A train state fault analysis system based on data analysis

Abstract

The present application relates to the field of rail transportation, and more particularly to a train state fault analysis system based on data analysis; comprising a receiving module, a determining module, a judging module and an output module; the present application receives the state parameters of the monitored equipment in real time through the receiving module, and confirms the directly associated equipment and the secondarily associated equipment of any equipment according to the electrical connection relationship with the confirming module, distinguishes the related equipment, sequentially traverses the directly associated equipment and the secondarily associated equipment according to the abnormal parameters with the judging module, and judges the running state of the equipment, discovers the abnormal equipment in time, outputs the diagnosis result with the output module, discovers the fault in time, and combines multi-dimensional data acquisition and analysis processing, so as to realize efficient processing and analysis of the train fault data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit, and in particular to a train state fault analysis system based on data analysis. BACKGROUND

[0002] With the rapid development of urban rail transit, the scale of the subway system is increasing, and the complexity of the equipment is significantly improved. The traditional fault management mode has been difficult to meet the needs of efficient operation and maintenance. At present, the subway fault management mainly relies on manual inspection and post-processing. With the increasing complexity and intelligentization of the train system, the traditional fault management method has been difficult to meet the needs of modern train operation.

[0003] The patent document with publication number CN119537969A discloses a train fault diagnosis method and system based on deep learning. The method includes deep feature learning of train fault feature chain data reported by the target train, generating low-dimensional embedding vectors and high-dimensional detail vectors of the train fault feature chain data; retrieving a plurality of prior fault feature chain data corresponding to the low-dimensional embedding vectors associated with the low-dimensional embedding vectors of the train fault feature chain data, generating a first prior fault feature chain data sequence, and retrieving a plurality of prior fault feature chain data corresponding to the high-dimensional detail vectors associated with the high-dimensional detail vectors of the train fault feature chain data, generating a second prior fault feature chain data sequence; calculating the correlation degree between each prior fault feature chain data in the first prior fault feature chain data sequence and the second prior fault feature chain data sequence and the train fault feature chain data; based on the correlation degree, extracting a plurality of associated mode fault feature chain data from the first prior fault feature chain data sequence and the second prior fault feature chain data sequence that exist in the train fault feature chain data; based on the train fault feature chain data and the plurality of associated mode fault feature chain data, diagnosing the fault of the target train.

[0004] However, in the prior art, the monitoring of train faults usually adopts a decentralized system, and the data collected by each decentralized system is independent, which cannot fuse the data collected by multiple decentralized systems, so that the monitoring and early warning of train faults have a lag. SUMMARY

[0005] Therefore, the present application provides a train state fault analysis system based on data analysis, which combines multi-dimensional data collection and analysis processing to overcome the problem of fault monitoring lag caused by insufficient data fusion in the prior art.

[0006] To achieve the above-mentioned purpose, the present application provides a train state fault analysis system based on data analysis, comprising:

[0007] The receiving module is used to receive the monitoring parameters of each device.

[0008] a determining module, connected with the receiving module, for determining direct associated device groups and secondary associated device groups of each device according to electrical connection relationship of each device;

[0009] a judging module, connected with the receiving module and the determining module respectively, for sequentially traversing direct associated devices in the direct associated device groups and secondary associated devices in the secondary associated device groups, and judging running states of the direct associated devices and the secondary associated devices;

[0010] an outputting module, connected with the judging module, for outputting a diagnosis result according to an influence of a difference between an abnormal fluctuation characteristic value of the direct associated devices and an abnormal fluctuation characteristic value of the secondary associated devices on the arbitrary device.

[0011] Further, the determining module comprises:

[0012] an image collecting unit for constructing an actual electrical topology graph according to actual electrical connection relationship of each device in the train collected;

[0013] a first determining unit, connected with the image collecting unit, for generating a plurality of direct associated actual devices according to nodes directly electrically connected with the arbitrary device determined based on the actual electrical topology graph;

[0014] a comparing unit, connected with the first determining unit, for comparing the direct associated actual devices with a pre-stored original electrical topology graph, so as to determine whether to update the original electrical topology graph according to the direct associated actual devices;

[0015] a monitoring unit, connected with the comparing unit, for monitoring maintenance information in real time, and changing a target electrical topology graph according to the maintenance information, wherein the target electrical topology graph is the same as the original electrical topology graph.

[0016] Further, the determining module further comprises:

[0017] a second determining unit, connected with the image collecting unit and the first determining unit respectively, for determining nodes directly electrically connected with the direct associated actual devices according to the actual electrical connection relationship;

[0018] a removing unit, connected with the second determining unit, for removing the arbitrary device from the nodes directly electrically connected with the direct associated actual devices, so as to form a plurality of secondary associated actual devices.

[0019] Further, the judging module comprises:

[0020] an acquisition unit configured to acquire historical maintenance frequencies of the directly associated actual devices and the secondarily associated actual devices within a preset period of time;

[0021] a traversal unit connected with the acquisition unit and configured to sequentially traverse the directly associated actual devices and the secondarily associated actual devices in descending order of the historical maintenance frequencies, and to traverse the secondarily associated actual devices after the directly associated actual devices are traversed;

[0022] a comparison unit connected with the traversal unit and configured to acquire actual parameters of the directly associated actual devices and the secondarily associated actual devices, to compare the actual parameters of the directly associated actual devices and the secondarily associated actual devices with a preset parameter fluctuation range, and to determine whether to generate an abnormal fluctuation characteristic value according to a comparison result.

