencies of use of trains, different operating conditions, etc. will have different degrees of impact on the analysis results. For example, in the early stage of the launch of new models, the operation and maintenance management was not perfect, and the accidents and failures caused by emergency handling and improper operation were high; after the China Railway Group organized various main manufacturers to carry out source quality rectification and strengthen operation control, the typical failure problems of various models were improved, and the safety situation tended to be stable. In other words, when coupling the model and speed level, the mobile equipment itself is specifically analyzed. The traditional analysis method is difficult to reflect the failure law of the vehicle throughout its life cycle, that is, it is impossible to reveal the impact of the advanced maintenance process of the EMU on the safety law.
[0073] At the same time, under the general background of cost saving and consumption reduction, cost reduction and efficiency improvement, it is necessary to systematically analyze the safety laws of EMUs of various platforms (categories), to macro-evaluate the reliability indicators of the whole vehicle, subsystem and even key components, and provide data support for the EMU reliability, availability, maintainability and safety RAMS system and fault prediction and health management (PHM).
[0074] Based on this, in an embodiment of the present application, a multi-dimensional risk analysis method and device for an EMU are provided; the multi-dimensional risk analysis method for an EMU obtains multi-source EMU data for the EMU from a pre-constructed EMU multi-type data source; the multi-source EMU data at least includes the operation data, fault data and accident data of each scenario of the EMU; a sub-dimensional label of each of the multi-source EMU data is determined, and a plurality of sub-dimensions are determined according to the sub-dimensional label; the multi-source EMU data of the EMU is associated with the advanced maintenance data of the EMU, and the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle is determined; based on the time attribute of the multi-source EMU data and the association relationship between the multi-source EMU data and the advanced maintenance cycle, the multi-source EMU data is associated with the advanced maintenance cycle from the year dimension. The year dimension and advanced maintenance dimension respectively determine the EMU under the multiple sub-dimensions to obtain the risk analysis results for the EMU, and the risk analysis results include fault analysis results and accident analysis results; based on this, the present application realizes the automated analysis of EMU risks in the year dimension and advanced maintenance dimension, greatly improves the efficiency of data processing and analysis, and provides cross-system and cross-space massive data aggregation and fusion, statistical analysis, and risk analysis in more dimensions. From the two perspectives of the EMU year dimension and advanced maintenance dimension, the accident and failure patterns of each subsystem of EMUs of different models in each year and advanced maintenance process are studied, providing data verification support for the extension of the maintenance process of the EMU and its subsystems, optimizing and strengthening the advanced maintenance projects for the subsystems that have not been paid attention to, and providing a basis for "saving money and reducing consumption".
[0075] Please refer to Figure 1 , Figure 1 A flowchart of a multi-dimensional risk analysis method for an EMU according to an embodiment of the present application is shown. The multi-dimensional risk analysis method for an EMU comprises the following steps S101-S104:
[0076] S101, acquiring multi-source EMU data for an EMU from pre-constructed EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU;
[0077] S102, determining a sub-dimension label for each of the multi-source EMU data, and determining a plurality of sub-dimensions according to the sub-dimension label;
[0078] S103, associating the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determining the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle;
[0079] S104. Based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, determine the risk analysis results of the EMU under the multiple sub-dimensions from the year dimension and the advanced maintenance dimension respectively; the risk analysis results include fault analysis results and accident analysis results.
[0080] In the step S101, multi-source EMU data for the EMU is obtained from pre-constructed EMU multi-type data sources; the multi-source EMU data at least includes the EMU's operating data, fault data and accident data for each scenario.
[0081] The pre-built EMU multi-type data sources refer to various pre-built information representing data sources, such as assignment, running, history, maintenance faults, driving faults, annual mileage of EMUs and advanced repair details of each model platform in the information management system EMIS; cross-disciplinary (engineering, electrical, etc.) data of "Safety Supervision Report 3" and external environment data.
[0082] Based on the characteristics of the data source system, storage, screening, processing and integration are carried out to form spatial data for building security law analysis models and establish spatial and temporal fields of different dimensions.
[0083] Use ETL tools to extract, clean, transform and load data from various sources into a data platform, establish a standardized data access and sharing interface, and allow data to be securely shared between different systems and organizations while ensuring data integrity and privacy.
[0084] The original EMU data obtained from the data source has problems such as missing, duplicate records, and fault location errors. For example, there are some manual backfilling standardization problems in the driving fault data in the EMIS system, resulting in unsatisfactory data validity: for example, there is a discrepancy between the driving fault phenomenon and the fault cause, and some faults are located based on the fault phenomenon rather than the actual fault cause, resulting in inaccurate positioning of the functional system; in addition, there is also the phenomenon of repeated fault recording, a typical example is that the fault of a double-coupled EMU occurs in one of the cars, but the front and rear cars are recorded once respectively. For another example, in the case of an EMU automatic pantograph lowering failure, the system structure fills in the main power supply system problem; but after the EMU is put into storage for inspection and analysis, it is determined that there is a problem with the network input and output control module, and the fault belongs to the network and auxiliary monitoring system.
[0085] Based on this, please refer to the embodiments of this application. Figure 2 The step of obtaining multi-source EMU data for an EMU from pre-built EMU multi-type data sources includes the following steps S201-S203:
[0086] S201, obtaining original EMU data from a pre-constructed EMU multi-type data source, and summarizing the operation data, fault data and accident data of each scenario in the original EMU data as original multi-source EMU data;
[0087] S202, deduplicating the original multi-source EMU data of the EMU to obtain deduplicated multi-source EMU data;
[0088] S203, reviewing and locating the deduplicated multi-source EMU data, and updating the multi-source EMU data according to the functional system in the standard dictionary table to obtain updated multi-source EMU data.
[0089] The operation data, fault data and accident data of each scenario in the original EMU data are aggregated as the original multi-source EMU data, so as to prevent data missing, especially prevent fault data and accident data missing.
[0090] The original multi-source EMU data of the EMU is deduplicated to solve the problem of duplicate data recording. The deduplicated multi-source EMU data is reviewed and located to correct the problem of irregular fault or accident location. For example, for fault A of the traction system, the staff mistakenly recorded its functional system as the body system.