[0023] Further, the traversal unit comprises:

[0024] a primary traversal subunit configured to traverse devices in a directly associated device group;

[0025] a secondary traversal subunit connected with the primary traversal subunit and configured to traverse devices in a secondarily associated device group;

[0026] an abnormality detection subunit connected with the primary traversal subunit and the secondary traversal subunit, and configured to detect the directly associated device group preferentially to obtain an abnormal fluctuation characteristic value.

[0027] Further, the comparison unit comprises:

[0028] a comparison subunit configured to compare the actual parameters with the preset parameter fluctuation range;

[0029] a first calculation subunit configured to calculate a difference between the actual parameter and a minimum value of the preset parameter fluctuation range and the minimum value, and to obtain an abnormal fluctuation characteristic value when the actual parameter is less than the minimum value;

[0030] a second calculation subunit configured to calculate a difference between the actual parameter and a maximum value of the preset parameter fluctuation range and the maximum value, and to obtain an abnormal fluctuation characteristic value when the actual parameter is greater than the maximum value.

[0031] Further, the output module comprises:

[0032] a calculation unit connected with the comparison unit and configured to calculate an absolute value of a difference between the abnormal fluctuation characteristic value of the directly associated device and the abnormal fluctuation characteristic value of the secondarily associated device, to obtain a real-time fluctuation difference value;

[0033] a difference output unit configured to output the real-time fluctuation difference value.

[0034] Further, the output module further comprises:

[0035] The analysis unit determines a warning level based on the real-time fluctuation difference, the real-time temperature rise and the actual fault influence range;

[0036] The output unit is connected with the difference output unit and the analysis unit, and outputs all fault information to generate a diagnosis result after the analysis unit determines the warning level.

[0037] Further, the analysis unit comprises:

[0038] The first comparison sub-unit determines the real-time fluctuation difference according to the first standard fluctuation difference and the second standard fluctuation difference to determine the warning level;

[0039] The second comparison sub-unit compares the standard temperature rise with the real-time temperature rise to determine the warning level when the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference.

[0040] The third comparison sub-unit determines the warning level based on the actual fault influence range when the real-time temperature rise is greater than the standard temperature rise.

[0041] Further, the analysis unit further comprises:

[0042] The first-level warning sub-unit determines that the fault belongs to a first-level fault when the real-time fluctuation difference is less than the second standard fluctuation difference, or the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference and the real-time temperature rise is less than or equal to the standard temperature rise.

[0043] The second-level warning sub-unit determines that the fault belongs to a second-level fault when the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference and the real-time temperature rise is greater than the standard temperature rise, and the actual fault influence range is less than the standard influence range.

[0044] The third-level warning sub-unit determines that the fault belongs to a third-level fault when the real-time fluctuation difference is greater than the first standard fluctuation difference, or the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference, the real-time temperature rise is greater than the standard temperature rise, and the actual fault influence range is greater than or equal to the standard influence range.

[0045] Compared with the prior art, the beneficial effects of the present application are that through the cooperative work of each module, the system can quickly and effectively identify and predict the equipment fault of the train, accurately locate the fault equipment and its associated equipment, reduce the fault troubleshooting time, improve the accuracy of fault diagnosis through dynamic adjustment according to real-time parameters, timely output the diagnosis result to help maintenance personnel take action quickly, and improve the operation safety and efficiency of the train.

[0046] Further, by comparing the actual collection with the pre-stored topology map, the topology deviation caused by equipment aging, human modification or temporary maintenance is eliminated, the accuracy of fault analysis is improved; the monitoring unit automatically synchronizes the maintenance record to ensure that the system always diagnoses based on the latest equipment connection relationship, avoiding interference from historical data; the comparison unit eliminates invalid or temporary changes through difference analysis, only retaining stable correlation to reduce the risk of misjudgment; the precise generation of direct correlation with actual equipment narrows down the scope of fault troubleshooting and reduces the detection time of irrelevant equipment.

[0047] Further, through the precise generation of secondary associated equipment, the system can analyze the impact of faults on indirectly associated equipment and predict fault diffusion paths; the elimination unit avoids including the original equipment itself in the secondary associated equipment group, preventing logical loops and misjudgments; it supports hierarchical resolution of complex electrical networks and adapts to large train systems; through hierarchical separation, it reduces the scope of invalid detection, such as high-frequency detection of directly associated equipment and low-frequency detection of secondary associated equipment. This makes the system have strong dynamic adaptability, eliminating the need for repeated work after the train undergoes line maintenance or equipment replacement, improving the accuracy of fault analysis and reducing additional costs.

[0048] Further, through the cooperation of each unit, the priority is sorted based on historical maintenance frequency, high-risk equipment is prioritized for troubleshooting, and the fault positioning time is shortened. By calculating the abnormal fluctuation representation value, the device state is expanded from binary judgment to continuous numerical evaluation, supporting fault level division. The update mechanism of historical maintenance frequency enables the system to learn about equipment aging patterns, automatically adjust detection strategies, reduce frequent detection of low-fault-rate equipment, and reduce system computing load. The system can comprehensively and accurately monitor the running state of equipment, timely detect and record abnormalities, and improve the efficiency and accuracy of fault analysis.

[0049] Further, through independent primary and secondary traversal subunits, resource competition issues during mixed detection are avoided, the total detection time is shortened, directly associated equipment is prioritized for detection, and interference caused by false positives of secondary equipment is reduced. In high-load scenarios, the system can suspend the tasks of the secondary traversal subunit, concentrate resources on processing directly associated equipment, support the addition of a tertiary traversal subunit without the need to reconstruct existing logic. This provides detailed data support for subsequent fault analysis and enables dynamic analysis of the time and change amplitude of abnormal parameters, enhancing the system's real-time monitoring capabilities.

[0050] Further, by clearly distinguishing between the two abnormal modes of parameter being too low and being too high, a differentiated fault handling strategy is supported, the hazard degree of different abnormal types is more truly reflected through independent calculation formulas, independent calculation rules are supported for different parameter types, complex working conditions are adapted, redundant operation is avoided through condition-triggered calculation subunits, and the flexibility of the system under complex scenarios is ensured.

[0051] Further, through difference quantization, the primary and secondary fault sources are determined by real-time fluctuation differences, global faults are avoided, difference data provides a direct basis for early warning level division, high difference scenarios preferentially process directly related equipment, low difference scenarios start secondary equipment for deep detection, difference output supports historical record storage, fault mode analysis and system iteration are facilitated, and diagnostic accuracy is improved.

[0052] Further, by comprehensively evaluating the fault risk through electrical abnormalities, thermodynamic states and influence ranges, fault levels are dynamically divided according to preset rules, and differentiated response strategies are matched, high-level warnings directly trigger train control instructions to reduce manual response delay, structured output supports seamless connection with operation and maintenance management systems to improve maintenance efficiency and fault positioning accuracy, which is conducive to dealing with more complex scenarios, enhancing system robustness, and ensuring that the determination logic remains scientific under different operating environments.

[0053] Further, unnecessary parameter calculation is reduced through condition progression, efficiency is greatly improved, the correlation of fluctuation difference, temperature rise and influence range is combined to avoid misjudgment caused by a single parameter, and the case of high fluctuation difference but normal temperature rise may be only a temporary abnormality; emergency response is accurate and low-risk scenarios only record without triggering control instructions to reduce operation and maintenance interference.

[0054] Further, through independent subunit parallel processing, the determination time is significantly shortened, rapid response is possible, multi-condition combination determination is supported, low-risk faults do not occupy high-level response resources, threshold values are dynamically adjusted according to operating environments to improve system robustness, and new early warning subunits or replacement determination parameters are supported to adapt to different maintenance scenario needs. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 The train state fault analysis system structure schematic diagram based on data analysis provided by the embodiment of the present application;

[0056] Figure 2 The structure schematic diagram of the determination module in the embodiment of the present application;

[0057] Figure 3 The structure schematic diagram of the determination module in the embodiment of the present application;

[0058] Figure 4A structure schematic diagram of a comparison unit in an embodiment of the present application is shown in the figure; DETAILED DESCRIPTION

[0059] In order to make the objects and advantages of the present application more clear, the present application will be further described in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0060] The preferred embodiments of the present application will be described in conjunction with the accompanying drawings. It should be understood that the embodiments are only used to explain the technical principles of the present application and should not be used to limit the protection scope of the present application.

[0061] It should be noted that, in the description of the present application, the terms indicating the direction or position relationship such as "upper", "lower", "left", "right", "inner", "outer" and the like are based on the direction or position relationship shown in the drawings, which is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0062] In addition, it should also be noted that, in the description of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connection" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be direct connection, or indirect connection through an intermediate medium, or the internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0063] Please refer to Figure 1 The figure is a schematic diagram of a train state fault analysis system based on data analysis in the embodiment;

[0064] The embodiment provides a train state fault analysis system based on data analysis, which comprises:

[0065] A receiving module is configured to receive monitoring parameters of each device;

[0066] A determining module is connected with the receiving module and is configured to determine direct associated device groups and secondary associated device groups of each device in sequence according to electrical connection relationships of the devices;

[0067] A judging module is connected with the receiving module and the determining module respectively and is configured to sequentially traverse direct associated devices in the direct associated device groups and secondary associated devices in the secondary associated device groups, and judge running states of the direct associated devices and the secondary associated devices;

[0068] An output module connected with the judging module, configured to output a diagnosis result according to the difference between the abnormal fluctuation characteristic value of the directly associated device and the abnormal fluctuation characteristic value of the secondarily associated device.