[0091] Specifically, the multi-source EMU data after deduplication is reviewed and located, that is, the functional system to which the multi-source EMU data belongs is reviewed and located, which can accurately reflect the relationship between faults, accidents and functional systems.
[0092] The standard dictionary table is a dictionary table published by relevant institutions, which records the classification of the EMU system and its subordinate functional components and can review and accurately locate faults.
[0093] Based on the above-mentioned aggregation, deduplication, review and positioning of the original multi-source EMU data, it is ensured that the quantity of multi-source EMU data is accurate, and the relationship between faults, accidents and functional systems is correct, so that the multi-source EMU data used for risk analysis accurately reflects the actual situation during the operation of the EMU, thereby improving the accuracy of the risk analysis results.
[0094] It should be noted that the original EMU data obtained from the pre-built EMU multi-type data source is converted and processed into multi-source EMU data for risk analysis, which not only undergoes the above-mentioned aggregation, deduplication, review and positioning, but also includes data storage, data screening, data processing and other processing.
[0095] Among them, data screening is to analyze the value contained in massive data, including data extraction, data cleaning, data loading and other processes, and finally extract valuable data and remove redundant and invalid data.
[0096] Data processing includes data cleaning, data conversion, data organization and other processes. Data cleaning is to remove duplicate data, missing data, abnormal data, etc., to improve data quality; data conversion is to convert the data format into a format that can be recognized and processed by analysis tools or systems, such as encoding conversion, data structure conversion, etc.; data organization is to organize the data, such as data normalization, data splitting, data merging, etc., to make the data easy to process and analyze.
[0097] In the actual operation process, it is necessary to conduct data mining, governance, processing and other aspects of the data source according to the type, characteristics and needs of the data source, and finally obtain high-quality data, laying the foundation for the subsequent multi-dimensional risk analysis of the EMU.
[0098] Specifically, when data mining, governance, processing, etc. are performed on the data source, the manually classified deviation data is classified and clustered using machine learning text annotation and other means, so as to further standardize the fault data.
[0099] In the step S102, a sub-dimension label of each of the multi-source EMU data is determined, and a plurality of sub-dimensions are determined according to the sub-dimension label.
[0100] Please refer to Figure 3 , Figure 3A flow chart of a method for determining multiple sub-dimensions according to an embodiment of the present application is shown; the sub-dimension label of each of the multi-source EMU data is determined, and multiple sub-dimensions are determined according to the sub-dimension label, including the following steps S301-S304:
[0101] S301, determining a vehicle model platform label, a function system label, and at least one target category label as sub-dimension labels; the target category label is customized according to business requirements;
[0102] S302: Determine, according to each sub-dimension label, a first sub-dimension corresponding to each sub-dimension label;
[0103] S303, according to the first sub-dimension corresponding to each sub-dimension label and the superior-subordinate relationship between different first sub-dimensions, the first sub-dimensions with superior-subordinate relationship are combined to obtain a second sub-dimension;
[0104] S304. Determine the first sub-dimension and the second sub-dimension as the multiple sub-dimensions; wherein the multiple sub-dimensions include a vehicle platform dimension, a target category dimension, a functional system dimension, a functional system dimension under the vehicle platform dimension, and a functional system dimension under the target category dimension.
[0105] In the vehicle platform dimension, the risks of failures, accidents, etc. of multiple vehicle models can be analyzed uniformly;
[0106] Customized target categories can be used to uniformly analyze the risks of a collection of vehicle models or functional systems based on user needs. For example, one target category is a middle-level concept between the vehicle platform and functional system levels.
[0107] Exemplarily, in an embodiment of the present application, the target category dimensions include: category one, category two, category three and category four.
[0108] The functional system dimension can specifically analyze the risks of individual functional systems in the train.
[0109] Exemplarily, the functional system dimensions include: traction system, air supply and braking, network and auxiliary monitoring, bogie and its auxiliary, auxiliary electrical, main power supply, others, and overall.
[0110] As vehicle complexity increases, the need for vehicle safety analysis becomes more detailed and diverse. By dividing sub-dimensions, various business needs can be better met, such as fault analysis, accident prevention, risk assessment, etc.
[0111] By setting multiple sub-dimensions, it is possible to analyze vehicle safety laws from a more granular and comprehensive perspective, rather than just analyzing the overall safety situation of the vehicle. Data from different dimensions provide different analysis perspectives. The vehicle platform dimension can analyze the risks of a certain type of vehicle from a macro perspective. The target category dimension can be customized according to business needs, thereby focusing on the risks of a specific set. The functional system dimension can go deep into the specific system level to analyze the safety of a single functional system. In this way, through the comprehensive analysis of multiple sub-dimensions, it is possible to more comprehensively identify the possible risk points of the vehicle and take corresponding preventive measures to improve the safety of the vehicle.
[0112] Process the multi-source EMU data to determine the sub-dimension label of each of the multi-source EMU data; specifically, classify and cluster by means of machine learning to determine the sub-dimension label of the multi-source EMU data. Label the core data used for EMU fault group analysis to create intermediate variables required for the next level of operation and realize the association of multiple data tables.
[0113] Please refer to the following Table 1 (fault analysis and location table) and Table 2 (accident analysis and location table), which exemplarily give fault data and accident data in multi-source EMU data.
[0114] Table 1
[0115]
[0116] Table 2
[0117]
[0118] In some embodiments, after determining the sub-dimensional label of each of the multi-source EMU data, a labeled feature database can be established.
[0119] According to the computing requirements, the labeled feature database is crossed with the advanced repair data of each vehicle platform to complete the association operation.
[0120] In the step S103, the multi-source EMU data of the EMU and the advanced maintenance data of the EMU are associated, and the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle is determined.
[0121] Advanced maintenance of train sets refers to the maintenance work of a higher level carried out on the train sets after the mileage reaches a certain standard value.
[0122] Advanced maintenance for EMUs includes level three, level four and level five maintenance, collectively referred to as advanced maintenance for EMUs. The maintenance cycle for level three maintenance is 1.2 million kilometers or 3 years; the maintenance cycle for level four maintenance is 2.4 million kilometers or 6 years; and the maintenance cycle for level five maintenance is 4.8 million kilometers or 12 years.