[0069] The receiving module is connected with the monitoring device on the train through a data interface to collect data and obtain accurate device operation parameters, which can include voltage, current, temperature, vibration and other key operation data. The receiving module receives the monitoring parameters of the specific device on the train in real time, ensures the timeliness and accuracy of the data, and transmits the data to the determining module. The collected data is processed by an analog-digital conversion circuit of the data collection module and stored in a buffer memory. Then, the real-time reception and transmission of data can be realized through a wireless sensor network or an Internet of Things technology.

[0070] The determining module is directly connected with the receiving module to facilitate fast data exchange. The determining module determines the directly associated device and the secondarily associated device according to the received device monitoring parameters and the preset electrical topology diagram. The preset electrical topology diagram is determined according to the electrical circuit structure of the train during manufacturing. If the device position structure is updated subsequently, the determining module generates a new electrical topology diagram to replace the old one to ensure the normal operation of the module. This process can use a database to store and query the electrical topology diagram, which improves the efficiency. Based on the electrical topology diagram and real-time data analysis, the system can quickly determine the fault source position and reduce the misjudgment rate.

[0071] The judging module is connected with the receiving module and the determining module, respectively, and is used to sequentially traverse the directly associated device and the secondarily associated device group to determine whether the running state of the devices is abnormal. The judgment basis includes whether the operation parameters of the devices are within the preset normal range.

[0072] The output module is connected with the judging module. The output module outputs a diagnosis result according to the abnormal fluctuation characteristic value of the directly associated device and the abnormal difference value of the secondarily associated device. The diagnosis result can include the health status of the device, the fault level and the recommended maintenance measures.

[0073] Through the cooperative work of the modules, the system can quickly and effectively identify and predict the device fault of the train, accurately locate the fault device and its associated devices, reduce the fault troubleshooting time, improve the accuracy of fault diagnosis by dynamically adjusting according to the real-time parameters, output the diagnosis result in time to help the maintenance personnel take action quickly, and improve the operation safety and efficiency of the train.

[0074] Please continue to refer to Figure 2 As shown in FIG. 6, which is a structure schematic diagram of the determining module of the embodiment;

[0075] Specifically, the determining module comprises:

[0076] an image acquisition unit configured to construct an actual electrical topology map according to an actual electrical connection relationship of each device in the train acquired;

[0077] a first determination unit connected with the image acquisition unit and configured to generate a plurality of directly associated actual devices according to nodes directly electrically connected with the arbitrary device determined based on the actual electrical topology map;

[0078] a comparison unit connected with the first determination unit and configured to compare the directly associated actual devices with a pre-stored original electrical topology map to determine whether to update the original electrical topology map according to the directly associated actual devices;

[0079] a monitoring unit connected with the comparison unit and configured to monitor maintenance information in real time and change a target electrical topology map according to the maintenance information, the target electrical topology map being the same as the original electrical topology map.

[0080] Specifically, the image acquisition unit acquires the actual electrical connection relationship of each device in the train and constructs an actual electrical topology map. The physical connection information of the device is obtained through a camera or a scanning device and is converted into a digital topology structure. The electrical topology map is a structure diagram of the electrical system of the train, which shows the connection relationship between each device. By traversing the topology map, the devices directly connected with the target device are found and classified as directly associated devices. Taking the voltage abnormality of a certain device as an example, the node unit will first find the relays, switches and other devices directly connected with the device and classify them as a directly associated device group.

[0081] The first determination unit is connected with the image acquisition unit and is configured to determine nodes directly electrically connected with the arbitrary device according to the actual electrical topology map, generate a plurality of directly associated actual devices, process the data of the image acquisition unit through a software algorithm, generate the actual electrical topology map and identify the directly associated devices.

[0082] The comparison unit compares the directly associated actual devices with the pre-stored original electrical topology map to determine whether to update the original electrical topology map according to the directly associated actual devices. The differences between the actually acquired associated devices and the pre-stored topology map are compared to determine whether the original topology map stored in the system needs to be updated.

[0083] The monitoring unit monitors the maintenance information in real time and changes the target electrical topology map according to the maintenance information, so as to ensure that the target electrical topology map is the same as the original electrical topology Figure 1 map. The maintenance operation is continuously tracked after the device is replaced and the line is adjusted, and the topology map is dynamically corrected to maintain consistency, so as to avoid system misjudgment caused by manual modification.

[0084] By comparing the actual collection with the pre-stored topology graph, the topology deviation caused by equipment aging, human modification or temporary maintenance is eliminated, the accuracy of fault analysis is improved, the monitoring unit automatically synchronizes the maintenance record, the system always diagnoses based on the latest equipment connection relationship, the historical data interference is avoided, the comparison unit eliminates invalid or temporary modification through difference analysis, only the stable correlation is retained, the misjudgment risk is reduced, the accurate generation of direct correlation with actual equipment is realized, the fault troubleshooting range is reduced, and the detection time of irrelevant equipment is reduced.

[0085] Specifically, the determination module further includes:

[0086] A second determination unit connected with the image acquisition unit and the first determination unit, configured to determine nodes directly electrically connected with the directly correlated actual equipment according to the actual electrical connection relationship;

[0087] A removal unit connected with the second determination unit, configured to remove the arbitrary device from the nodes directly electrically connected with the directly correlated actual equipment, to form a plurality of secondarily correlated actual devices.