[0123] The advanced maintenance data of EMUs, also known as the detailed advanced maintenance data of trains, is a detailed record of the advanced maintenance plan and actual implementation of EMUs or vehicles. These data are important bases for ensuring the safe operation of vehicles, extending their service life and optimizing maintenance strategies.
[0124] Determine the correlation between the multi-source EMU data and the advanced maintenance cycle of the EMU, so that the multi-source EMU data and the advanced maintenance cycle can be matched, and the multi-source EMU data corresponding to the advanced maintenance cycle is used to perform risk analysis from the advanced maintenance cycle dimension.
[0125] In the step S104, based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle, the risk analysis results of the EMU under the multiple sub-dimensions are determined from the year dimension and the advanced maintenance dimension respectively; the risk analysis results include fault analysis results and accident analysis results.
[0126] In the embodiment of the present application, specifically, risk analysis is first performed from two dimensions: year dimension and advanced repair dimension; under the year dimension and advanced repair dimension, there are multiple sub-dimensions, including vehicle platform dimension, target category dimension, functional system dimension, functional system dimension under vehicle platform dimension, and functional system dimension under target category dimension; among these sub-dimensions, vehicle platform dimension, target category dimension, and functional system dimension belong to sub-dimensions with larger granularity, while functional system dimension under vehicle platform dimension and functional system dimension under target category dimension belong to sub-dimensions with smaller granularity.
[0127] For each sub-dimension, the risk analysis results are divided into two types: failure analysis results and accident analysis results.
[0128] Exemplarily, the multi-dimensional risk analysis method described in the embodiment of the present application can specifically obtain: risk analysis results of year dimension-vehicle model platform dimension, risk analysis results of year dimension-target category dimension, risk analysis results of year dimension-functional system dimension, risk analysis results of year dimension-vehicle model platform dimension-functional system dimension, risk analysis results of year dimension-target category dimension-functional system dimension; and can obtain: risk analysis results of advanced repair dimension-vehicle model platform dimension, risk analysis results of advanced repair dimension-target category dimension, risk analysis results of advanced repair dimension-functional system dimension, risk analysis results of advanced repair dimension-vehicle model platform dimension-functional system dimension, risk analysis results of advanced repair dimension-target category dimension-functional system dimension; the risk analysis results include two types: fault analysis results and accident analysis results.
[0129] It can be seen that the multi-dimensional risk analysis method described in the embodiment of the present application starts from the advanced maintenance and years of EMUs, and comprehensively analyzes the accidents and failure patterns of each subsystem of EMUs of different models in each advanced maintenance process and year.
[0130] In the multi-dimensional risk analysis method for a motor vehicle group described in the embodiment of the present application, after obtaining the risk analysis result for the motor vehicle group, the method further includes:
[0131] Visualize the risk analysis results of multiple sub-dimensions under the year dimension and advanced repair dimension;
[0132] Taking the year dimension or the advanced repair dimension as the X-axis, taking the failure rate per million kilometers in the risk analysis result corresponding to the dimension as the Y-axis, respectively determining the broken line marks corresponding to different sub-dimensions, and generating a two-dimensional chart of the risk analysis result of the year dimension or the advanced repair dimension;
[0133] and / or,
[0134] The year dimension or advanced repair dimension is used as the X-axis, the failure rate per million kilometers in the risk analysis result corresponding to the dimension is used as the Y-axis, and different sub-dimensions are corresponded to the Z-axis to generate a three-dimensional chart of the risk analysis results of the year dimension or advanced repair dimension.
[0135] Based on the two-dimensional chart and the three-dimensional chart, the risk analysis result is displayed visually, so that the user can better understand the risk analysis result.
[0136] Exemplarily, the two-dimensional chart may be a two-dimensional line chart, a scatter plot, etc.; the three-dimensional chart may be a three-dimensional bar chart.
[0137] For details, please refer to Figure 4 , Figure 4 A flow chart of a method for determining the risk analysis results of the EMU under the multiple sub-dimensions from the year dimension in an embodiment of the present application is shown; Figure 4 As shown, based on the time attribute of the multi-source EMU data, the risk analysis results of the EMU under the multiple sub-dimensions are determined from the year dimension, including the following steps S401-S402:
[0138] S401. Based on the time attribute of the multi-source EMU data, determine the annual failure statistics and annual accident statistics of multiple sub-dimensions under the year dimension; the annual failure statistics of the sub-dimensions include the number of failures and annual mileage of the sub-dimensions in multiple years; the annual accident statistics of the sub-dimensions include the number of accidents and accident mileage of the sub-dimensions in multiple years;
[0139] S402, based on the annual failure statistics results and annual accident statistics results of the multiple sub-dimensions under the year dimension, determine the annual failure analysis results and the annual accident analysis results of the multiple sub-dimensions under the year dimension;
[0140] The annual failure analysis result summarizes the failure rates per million kilometers of multiple years; the failure rates per million kilometers of the year are determined based on the number of failures in the year, the annual mileage and the first conversion coefficient;
[0141] The accident analysis results for the year summarize the accident rates per million kilometers for multiple years; the accident rates per million kilometers for the year are determined based on the number of accidents in that year, the annual mileage and the second conversion factor.
[0142] The number of faults, also called fault items.
[0143] Exemplarily, for the target categories, by using the target category-technology platform dictionary data and multi-source EMU data as algorithm input information, the failure rate per million kilometers of different target categories in each year is counted.
[0144] For example, please refer to Table 3, which shows an example of annual failure statistics results of the target category dimension under the year dimension.
[0145] Table 3
[0146]
[0147] Based on the above data, the failure rate per million kilometers of the year dimension-target category dimension can be calculated.
[0148] Please refer to Table 3, which shows the number of failures and annual mileage of the target category dimension in multiple years; the failure rate per million kilometers based on the year dimension-target category dimension is determined based on the number of failures in the target category dimension in that year, the annual mileage and the first conversion factor.
[0149] The annual failure analysis results summarize the failure rates per million kilometers for multiple years; the failure rates per million kilometers for the year are determined based on the number of failures in that year, the annual mileage and the first conversion coefficient; specifically, the number of failures is used as the numerator and the annual mileage data is used as the denominator, and the two are divided. The result is multiplied by the first conversion coefficient 1,000,000 to obtain the annual failure rate per million kilometers statistics of the year dimension-target category dimension.