[0088] The second determination unit is connected with the image acquisition unit and the first determination unit, configured to determine nodes directly electrically connected with the directly correlated actual equipment according to the actual electrical connection relationship; the removal unit is connected with the second determination unit, configured to remove the arbitrary device from the nodes, to form a plurality of secondarily correlated actual devices; the second determination unit expands the correlation level, and on the basis of the directly correlated device group, the connection nodes of the next level are acquired through a topology graph traversal algorithm. The removal unit excludes redundant nodes, ensures that the secondarily correlated devices only include indirectly correlated and independent devices, filters out the nodes related to the original device, and avoids the repeated detection problem caused by loop connection.

[0089] Through the accurate generation of the secondarily correlated devices, the system can analyze the influence of the fault on the indirectly correlated devices, and predict the fault diffusion path; the removal unit avoids including the original device itself in the secondarily correlated device group, prevents logical cycle and misjudgment, supports hierarchical analysis of complex electrical networks, and adapts to large train systems; through hierarchical separation, the invalid detection range is reduced, for example, only the directly correlated devices are detected at high frequency, and the secondarily correlated devices are detected at low frequency. The system has strong dynamic adaptability, does not need to perform repeated work after the train is maintained or the device is replaced, the accuracy of fault analysis is improved, and the additional cost is reduced.

[0090] As shown in Figure 3 the structure schematic diagram of the judgment module in this embodiment;

[0091] Specifically, the judgment module includes:

[0092] The acquisition unit is configured to acquire historical maintenance frequencies of the directly associated actual devices and the secondarily associated actual devices within a preset period of time;

[0093] The traversal unit is connected with the acquisition unit and is configured to sequentially traverse the directly associated actual devices and the secondarily associated actual devices in descending order of the historical maintenance frequencies, and after the directly associated actual devices are traversed, the secondarily associated actual devices are traversed;

[0094] The comparison unit is connected with the traversal unit and is configured to acquire actual parameters of the directly associated actual devices and the secondarily associated actual devices, compare the actual parameters of the directly associated actual devices and the secondarily associated actual devices with a preset parameter fluctuation range, and determine whether to generate an abnormal fluctuation characteristic value according to a comparison result.

[0095] The acquisition unit is configured to acquire historical maintenance frequencies of the directly associated actual devices and the secondarily associated actual devices within a preset period of time; the traversal unit is connected with the acquisition unit and is configured to sequentially traverse the devices in descending order of the historical maintenance frequencies, and the directly associated actual devices are traversed preferentially, and then the secondarily associated actual devices are traversed;

[0096] The comparison unit is connected with the traversal unit and is configured to compare actual parameters of the directly associated actual devices and the secondarily associated actual devices with a preset parameter fluctuation range, determine that an abnormality occurs when the actual parameters exceed the preset parameter fluctuation range, and generate an abnormal fluctuation characteristic value, optimize a detection order based on historical maintenance data, and quantify a degree of abnormality of the devices. The acquisition unit is configured to extract device maintenance records through data mining and quantify a fault tendency of the devices. The traversal unit is configured to dynamically adjust a detection priority, preferentially detect a device with a high fault rate, and improve a fault discovery efficiency. The comparison unit is configured to determine an abnormality through parameter fluctuation range and generate a quantifiable abnormal fluctuation characteristic value, and provide a numerical basis for subsequent diagnosis.

[0097] Through cooperation of the units, a priority order based on historical maintenance frequencies is determined, high-risk devices are preferentially investigated, fault positioning time is shortened, a device state is expanded from a binary determination of “normal / abnormal” to continuous numerical evaluation through calculation of an abnormal fluctuation characteristic value, fault grading is supported, and an updating mechanism of the historical maintenance frequencies enables the system to learn a device aging rule, automatically adjust a detection strategy, reduce frequent detection of a device with a low fault rate, and reduce a system calculation load.

[0098] Specifically, the traversal unit comprises:

[0099] The first traversal subunit is configured to traverse devices in a directly associated device group;

[0100] The second traversal subunit is connected with the first traversal subunit and is configured to traverse devices in a secondarily associated device group;

[0101] Anomaly detection subunit connected with the first-level traversal subunit and the second-level traversal subunit, preferentially detecting the directly associated device group to obtain an abnormal fluctuation characteristic value.

[0102] The first-level traversal subunit is used for traversing the devices in the directly associated device group; the second-level traversal subunit is connected with the first-level traversal subunit and is used for traversing the devices in the second-level associated device group; and the anomaly detection subunit is connected with the first-level traversal subunit and the second-level traversal subunit, preferentially detects the directly associated device group, processes the directly associated devices and the second-level associated devices through the first-level and second-level traversal subunits respectively, realizes layered management of the detection range, forces the anomaly detection subunit to preferentially detect the directly associated device group, ensures the fault source troubleshooting efficiency, and directly transfers the abnormal data to a subsequent module to form a closed-loop analysis process.