[0150] Based on the target category, the annual failure statistics under the year dimension are summarized to generate a line chart of the failure rate per million kilometers in the year dimension - target category dimension.
[0151] Please refer to Figure 5 , Figure 5 A line chart showing the failure rate per million kilometers in the year dimension-target category dimension described in the embodiment of the present application is shown. Figure 5 In the example, the X-axis represents the year dimension, the Y-axis represents the failure rate per million kilometers under the target category dimension, and different broken lines represent different target categories; thus, Figure 5 The figure comprehensively displays the statistical results of the failure rate per million kilometers of Category 1, Category 2, Category 3, and Category 4 under the year dimension.
[0152] Exemplarily, based on the target category-functional system, the annual failure statistics under the year dimension are summarized to generate a line graph of the failure rate per million kilometers in the year dimension-target category dimension-functional system dimension.
[0153] Please refer to Figure 6 , Figure 6 A line graph showing the failure rate per million kilometers of the functional system in the year dimension-target category one according to the embodiment of the present application is shown; Figure 6 The statistical results of the annual changes of various functional systems (traction, air supply and braking, network and auxiliary monitoring, bogie and its auxiliary, auxiliary electrical, main power supply, others, and overall) of the target category under the year dimension are displayed; among them, the X-axis represents the year dimension, the Y-axis represents the failure rate per million kilometers of various functional systems in the category dimension, and different broken lines represent different functional systems.
[0154] Please refer to Figure 5 and Figure 6 , users can select the risk analysis results of a specific dimension in the graphical user interface to view, Figure 5 In the display, if the user selects the year dimension failure details and the target category dimension, the failure rate per million kilometers of the year dimension-target category dimension will be displayed; Figure 6 If the user selects Category 1, the failure rate per million kilometers of the functional system under the year dimension - target Category 1 - will be further displayed.
[0155] Therefore, users can select operations based on the dimensions in the graphic user interface and choose to view the risk analysis results of specific dimensions. The operation is convenient, fast and intuitive.
[0156] That is to say, in the multi-dimensional risk analysis method for EMU, the method further includes:
[0157] In response to a selection operation on a year dimension or an advanced repair dimension in a graphical user interface, displaying the fault analysis results and / or accident analysis results of the EMU under the multiple sub-dimensions from the year dimension or the advanced repair dimension selected by the user;
[0158] In response to an expansion operation on the displayed fault analysis result and / or accident analysis result, the fault statistical result corresponding to the fault analysis result and / or the accident statistical result corresponding to the accident analysis result are displayed.
[0159] It should be noted that in the embodiments of the present application, the fault analysis results and / or accident analysis results may be presented in the form of visual charts, or in other forms such as tables and data.
[0160] The fault analysis results are associated with their corresponding fault statistical results, and the accident analysis results are associated with their corresponding accident analysis results; the fault statistical results include data corresponding to the fault analysis results; the accident analysis results include corresponding data in the accident analysis results, so that users can not only see the risk analysis results, but also easily see the corresponding data, which is more convenient.
[0161] The fault statistics results and / or accident statistics results can be exported to facilitate the user's subsequent use of the data and provide data verification support for extending the system maintenance period.
[0162] Exemplarily, the annual failure statistics results of the vehicle model platform dimension under the year dimension and the corresponding annual failure analysis results are described below.
[0163] Please refer to Table 4, which shows the annual failure statistics of the year dimension-model platform dimension described in the embodiment of the present application. Table 4 counts the number of failures and annual mileage of different model platforms at different levels of difficulty.
[0164] The accident analysis results of the year summarize the accident rates per million kilometers for multiple years; the accident rates per million kilometers for the year are determined based on the number of accidents in that year, the annual mileage and the second conversion coefficient. Specifically, the statistical data on the number of failures is used as the numerator and the annual mileage data is used as the denominator, and the two are divided. The result is multiplied by the second conversion coefficient 1,000,000 to obtain the annual failure rate statistics per million kilometers for different vehicle technology platforms under the year dimension.
[0165] Table 4
[0166]
[0167] By classifying and summarizing the annual failure rates per million kilometers based on vehicle model platforms, a statistical line chart of the year dimension - vehicle model platform - annual failure rates per million kilometers is generated.
[0168] Please refer to Figure 7 , Figure 7The statistical line graph of the year dimension-model platform-annual million-kilometer failure rate in the embodiment of the present application is shown. In the line graph, the X-axis represents the year dimension, and the Y-axis represents the million-kilometer failure rate, specifically the million-kilometer failure rate in the model platform dimension. Different broken lines represent different model platforms, so that the annual million-kilometer failure rate of the two dimensions of year dimension-model platform is represented in one line graph.
[0169] It should be noted that we can further summarize the annual failure statistics under the year dimension based on the vehicle platform-functional system, and generate a line graph of the failure rate per million kilometers in the year dimension-vehicle platform dimension-functional system dimension.
[0170] In addition to determining the generation of a statistical line chart of the year dimension-model platform-annual million-kilometer failure rate based on the annual failure rate per million kilometers of different model platforms shown in Table 4, the algorithm execution results can also be classified and summarized based on the model platform, year, and million-kilometer failure rate to generate a three-dimensional chart of the annual million-kilometer failure rate of the year dimension-model platform dimension-functional system dimension, so as to more intuitively display the risk analysis results in the form of a 3D rendering.
[0171] Please refer to Figure 8 , Figure 8 A three-dimensional bar chart of the year dimension-model platform-annual million-kilometer failure rate described in the embodiment of the present application is shown; Figure 8 In the figure, the X-axis is the year dimension, the Y-axis is the year dimension-the failure rate per million kilometers of the vehicle model platform, and the Z-axis is the vehicle model platform. Thus, the risk analysis result of the year dimension-vehicle model platform-annual failure rate per million kilometers is more intuitive in the form of a 3D rendering.
[0172] Please refer to Figure 8 For the same dimension, such as the year dimension-model platform dimension, based on the selection operation of the two-dimensional chart or the three-dimensional chart, the two-dimensional chart or the three-dimensional chart under the dimension can be displayed accordingly to better meet user needs.
[0173] It should be noted that a specific vehicle model platform can be further selected to obtain the failure rate per million kilometers in the year dimension - vehicle model platform dimension - functional system dimension, and generate a corresponding two-dimensional line graph or three-dimensional bar graph.