[0103] Through the independent first-level and second-level traversal subunits, the resource competition problem in mixed detection is avoided, the total detection time is shortened, the directly associated devices are preferentially detected, and the interference caused by false positives of the second-level devices is reduced. In a high-load scenario, the system can suspend the task of the second-level traversal subunit, concentrates resources on processing the directly associated devices, supports adding a third-level traversal subunit, and does not need to reconstruct the existing logic. Detailed data support is provided for subsequent fault analysis, the time and change amplitude of abnormal parameters can be dynamically analyzed, and the real-time monitoring capability of the system is enhanced.

[0104] Please continue to refer to Figure 4 As shown in the figure, which is a comparison unit schematic diagram of the embodiment;

[0105] Specifically, the comparison unit includes:

[0106] The comparison subunit is used to compare the actual parameter with the preset parameter fluctuation range.

[0107] The first calculation subunit calculates the difference between the actual parameter and the minimum value of the preset parameter fluctuation range and the minimum value, and obtains an abnormal fluctuation characteristic value when the actual parameter is less than the minimum value of the preset parameter fluctuation range.

[0108] The second calculation subunit calculates the difference between the actual parameter and the maximum value of the preset parameter fluctuation range and the maximum value, and obtains an abnormal fluctuation characteristic value when the actual parameter is greater than the maximum value of the preset parameter fluctuation range.

[0109] The actual parameter p of the direct correlation actual device and the secondary correlation actual device is recorded, the minimum preset parameter a and the maximum preset parameter b in the preset parameter fluctuation range a-b, and the abnormal fluctuation characteristic value K. Taking the air conditioner current as an example, if the generated current parameter is 0.7A, the preset current fluctuation range is 0.3 to 0.5A, then the current generated by the sound level meter (0.7-0.5) / 0.5, the abnormal fluctuation characteristic value of the air conditioner current is 40%. If the current is 0.15A, then (0.3-0.15) / 0.3 is obtained. The abnormal fluctuation characteristic value is 50%.

[0110] By clearly distinguishing the two abnormal modes of parameter being too low and being too high, the differentiated fault processing strategy is supported,

[0111] By the independent calculation formula, the harm degree of different abnormal types is more truly reflected, such as temperature exceeding limit 10% is more urgent than voltage exceeding limit 10%, the independent calculation rule is set for different parameter types, the complex working conditions are adapted, the condition triggering calculation subunit is triggered, the redundant operation is avoided, and the flexibility of the system in complex scenes is ensured.

[0112] Specifically, the output module comprises:

[0113] The calculation unit is connected with the comparison unit, and is used to calculate the absolute value of the difference between the abnormal fluctuation characteristic value of the direct correlation device and the abnormal fluctuation characteristic value of the secondary correlation device, to obtain a real-time fluctuation difference value;

[0114] The difference output unit is used to output the real-time fluctuation difference value.

[0115] The calculation unit is connected with the comparison unit, and is used to calculate the absolute value of the difference between the abnormal fluctuation characteristic value of the direct correlation device and the abnormal fluctuation characteristic value of the secondary correlation device, to generate a real-time fluctuation difference value; the difference output unit is used to output the real-time fluctuation difference value and transmit it to a subsequent analysis module or a user interface. This is a refinement of the output module function, and the core is to quantify the fault influence range and provide operable difference data. The absolute value calculation eliminates directional bias, and focuses on the difference in abnormal degree between the direct correlation device and the secondary correlation device. The difference output unit converts the calculation result into a readable or transmissible data format, and supports subsequent decision-making.

[0116] By difference quantization, the real-time fluctuation difference value clearly distinguishes the primary and secondary fault sources, avoids misjudgment of global faults, and provides a direct basis for pre-warning level division. In a high-difference value scene, the direct correlation device is preferentially processed, and in a low-difference value scene, the secondary device is started for deep detection. The difference output supports historical record storage, which is convenient for fault mode analysis and system iteration, and improves the diagnosis accuracy.

[0117] Specifically, the output module further comprises:

[0118] an analysis unit configured to determine a warning level based on the real-time fluctuation difference, the real-time temperature rise, and the fault influence range;

[0119] an output unit connected to the difference output unit and the analysis unit, configured to output all fault information and generate a diagnosis result after the analysis unit determines the warning level.

[0120] determine a warning level based on the real-time fluctuation difference, the real-time temperature rise, and the fault influence range; an output unit connected to the difference output unit and the analysis unit, configured to output all fault information and generate a diagnosis result after the analysis unit determines the warning level. First, the analysis unit receives input parameters: fluctuation difference = 35%, temperature rise = 55°C, and influence range = 4 devices. Then, according to the preset rules, when the fluctuation difference ≥ 30% and the temperature rise ≥ 50°C, it meets the "serious" level; when the influence range > 5 devices, it does not meet; finally, the final warning level is determined to be "serious", triggering the corresponding control instruction. For example, if the train high-voltage circuit breaker contact temperature is abnormal, the real-time fluctuation difference is 40%, which is determined to be directly related to the equipment insulation resistance abnormality, the real-time temperature rise is 70°C (preset safety threshold = 60°C), and the fault influence range is 2 directly related devices. The analysis unit determines that the fluctuation difference ≥ 30% and the temperature rise ≥ 60°C meet the "serious" level, and the output unit executes, displaying a red alarm: "serious fault" and sending an instruction to the circuit breaker control unit through the TCMS (train control management system) to perform a power-off operation, generate a diagnosis report and upload it to the operation and maintenance platform, including the recommended measures: "replace the circuit breaker contact and check the insulation resistance".