[0174] In the embodiment of the present application, based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions are determined from the advanced maintenance dimension, including:
[0175] Determine the multi-source EMU data corresponding to the advanced maintenance cycle based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle;
[0176] Based on the advanced maintenance data, determine whether the EMU has actually completed advanced maintenance within a given period after the fault occurs;
[0177] If it is not repaired, the fault data or accident data is removed from the multi-source EMU data, and the multi-source EMU data is updated. The updated multi-source EMU data includes fault samples and accident samples;
[0178] Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions are determined from the advanced maintenance dimension.
[0179] That is to say, for the advanced maintenance data, it is necessary to further process the multi-source EMU data based on whether the EMU has actually completed the advanced maintenance within a predetermined period after the fault occurs, and eliminate the fault data and accident data that have not been repaired, so as to obtain the processed fault data and accident data as fault samples and accident samples.
[0180] Please refer to Fig. 9 , Fig. 9 A flowchart of a method for determining the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions from the high-level maintenance dimension according to an embodiment of the present application is shown; the method for determining the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions from the high-level maintenance dimension based on the updated multi-source EMU data corresponding to the high-level maintenance cycle includes the following steps S901-S902:
[0181] S901. Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, respectively determine the advanced maintenance failure statistics and advanced maintenance accident statistics of multiple sub-dimensions under the advanced maintenance dimension; the advanced maintenance failure statistics of the sub-dimensions include the number of advanced maintenance failure samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension; the advanced maintenance accident statistics of the sub-dimensions include the number of advanced maintenance accident samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension;
[0182] S902, based on the advanced repair failure statistics results and advanced repair accident statistics results of multiple sub-dimensions under the advanced repair dimension, determine the advanced repair failure analysis results and advanced repair accident analysis results of multiple sub-dimensions under the advanced repair dimension;
[0183] The high-level maintenance failure analysis results are summarized with failure rates per million kilometers of multiple speed levels; the failure rates per million kilometers of the speed level are determined based on the number of high-level maintenance failure samples of the speed level, the annual mileage and the third conversion factor;
[0184] The high-level maintenance accident analysis results summarize accident rates per million kilometers at multiple speed levels; the accident rates per million kilometers at each speed level are determined based on the number of high-level maintenance accident samples at that speed level, the annual mileage, and a fourth conversion factor.
[0185] It is particularly important to note that in the embodiment of the present application, the annual mileage of the advanced maintenance dimension is calculated by the following method:
[0186] Determine the number of standard groups of the vehicle model platform according to the number of groups of specific vehicle models under the vehicle model platform and the preset standard group conversion coefficient corresponding to the group;
[0187] When the advanced maintenance cycle intervals corresponding to the target category dimensions are different, determining the estimated coefficient of annual mileage based on the advanced maintenance cycle intervals corresponding to the target category dimensions;
[0188] The product of the estimated coefficient and the number of standard groups of the vehicle platform is taken as the annual mileage of the advanced repair.
[0189] During the advanced maintenance cycle, different trains may enter maintenance early or late, and the annual mileage of each train during advanced maintenance may be different. Therefore, the embodiment of the present application estimates the annual mileage based on the number of train sets of specific models under the model platform and the advanced maintenance cycle interval, thereby reducing the computational flow and statistical difficulty in the risk analysis process of the advanced maintenance dimension.
[0190] Please refer to Table 5, which shows the advanced repair fault statistics results in the customized target category dimension described in the embodiment of the present application; the advanced repair fault statistics results include the number of advanced repair fault samples at different speed levels and the annual mileage.
[0191] The number of advanced repair failure samples is the statistical advanced repair items.
[0192] Table 5
[0193]
[0194] Please refer to Table 5. In the embodiment of the present application, the advanced maintenance cycle interval corresponding to category 1 and category 2 is 600,000 kilometers, the advanced maintenance cycle interval corresponding to category 3 and category 4 is 1.2 million kilometers, the estimated coefficient corresponding to category 1 and category 2 is 600,000, and the estimated coefficient corresponding to category 3 and category 4 is 1,200,000.
[0195] Analyze the statistical results of advanced repair faults, and obtain the number of advanced repair fault samples for each speed level by calculating the advanced repair data of the vehicle group and the driving fault matching data according to the speed level; for the mileage data of the advanced repair dimension, when the target category is one or two, the standard number of groups is multiplied by 600,000, and if the target category is three or four, the standard number of groups is multiplied by 1,200,000.
[0196] The advanced repair failure analysis results summarize the failure rates per million kilometers for multiple speed levels; the failure rates per million kilometers for each speed level are determined based on the number of advanced repair failure samples for that speed level, the annual mileage and a third conversion coefficient. Specifically, the number of advanced repair failure samples is taken as the numerator and the annual mileage data is taken as the denominator, and the two are divided. The result is multiplied by the third conversion coefficient 1,000,000 to obtain the failure rate statistics per million kilometers for advanced repair dimensions of different target categories.
[0197] By classifying and summarizing the high-level maintenance dimension failure rate per million kilometers based on the custom target category, a statistical line chart of the high-level maintenance dimension failure rate per million kilometers is generated.
[0198] Please refer to Fig.10 , Fig.10 A line chart showing the failure rate statistics for one million kilometers in the advanced maintenance dimension-target category dimension described in an embodiment of the present application is shown; Fig.10 In the figure, the X-axis is the advanced maintenance dimension, specifically different speed levels, and the Y-axis is the advanced maintenance dimension-the failure rate per million kilometers of the target category. Different broken lines correspond to different categories.
[0199] It should be noted that when a specific target category is selected, a statistical line chart of the failure rate per million kilometers of the advanced maintenance dimension-target category dimension-functional system can be generated.
[0200] Alternatively, when selecting to generate a three-dimensional chart, a three-dimensional bar chart of the failure rate per million kilometers of the advanced maintenance dimension-target category dimension is generated, and the Z axis of the three-dimensional bar chart represents different target category dimensions.
[0201] Exemplarily, the following Table 6 shows the advanced repair fault statistics results in the vehicle model platform dimension described in the embodiment of the present application.
[0202] By taking the number of advanced repair failure samples as the numerator and the annual mileage data as the denominator, dividing them and multiplying the result by the third conversion coefficient 1,000,000, we can obtain the statistics of the failure rate per million kilometers of the advanced repair dimension for different vehicle models and platforms.
[0203] By classifying and summarizing the failure rates per million kilometers based on the vehicle model platform dimension, a statistical line chart of the failure rates per million kilometers based on the advanced repair dimension - vehicle model platform dimension is generated.
[0204] Please refer to Fig.11, Fig.11 A line chart showing the failure rate statistics of one million kilometers in the advanced maintenance dimension-vehicle platform dimension described in an embodiment of the present application is shown; Fig.11 In the figure, the X-axis is the advanced maintenance dimension, specifically different speed levels, and the Y-axis is the advanced maintenance dimension - the failure rate per million kilometers of the vehicle platform. Different broken lines correspond to different vehicle platforms.
[0205] It should be noted that when a specific vehicle model platform is selected, a statistical line graph of the failure rate per million kilometers of the advanced maintenance dimension - vehicle model platform - functional system can be generated.
[0206] Alternatively, when selecting to generate a three-dimensional chart, a three-dimensional bar chart of the failure rate per million kilometers of the advanced repair dimension-vehicle platform dimension is generated, and the Z axis of the three-dimensional bar chart represents different vehicle platform dimensions.
[0207] Table 6
[0208]
[0209] It should be noted that in the embodiment of the present application, annual accident statistics are also obtained and further processed to obtain annual accident analysis results. For example, a statistical line chart of year dimension-target category dimension-accident rate per million kilometers in the annual accident analysis results is obtained, and a three-dimensional bar chart of year dimension-target category dimension-accident rate per million kilometers is obtained through processing, and so on.
[0210] In an embodiment of the present application, advanced repair accident statistics are also obtained, and further processed to obtain advanced repair accident analysis results, for example, a statistical line chart of year dimension-target category dimension-accident rate per million kilometers in the advanced repair accident analysis results is obtained, and a three-dimensional bar chart of advanced repair dimension-target category dimension-accident rate per million kilometers is obtained, and the like.
[0211] In an embodiment of the present application, various dimensions of the risk analysis results are updated in a dedicated tool in the form of structural charts to facilitate data traceability. For example, the failure rate of each vehicle platform is counted according to the advanced maintenance process, and the failure rate of each subsystem of a certain vehicle platform is counted according to the advanced maintenance process of the platform. This is convenient for users to view and quickly interpret the risk analysis results of the EMU from different dimensions, and to more comprehensively grasp and understand the failure patterns of the vehicle throughout its life cycle.
[0212] Based on the same inventive concept, the embodiment of the present application also provides a multidimensional risk analysis device for EMUs corresponding to the multidimensional risk analysis method for EMUs. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the multidimensional risk analysis method for EMUs mentioned above in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0213] Please refer to Fig.12 , Fig.12 The structure diagram of the multi-dimensional risk analysis device for EMUs according to the embodiment of the present application is shown; the multi-dimensional risk analysis device for EMUs comprises:
[0214] The acquisition module 1201 is used to acquire multi-source EMU data for the EMU from pre-constructed EMU multi-type data sources; the multi-source EMU data at least includes the EMU operation data, fault data and accident data of each scenario;
[0215] A first determination module 1202 is used to determine a sub-dimension label of each of the multi-source EMU data, and determine a plurality of sub-dimensions according to the sub-dimension label;
[0216] The second determination module 1203 is used to associate the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determine the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle;
[0217] The third determination module 1204 is used to determine the risk analysis results of the EMU in the multiple sub-dimensions from the year dimension and the advanced maintenance dimension respectively based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle; the risk analysis results include fault analysis results and accident analysis results.
[0218] In some embodiments, in the multi-dimensional risk analysis device for EMUs, the first determination module, when determining the sub-dimensional label of each of the multi-source EMU data and determining multiple sub-dimensions according to the sub-dimensional label, is specifically used to:
[0219] Determine a vehicle model platform label, a function system label and at least one target category label as sub-dimension labels; the target category label is customized according to business requirements;
[0220] According to each sub-dimension label, determine the first sub-dimension corresponding to each sub-dimension label;
[0221] According to the first sub-dimension corresponding to each sub-dimension label and the superior-subordinate relationship between different first sub-dimensions, the first sub-dimensions with the superior-subordinate relationship are combined to obtain the second sub-dimension;
[0222] The first sub-dimension and the second sub-dimension are determined as the multiple sub-dimensions; wherein the multiple sub-dimensions include a vehicle platform dimension, a target category dimension, a functional system dimension, a functional system dimension under the vehicle platform dimension, and a functional system dimension under the target category dimension.
[0223] In some embodiments, in the multi-dimensional risk analysis device for an EMU, the third determination module, when determining the risk analysis results of the EMU under the multiple sub-dimensions from the year dimension based on the time attribute of the multi-source EMU data, is specifically used to:
[0224] Based on the time attribute of the multi-source EMU data, annual failure statistics and annual accident statistics of multiple sub-dimensions under the year dimension are determined; the annual failure statistics of the sub-dimensions include the number of failures and annual mileage of the sub-dimensions in multiple years; the annual accident statistics of the sub-dimensions include the number of accidents and accident mileage of the sub-dimensions in multiple years;
[0225] Based on the annual failure statistics results and annual accident statistics results of the multiple sub-dimensions under the year dimension, determine the annual failure analysis results and the annual accident analysis results of the multiple sub-dimensions under the year dimension;
[0226] The annual failure analysis result summarizes the failure rates per million kilometers of multiple years; the failure rates per million kilometers of the year are determined based on the number of failures in the year, the annual mileage and the first conversion coefficient;
[0227] The accident analysis results for the year summarize the accident rates per million kilometers for multiple years; the accident rates per million kilometers for the year are determined based on the number of accidents in that year, the annual mileage and the second conversion factor.
[0228] In some embodiments, in the multi-dimensional risk analysis device for EMUs, the third determination module, when determining the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions from the advanced maintenance dimension based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle, is specifically used to:
[0229] Determine the multi-source EMU data corresponding to the advanced maintenance cycle based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle;
[0230] Based on the advanced maintenance data, determine whether the EMU has actually completed advanced maintenance within a given period after the fault occurs;
[0231] If it is not repaired, the fault data or accident data is removed from the multi-source EMU data, and the multi-source EMU data is updated. The updated multi-source EMU data includes fault samples and accident samples;
[0232] Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions are determined from the advanced maintenance dimension.