[0121] By comprehensively evaluating the fault risk through the combination of electrical abnormalities, thermodynamic state, and system topology, dynamically dividing the fault level according to the preset rules, matching the differentiated response strategies, and triggering the train control instruction directly for high-level warning, the manual response delay is reduced; the structured output supports seamless connection with the operation and maintenance management system, improves maintenance efficiency, improves fault positioning accuracy, is conducive to dealing with more complex scenarios, enhances system robustness, and the introduction of preset risk threshold ensures that the determination logic remains scientific in different operating environments.

[0122] Specifically, the analysis unit comprises:

[0123] a first comparison sub-unit configured to determine the real-time fluctuation difference based on the first standard fluctuation difference and the second standard fluctuation difference to determine the warning level;

[0124] a second comparison sub-unit configured to, when the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference, obtain the corresponding real-time temperature rise, compare the standard temperature rise with the real-time temperature rise, and determine the warning level;

[0125] The third comparison subunit determines the early warning level based on the actual fault influence range when the real-time temperature rise is greater than the standard temperature rise.

[0126] The first comparison subunit determines the early warning level according to the real-time fluctuation difference, the first standard fluctuation difference and the second standard fluctuation difference. When the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference, the corresponding real-time temperature rise is obtained, and the standard temperature rise and the real-time temperature rise are compared to determine the early warning level. When the real-time temperature rise is greater than the standard temperature rise, the early warning level is determined based on the actual fault influence range. The early warning level includes: a slight fault, recording and planning maintenance; a moderate fault, starting a redundant system or a degraded operation mode; and a serious fault, triggering an emergency brake and contacting a dispatch center to start rescue. The data understanding threshold is reduced through a visualization tool, helping maintenance personnel to quickly develop a maintenance plan, and long-term accumulated fault data can be used for predictive maintenance.

[0127] The efficiency is greatly improved by condition progression to reduce unnecessary parameter calculation. The correlation of fluctuation difference, temperature rise and influence range is combined to avoid misjudgment caused by a single parameter. The emergency response is accurate and low-risk scenarios only record without triggering control instructions, reducing operation and maintenance interference.

[0128] Specifically, the analysis unit further includes:

[0129] The first-level early warning subunit determines that the fault belongs to a first-level fault when the real-time fluctuation difference is less than the second standard fluctuation difference, or the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference and the real-time temperature rise is less than or equal to the standard temperature rise.

[0130] The second-level early warning subunit determines that the fault belongs to a second-level fault when the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference and the real-time temperature rise is greater than the standard temperature rise, and the actual fault influence range is less than the standard influence range.

[0131] The third-level early warning subunit determines that the fault belongs to a third-level fault when the real-time fluctuation difference is greater than the first standard fluctuation difference, or the real-time fluctuation difference is between the first standard fluctuation difference and the second standard fluctuation difference, the real-time temperature rise is greater than the standard temperature rise, and the actual fault influence range is greater than or equal to the standard influence range.

[0132] For low-risk scenarios (small fluctuation difference or normal temperature rise), only record fault information, do not trigger control instructions; for medium-risk scenarios (fluctuation difference is moderate and temperature rise is over standard, but the impact is limited), start the redundant system or run in degraded mode; for high-risk scenarios (large fluctuation difference or wide impact range), trigger emergency braking and contact the dispatch center. The first standard deviation is 30%, and the second standard deviation is 60%. The multiple levels of sub-units cooperate with each other, accurately determine the fault level by defining the size relationship between the real-time fluctuation difference and the first standard deviation and the second standard deviation.

[0133] Through independent sub-unit parallel processing, the determination time is significantly shortened, rapid reaction can be performed, multi-condition combination determination is performed, low-risk faults are avoided from occupying high-level response resources, threshold values are dynamically adjusted according to the running environment, the system robustness is improved, and new early warning sub-units or replacement determination parameters are supported to adapt to different maintenance scene needs.

[0134] So far, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

[0135] The above description is only the preferred embodiments of the present application and is not intended to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A train condition fault analysis system based on data analysis, characterized in that, include: The receiving module is used to receive monitoring parameters from various devices; A determining module, which is connected to the receiving module, is used to sequentially determine the direct associated device group and the secondary associated device group of each device according to the electrical connection relationship of each device; The judgment module is connected to the receiving module and the determining module respectively, and is used to sequentially traverse the directly associated devices in several directly associated device groups and the secondary associated devices in the secondary associated device groups, and to determine the operating status of the directly associated devices and the secondary associated devices. An output module, connected to the judgment module, is used to output a diagnostic result based on the impact of the difference between the abnormal fluctuation characterization values ​​of several directly related devices and the abnormal fluctuation characterization values ​​of the secondary related devices on any device.