[0233] In some embodiments, in the multi-dimensional risk analysis device for an EMU, the third determination module, when determining the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions from the advanced maintenance dimension based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, is specifically used to:
[0234] Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the advanced maintenance failure statistics and advanced maintenance accident statistics of multiple sub-dimensions under the advanced maintenance dimension are respectively determined; the advanced maintenance failure statistics of the sub-dimensions include the number of advanced maintenance failure samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension; the advanced maintenance accident statistics of the sub-dimensions include the number of advanced maintenance accident samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension;
[0235] Based on the advanced repair failure statistics results and advanced repair accident statistics results of multiple sub-dimensions under the advanced repair dimension, determine the advanced repair failure analysis results and advanced repair accident analysis results of multiple sub-dimensions under the advanced repair dimension;
[0236] The high-level maintenance failure analysis results are summarized with failure rates per million kilometers of multiple speed levels; the failure rates per million kilometers of the speed level are determined based on the number of high-level maintenance failure samples of the speed level, the annual mileage and the third conversion factor;
[0237] The high-level maintenance accident analysis results summarize accident rates per million kilometers at multiple speed levels; the accident rates per million kilometers at each speed level are determined based on the number of high-level maintenance accident samples at that speed level, the annual mileage, and a fourth conversion factor.
[0238] In some embodiments, in the multi-dimensional risk analysis device for EMUs, the annual mileage of the advanced maintenance dimension is calculated by the following method:
[0239] Determine the number of standard groups of the vehicle model platform according to the number of groups of specific vehicle models under the vehicle model platform and the preset standard group conversion coefficient corresponding to the group;
[0240] When the advanced maintenance cycle intervals corresponding to the target category dimensions are different, determining the estimated coefficient of annual mileage based on the advanced maintenance cycle intervals corresponding to the target category dimensions;
[0241] The product of the estimated coefficient and the number of standard groups of the vehicle platform is taken as the annual mileage of the advanced repair.
[0242] In some embodiments, the multi-dimensional risk analysis device for EMUs further includes:
[0243] A visualization module, for visualizing the risk analysis results of multiple sub-dimensions under the year dimension and the advanced repair dimension after obtaining the risk analysis results for the EMU;
[0244] Taking the year dimension or the advanced repair dimension as the X-axis, taking the failure rate per million kilometers in the risk analysis result corresponding to the dimension as the Y-axis, respectively determining the broken line marks corresponding to different sub-dimensions, and generating a two-dimensional chart of the risk analysis result of the year dimension or the advanced repair dimension;
[0245] and / or,
[0246] The year dimension or advanced repair dimension is used as the X-axis, the failure rate per million kilometers in the risk analysis result corresponding to the dimension is used as the Y-axis, and different sub-dimensions are corresponded to the Z-axis to generate a three-dimensional chart of the risk analysis results of the year dimension or advanced repair dimension.
[0247] In some embodiments, the multi-dimensional risk analysis device for EMUs also includes
[0248] A selection module is used for displaying the fault analysis results and / or accident analysis results of the EMU under the multiple sub-dimensions from the year dimension or the advanced repair dimension selected by the user in response to the selection operation on the year dimension or the advanced repair dimension in the graphical user interface;
[0249] In response to an expansion operation on the displayed fault analysis result and / or accident analysis result, the fault statistical result corresponding to the fault analysis result and / or the accident statistical result corresponding to the accident analysis result are displayed.
[0250] In some embodiments, the acquisition module in the multi-dimensional risk analysis device for EMUs, when acquiring multi-source EMU data for EMUs from pre-built EMU multi-type data sources, is specifically used to:
[0251] Acquire original EMU data from a pre-built EMU multi-type data source, and aggregate the operation data, fault data, and accident data of each scenario in the original EMU data as original multi-source EMU data;
[0252] Deduplication is performed on the original multi-source EMU data of the EMU to obtain deduplication-free multi-source EMU data;
[0253] The multi-source EMU data after deduplication is reviewed and located, and the multi-source EMU data is updated according to the functional system in the standard dictionary table to obtain updated multi-source EMU data.
[0254] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.
[0255] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0256] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0257] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a platform server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.
[0258] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A multi-dimensional risk analysis method for EMUs, characterized in that: The method comprises the following steps: Acquire multi-source EMU data for the EMU from pre-built EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU; Determine a sub-dimension label for each of the multi-source EMU data, and determine a plurality of sub-dimensions according to the sub-dimension label; Associating the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determining the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle; Based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, the risk analysis results of the EMU in the multiple sub-dimensions are determined from the year dimension and the advanced maintenance dimension respectively; the risk analysis results include fault analysis results and accident analysis results.
2. The multi-dimensional risk analysis method for EMU according to claim 1 is characterized in that: Determining the sub-dimension label of each of the multi-source EMU data, and determining a plurality of sub-dimensions according to the sub-dimension label, includes: Determine a vehicle model platform label, a function system label and at least one target category label as sub-dimension labels; the target category label is customized according to business requirements; According to each sub-dimension label, determine the first sub-dimension corresponding to each sub-dimension label; According to the first sub-dimension corresponding to each sub-dimension label and the superior-subordinate relationship between different first sub-dimensions, the first sub-dimensions with the superior-subordinate relationship are combined to obtain the second sub-dimension; The first sub-dimension and the second sub-dimension are determined as the multiple sub-dimensions; wherein the multiple sub-dimensions include a vehicle platform dimension, a target category dimension, a functional system dimension, a functional system dimension under the vehicle platform dimension, and a functional system dimension under the target category dimension.