2. The train status fault analysis system based on data analysis according to claim 1, characterized in that, The determining module includes: The image acquisition unit is used to construct an actual electrical topology diagram based on the actual electrical connection relationships of each device in the train. The first determining unit is connected to the image acquisition unit and is used to generate a number of directly associated actual devices based on the nodes that are directly electrically connected to any device, determined based on the actual electrical topology diagram. A comparison unit, connected to the first determining unit, is used to compare the directly associated actual device with a pre-stored original electrical topology diagram to determine whether to update the original electrical topology diagram based on the directly associated actual device. A monitoring unit, connected to the comparison unit, is used to monitor maintenance information in real time and change it to a target electrical topology diagram based on the maintenance information. The target electrical topology diagram is the same as the original electrical topology diagram.

3. The train status fault analysis system based on data analysis according to claim 2, characterized in that, The determining module further includes: The second determining unit is connected to the image acquisition unit and the first determining unit respectively, and is used to determine the node that is directly electrically connected to the directly associated actual device according to the actual electrical connection relationship. The elimination unit, which is connected to the second determining unit, is used to eliminate any device from the nodes that are directly electrically connected to the directly associated actual device, so as to form a number of secondary associated actual devices.

4. The train status fault analysis system based on data analysis according to claim 3, characterized in that, The judgment module includes: The acquisition unit is used to acquire the historical maintenance frequency of the directly associated actual equipment and the secondary associated actual equipment within a preset time period; The traversal unit, which is connected to the acquisition unit, is used to traverse sequentially according to the historical maintenance frequency from large to small. After all the directly associated actual devices have been traversed, the secondary associated actual devices are traversed. The comparison unit, which is connected to the traversal unit, is used to obtain the actual parameters of the directly associated actual devices and the secondary associated actual devices, compare the actual parameters of the directly associated actual devices and the secondary associated actual devices with the preset parameter fluctuation range, and determine whether to generate an abnormal fluctuation characterization value based on the comparison result.

5. The train condition fault analysis system based on data analysis according to claim 4, characterized in that, The traversal unit includes: The first-level traversal sub-unit is used to traverse the devices within the directly associated device group; A second-level traversal subunit, which is connected to the first-level traversal subunit, is used to traverse the devices within the second-level associated device group; An anomaly detection subunit, which is connected to the first-level traversal subunit and the second-level traversal subunit, prioritizes the detection of the directly associated device group to obtain anomaly fluctuation characterization values.

6. The train condition fault analysis system based on data analysis according to claim 4, characterized in that, The comparison unit includes: The comparison sub-unit is used to compare the actual parameters with the preset parameter fluctuation range; The first calculation subunit calculates the difference between the actual parameter and the minimum value of the preset parameter fluctuation range when the actual parameter is less than the minimum value, and compares the difference with the minimum value to obtain the abnormal fluctuation characterization value. The second calculation subunit calculates the difference between the actual parameter and the maximum value of the preset parameter fluctuation range when the actual parameter is greater than the maximum value of the preset parameter fluctuation range. It then compares this difference with the maximum value to obtain the abnormal fluctuation characterization value.

7. The train condition fault analysis system based on data analysis according to claim 4, characterized in that, The output module includes: A calculation unit, connected to the comparison unit, is used to calculate the absolute value of the difference between the abnormal fluctuation characterization value of the directly associated device and the abnormal fluctuation characterization value of the secondary associated device, so as to obtain the real-time fluctuation difference. The difference output unit is used to output the real-time fluctuation difference.

8. The train condition fault analysis system based on data analysis according to claim 7, characterized in that, The output module also includes: The analysis unit determines the warning level based on the real-time fluctuation difference, real-time temperature rise, and fault impact range; The output unit, which is connected to the difference output unit and the analysis unit, outputs all fault information and generates diagnostic results after the analysis unit determines the warning level.

9. The train status fault analysis system based on data analysis according to claim 8, characterized in that, The analysis unit includes: The first comparison subunit determines the real-time fluctuation difference based on the first standard fluctuation difference and the second standard fluctuation difference to determine the warning level; The second comparison subunit acquires the corresponding real-time temperature rise when the real-time fluctuation difference is between the first standard deviation and the second standard deviation, and compares the standard temperature rise with the real-time temperature rise to determine the warning level. The third comparison subunit determines the warning level based on the actual fault impact range when the real-time temperature rise is greater than the standard temperature rise.

10. The train condition fault analysis system based on data analysis according to claim 9, characterized in that, The analysis unit further includes: The first-level early warning subunit determines that the fault belongs to the first-level fault when the real-time fluctuation difference is less than the second standard deviation, or when the real-time fluctuation difference is between the first and second standard deviations and the real-time temperature rise is less than or equal to the standard temperature rise. The secondary early warning subunit determines that the fault is a secondary fault when the real-time fluctuation difference is between the first standard deviation and the second standard deviation, the real-time temperature rise is greater than the standard temperature rise, and the actual fault impact range is less than the standard impact range. In a Level 3 early warning subunit, if the real-time fluctuation difference is greater than the first standard deviation, or if the real-time fluctuation difference is between the first and second standard deviations, the real-time temperature rise is greater than the standard temperature rise, and the actual fault impact range is greater than or equal to the standard impact range, the fault is determined to be a Level 3 fault.

Images