3. The multi-dimensional risk analysis method for EMU according to claim 2 is characterized in that: Based on the time attribute of the multi-source EMU data, the risk analysis results of the EMU under the multiple sub-dimensions are determined from the year dimension, including: Based on the time attribute of the multi-source EMU data, annual failure statistics and annual accident statistics of multiple sub-dimensions under the year dimension are determined; the annual failure statistics of the sub-dimensions include the number of failures and annual mileage of the sub-dimensions in multiple years; the annual accident statistics of the sub-dimensions include the number of accidents and accident mileage of the sub-dimensions in multiple years; Based on the annual failure statistics results and annual accident statistics results of the multiple sub-dimensions under the year dimension, determine the annual failure analysis results and the annual accident analysis results of the multiple sub-dimensions under the year dimension; The annual failure analysis result summarizes the failure rates per million kilometers of multiple years; the failure rates per million kilometers of the year are determined based on the number of failures in the year, the annual mileage and the first conversion coefficient; The accident analysis results for the year summarize the accident rates per million kilometers for multiple years; the accident rates per million kilometers for the year are determined based on the number of accidents in that year, the annual mileage and the second conversion factor.
4. The multi-dimensional risk analysis method for EMU according to claim 2 is characterized in that: Based on the time attribute of the multi-source EMU data and the correlation between the multi-source EMU data and the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU under the multiple sub-dimensions are determined from the advanced maintenance dimension, including: Determine the multi-source EMU data corresponding to the advanced maintenance cycle based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle; Based on the advanced maintenance data, determine whether the EMU has actually completed advanced maintenance within a given period after the fault occurs; If it is not repaired, the fault data or accident data is removed from the multi-source EMU data, and the multi-source EMU data is updated. The updated multi-source EMU data includes fault samples and accident samples; Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions are determined from the advanced maintenance dimension.
5. The multi-dimensional risk analysis method for EMUs according to claim 4 is characterized in that: Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the fault analysis results and accident analysis results of the EMU in the multiple sub-dimensions are determined from the advanced maintenance dimension, including: Based on the updated multi-source EMU data corresponding to the advanced maintenance cycle, the advanced maintenance failure statistics and advanced maintenance accident statistics of multiple sub-dimensions under the advanced maintenance dimension are respectively determined; the advanced maintenance failure statistics of the sub-dimensions include the number of advanced maintenance failure samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension; the advanced maintenance accident statistics of the sub-dimensions include the number of advanced maintenance accident samples and annual mileage at different speed levels of the advanced maintenance cycle of the sub-dimension; Based on the advanced repair failure statistics results and advanced repair accident statistics results of multiple sub-dimensions under the advanced repair dimension, determine the advanced repair failure analysis results and advanced repair accident analysis results of multiple sub-dimensions under the advanced repair dimension; The high-level maintenance failure analysis results are summarized with failure rates per million kilometers of multiple speed levels; the failure rates per million kilometers of the speed level are determined based on the number of high-level maintenance failure samples of the speed level, the annual mileage and the third conversion factor; The high-level maintenance accident analysis results summarize accident rates per million kilometers at multiple speed levels; the accident rates per million kilometers at each speed level are determined based on the number of high-level maintenance accident samples at that speed level, the annual mileage, and a fourth conversion factor.
6. The multi-dimensional risk analysis method for EMUs according to claim 5 is characterized in that: The annual mileage of the advanced repair dimension is calculated by the following method: Determine the number of standard groups of the vehicle model platform according to the number of groups of specific vehicle models under the vehicle model platform and the preset standard group conversion coefficient corresponding to the group; When the advanced maintenance cycle intervals corresponding to the target category dimensions are different, determining the estimated coefficient of annual mileage based on the advanced maintenance cycle intervals corresponding to the target category dimensions; The product of the estimated coefficient and the number of standard groups of the vehicle model platform is taken as the annual mileage of the advanced repair.
7. The multi-dimensional risk analysis method for EMUs according to claim 2 is characterized in that: After obtaining the risk analysis result for the EMU, the method further includes: Visualize the risk analysis results of multiple sub-dimensions under the year dimension and advanced repair dimension; Taking the year dimension or the advanced repair dimension as the X-axis, taking the failure rate per million kilometers in the risk analysis result corresponding to the dimension as the Y-axis, respectively determining the broken line marks corresponding to different sub-dimensions, and generating a two-dimensional chart of the risk analysis result of the year dimension or the advanced repair dimension; and / or, The year dimension or advanced repair dimension is used as the X-axis, the failure rate per million kilometers in the risk analysis result corresponding to the dimension is used as the Y-axis, and different sub-dimensions are corresponded to the Z-axis to generate a three-dimensional chart of the risk analysis results of the year dimension or advanced repair dimension.
8. The multi-dimensional risk analysis method for EMUs according to claim 1 is characterized in that: The method further comprises: In response to a selection operation on a year dimension or an advanced repair dimension in a graphical user interface, displaying the fault analysis results and / or accident analysis results of the EMU under the multiple sub-dimensions from the year dimension or the advanced repair dimension selected by the user; In response to an expansion operation on the displayed fault analysis result and / or accident analysis result, the fault statistical result corresponding to the fault analysis result and / or the accident statistical result corresponding to the accident analysis result are displayed.
9. The multi-dimensional risk analysis method for EMU according to claim 1 is characterized in that: The step of obtaining multi-source EMU data for the EMU from the pre-built EMU multi-type data sources includes: Acquire original EMU data from a pre-built EMU multi-type data source, and aggregate the operation data, fault data, and accident data of each scenario in the original EMU data as original multi-source EMU data; Deduplication is performed on the original multi-source EMU data of the EMU to obtain deduplication-free multi-source EMU data; The multi-source EMU data after deduplication is reviewed and located, and the multi-source EMU data is updated according to the functional system in the standard dictionary table to obtain updated multi-source EMU data.
10. A multi-dimensional risk analysis device for EMUs, characterized in that: The device comprises: An acquisition module, used to acquire multi-source EMU data for an EMU from pre-built EMU multi-type data sources; the multi-source EMU data at least includes operation data, fault data and accident data of each scenario of the EMU; A first determination module is used to determine a sub-dimension label of each of the multi-source EMU data, and determine a plurality of sub-dimensions according to the sub-dimension label; A second determination module is used to associate the multi-source EMU data of the EMU with the advanced maintenance data of the EMU, and determine the association relationship between the multi-source EMU data of the EMU and the advanced maintenance cycle; The third determination module is used to determine the risk analysis results of the EMU in the multiple sub-dimensions from the year dimension and the advanced maintenance dimension respectively based on the time attribute of the multi-source EMU data and the association between the multi-source EMU data and the advanced maintenance cycle; the risk analysis results include fault analysis results and accident analysis results.
